(Ghiduri) (Cancer) ACCP Diagnosis and Management of Lung Cancer - Evidence-Based Guidelines

DOI:10.1378/chest.123.1_suppl.
1S
2003;123;1-2 Chest
W. Michael Alberts
Lung Cancer Guidelines: Introduction
This information is current as of September 14, 2006
http://www.chestjournal.org/cgi/content/full/123/1_suppl/1S
located on the World Wide Web at:
The online version of this article, along with updated information and services, is
ISSN: 0012-3692.
may be reproduced or distributed without the prior written permission of the copyright holder.
3300 Dundee Road, Northbrook IL 60062. All rights reserved. No part of this article or PDF
published monthly since 1935. Copyright 2005 by the American College of Chest Physicians,
CHEST is the official journal of the American College of Chest Physicians. It has been
by on September 14, 2006 www.chestjournal.org Downloaded from
Lung Cancer Guidelines*
Introduction
W. Michael Alberts, MD, MBA, FCCP
(CHEST 2003; 123:1S2S)
T
he numbers are staggering. It is projected that
169,400 individuals in the United States will
receive diagnoses of cancer of the lung in 2002
(90,200 men and 79,200 women).
1
More disconcert-
ing is that 154,900 individuals will succumb to this
disease (89,200 men and 65,700 women) during the
year.
1
The numbers from abroad are no more com-
forting (and in many cases, more ominous). Approx-
imately 1 million people worldwide die of this dis-
ease each year.
2
Mortality
To personalize the scope of this problem, one
need only refer to an oft-quoted analogy. A mortality
rate of 154,900 individuals is approximately equiva-
lent to the death toll from a jumbo jet crashing every
day of the year, year after year. One can only imagine
the public outcry and Congressional hearings that
would result from such a series of events. The
analogy fails at this point for the public response to
lung cancer incidence and mortality is somewhat
muted. There are many reasons behind the current
public attitude, but suffice it to say that lung cancer
is a major public health problem.
Lung cancer is currently the leading cause of
cancer deaths in both men and women in the United
States. Deaths from lung cancer in women surpassed
those due to breast cancer in 1987 and are expected
to account for about 25% of all female cancer deaths
in 2002.
1
Thirty-one percent of cancer deaths in men
are attributable to lung cancer.
1
Lung cancer causes
more deaths than the next three most common
cancers combined (colon cancer, 48,100 deaths;
breast cancer, 40,000 deaths; and prostate, 30,200
deaths).
1
Prior to returning to the subject at hand, it must
be said that much of the effort evidenced in this
publication might not be necessary but for the real
culprit, namely, tobacco and tobacco products. To-
bacco use is the leading cause of preventable death
in this country and accounts for one of every five
deaths.
3
Half of regular smokers die prematurely of
a tobacco-related disease.
3
Not to minimize the
efforts of clinicians and clinical researchers, but let
us be honest; the biggest bang for the buck comes
in the form of lung cancer prevention. Whether
primary, secondary, or tertiary, the prevention of
cigarette smoking has the biggest potential to im-
prove the dismal statistics associated with lung
cancer.
Unfortunately, should tobacco and its products
magically disappear tomorrow, the health of the
population would continue to be victimized for
decades to come. Even today, more former smokers
than active smokers develop lung cancer. Eventually
though, lung cancer would be relegated to case
report status, a spot it enjoyed in the 19th century
and up to the advent of widespread cigarette use in
the 20th century. Most chest physicians would cheer
the day that their efforts could be refocused from
tobacco-induced disease to other diseases of the
chest.
Treatment
The status of the treatment of lung cancer is no
more encouraging. The expected 5-year survival rate
for all patients in whom lung cancer is diagnosed is
15%, compared with 61% for colon cancer, 86% for
breast cancer, and 96% for prostate cancer.
1
The
median survival time of patients with untreated
metastatic non-small cell lung cancer is 4 to 5
months, with a survival rate at 1 year of 10%.
4
In
2002, state-of-the-art treatment for this population
provides a median survival time of approximately
8 months (an extension of a mere 3 to 4 months) and
a 1-year survival rate of 33%.
4
For localized lung
cancer, the expectations of treatment are better but
not good. The 5-year survival rate for patients with
potentially resectable lung cancer is significantly
100% (stage IA, 67%; stage IB, 57%; stage IIA,
55%; stage IIB, 39%; and stage IIIA, 23%).
5
Fur-
thermore, progress in treatment has been slow. The
current overall 5-year survival rate of 15% is only
slightly better than the 8% survival rate of the early
*From the H. Lee Moffitt Cancer Center and Research Institute,
Tampa, FL.
Correspondence to: W. Michael Alberts, MD, MBA, FCCP,
Associate Center Director for Clinical Affairs, H. Lee Moffitt
Cancer Center and Research Institute, 12902 Magnolia Dr,
Tampa, FL 33612
www.chestjournal.org CHEST / 123 / 1 / JANUARY, 2003 SUPPLEMENT 1S
1960s. Given these data, many physicians have as-
sumed a nihilistic approach to the patient with lung
cancer.
Although a 15% 5-year survival rate is meager and
still dismal, the near doubling of the 5-year survival
rate has provided some room for optimism and has
begun to shift the nihilism associated with lung
cancer treatment into a guarded optimism. One
might take solace from the fact that 7% of newly
diagnosed patients (or nearly 12,000 patients) will
survive in 2002 but would not have been successfully
treated in 1960. In addition, a number of promising
new drugs have been incorporated into clinical trials,
and many more are in the pipeline. Specifically
targeted biological therapies are particularly promis-
ing. New diagnostic modalities, such as positron
emission tomography, are finding widespread use
and may alter our diagnostic and therapeutic algo-
rithms. New surgical procedures and techniques
have been developed and perfected. Safer and more
effective methods of delivering radiation are coming
into clinical use, and many people in the medical
community are cautiously hopeful that lung cancer
screening will prove able to convey a survival benefit
and be cost effective.
Lung Cancer Guidelines Project
It is against this backdrop of simultaneous pessi-
mism and optimism that the American College of
Chest Physicians, through the Health and Science
Policy Committee, commissioned the development
of evidence-based lung cancer guidelines in hopes
that a review, evaluation, and synthesis of the pub-
lished literature, along with expert consensus when
necessary, might lead to a series of recommendations
that would assist physicians in achieving the best
possible outcomes for their patients, given the knowl-
edge and capabilities available at this time.
The size of the task was daunting, but the goal was
laudable. The effort expended on this project by
many individuals has been truly heroic. The volun-
tary effort of the section editors, the writing commit-
tees, and the review panels in support of this publi-
cation and our patients has been nothing less than
impressive. I am very pleased with the final product
and hope that it proves to be of benefit to you and
your patients.
Special thanks go to Gene Colice, MD, as Vice-
Chair of the Lung Cancer Guidelines Project, and
Doug McCrory, MD, as the Principal Investigator
with the Duke University Evidence-based Practice
Center. Both devoted many extra hours, nights, and
weekends over the past 2 years to ensure the success
of the project. Members of the Health and Science
Policy Committee and the American College of
Chest Physicians Board of Regents deserve recogni-
tion for their review and editing of the final manu-
script. The true driving force, however, behind this
effort has been Sandra Lewis, PhD, who, as Project
Manager, has brought the project to this point
through sheer effort and diplomatic prodding. A
thank you to her is certainly in order.
References
1 Jemal A, Thomas A, Murray T, et al. Cancer statistics, 2002.
CA Cancer J Clin 2002; 52:2347
2 Carney DN. Lung cancer: time to move on from chemother-
apy. N Engl J Med 2002; 346:126127
3 Centers for Disease Control Prevention. Tobacco use: United
States, 1999. MMWR Morb Mortal Wkly Rep 1999; 48:986
1900
4 Schiller JH, Harrington D, Belani CP, et al. Comparison of
four chemotherapy regimens for advanced non-small-cell
lung cancer. N Engl J Med 2002; 346:9298
5 Mountain CF. Revisions in the International System for
Staging Lung Cancer. Chest 1997; 111:17101717
2S Lung Cancer Guidelines
DOI:10.1378/chest.123.1_suppl.1S
2003;123;1-2 Chest
W. Michael Alberts
Lung Cancer Guidelines: Introduction
& Services
Updated Information
figures, can be found at:
Updated information and services, including high-resolution
References
BIBL
http://www.chestjournal.org/cgi/content/full/123/1_suppl/1S#
at:
This article cites 5 articles, 4 of which you can access for free
Permissions & Licensing
http://www.chestjournal.org/misc/reprints.shtml
tables) or in its entirety can be found online at:
Information about reproducing this article in parts (figures,
Reprints
Information about ordering reprints can be found online:
Email alerting service
sign up in the box at the top right corner of the online article.
Receive free email alerts when new articles cite this article
Images in PowerPoint format
article figure for directions.
teaching purposes in PowerPoint slide format. See any online
Figures that appear in CHEST articles can be downloaded for
2003;123;3-6 Chest
and Sydney Parker
Douglas C. McCrory, Gene L. Colice, Sandra Zelman Lewis, W. Michael Alberts
Development
Overview of Methodology for Lung Cancer Evidence Review and Guideline
ISSN: 0012-3692.
Overview of Methodology for Lung
Cancer Evidence Review and Guideline
Development*
Douglas C. McCrory, MD, MHS; Gene L. Colice, MD, FCCP;
Sandra Zelman Lewis, PhD; W. Michael Alberts, MD, FCCP; and
Sydney Parker, PhD
A multidisciplinary panel was convened by the American College of Chest Physicians (ACCP)
to develop clinical practice guidelines for lung cancer prevention, diagnosis, and treatment.
The ACCP Expert Panel on Lung Cancer Guidelines produced 20 guidelines, each related to
a distinct set of management decisions. This article describes the approach used to develop
the guidelines, including identifying, evaluating, and synthesizing evidence, assessing the
strength of evidence pertinent to individual guidelines, and grading guideline
recommendations. (CHEST 2003; 123:3S6S)
Key words: advisory committees; evidence-based medicine; lung neoplasms; practice guidelines
Abbreviations: ACCP American College of Chest Physicians; NSCLC non-small cell lung
cancer; USPSTF US Preventive Services Task Force
T
he American College of Chest Physicians
(ACCP) Health and Science Policy Committee,
through a poll of the ACCP membership and a
structured internal review of all nominated topics,
selected lung cancer as a high-priority topic for
developing evidence-based clinical practice guide-
lines. The ACCP, with support from the Duke
University Center for Clinical Health Policy Re-
search, commissioned an expert panel, the goal of
which was to produce clinically relevant, evidence-
based guidelines for lung cancer prevention, diagno-
sis, and treatment that were aimed at primary care
physicians and pulmonary specialists. Because the
target audience is primary care physicians and pul-
monologists, the guidelines are somewhat more de-
tailed about issues of prevention, screening, and
diagnostic evaluation compared to treatment issues.
The guidelines avoid detailed recommendations
about how to administer treatments, which usually
are provided by other specialists, and instead focus
on recommendations for treatment approaches that
are sufficient to make referral decisions.
The guideline topics are as follows: (1) prevention;
(2) screening and early detection; (3) diagnosis;
(4) initial evaluation; (5) physiologic and functional
assessment; (6) solitary pulmonary nodules; (7) non-
invasive staging; (8) invasive staging; (9) treatment of
early-stage non-small cell lung cancer (NSCLC);
(10) treatment of stage I NSCLC; (11) treatment of
stage II NSCLC; (12) treatment of stage IIIA
NSCLC; (13) treatment of stage IIIB NSCLC;
(14) lung cancers with special considerations;
(15) chemotherapeutic management of stage IV
NSCLC; (16) small cell lung cancer evaluation and
treatment; (17) follow-up/surveillance; (18) palliative
treatment; (19) end-of-life care; and (20) practice or-
ganization. Not included among the topics is the man-
agement of other malignancies that also may occur in
the lung, including mesothelioma, hamartoma, thy-
moma, and carcinoid and neuroendocrine tumors. Also
omitted from these topics are complementary and
alternative medicine treatments.
Materials and Methods
Evidence Review
As a first step in identifying the evidence for each topic, we
sought existing evidence syntheses including guidelines, system-
atic reviews, and meta-analyses. We searched computerized
*From the Department of Medicine (Dr. McCrory), Duke
University Medical Center, Durham, NC; Pulmonary, Critical
Care, and Respiratory Services (Dr. Colice), Washington Hospi-
tal Center, Washington, DC; the American College of Chest
Physicians (Drs. Lewis and Parker), Northbrook, IL; and the H.
Lee Moffitt Cancer Center and Research Institute (Dr. Alberts),
Tampa, FL.
This research was supported by a contract from the American
College of Chest Physicians.
Correspondence to: Douglas McCrory, MD, MHS, Duke Center
for Clinical Health Policy Research, 2200 W Main St, Suite 220,
Durham, NC 27705; e-mail: douglas.mccrory@duke.edu
bibliographic databases including MEDLINE, Cancerlit,
CINAHL and HealthStar, the Cochrane Collaboration Database
of Abstracts of Reviews of Effectiveness, the National Guideline
Clearinghouse, and the National Cancer Institute Physician Data
Query database. Computerized searches through July 2001 used
the MeSH terms lung neoplasms (exploded) and bronchial
neoplasms or text searches for lung cancer combined with review
articles, practice guidelines, guidelines, and meta-analyses. We
also searched and included studies from the reference lists of
review articles, and queried experts in the field. An international
search was conducted of Web sites of provider organizations that
were likely to have developed guidelines. Abstracts of candidate
English language articles were reviewed by two physicians (one
with methodological expertise and one with content area exper-
tise) and a subset was selected for review in full text. Full-text
articles were reviewed again by two physicians to determine
whether they were original publications of a synthesis and were
pertinent to at least one of the topics of the guideline. Articles
described as practice guidelines, systematic reviews, or meta-
analyses were included, as were review articles that included a
Methods section. Included articles were classified according to
topic.
The methodological quality of existing clinical practice guide-
lines were evaluated using a rigorously developed and validated
instrument (see the article on the critical appraisal of lung cancer
guidelines in this supplement).
1
An initial review of published syntheses permitted the execu-
tive committee to select topics for new evidence reviews. Five
topics for which existing syntheses were inadequate were selected
for new evidence reviews conducted by the Duke Center for
Clinical Health Policy Research. These topics included preven-
tion, screening, diagnosis, and staging (invasive and noninvasive).
The methods for these reviews are described in separate articles
in this supplement. For other topics, identified guidelines, meta-
analyses, and systematic reviews were made available to the
writing committees described below.
Guideline Development
The guideline development panel was composed of members
and nonmembers of the ACCP who were known to have
expertise in various areas of lung cancer management and care,
representing multiple specialties, and 13 national and interna-
tional medical associations (Table 1), including the Alliance for
Lung Cancer Advocacy, Support, and Education (a patient
support group) and the Oncology Nurses Society. The specialties
included pulmonary/respiratory medicine, critical care, medical
oncology, thoracic surgery, radiation oncology, epidemiology,
law, and medical ethics. The panel was led by an executive
committee including the chair (MA) and vice chair (GC), the
leader of the methodology support group (DM), the ACCP
project manager (SL), and the vice-president for Health and
Science Policy (SP). The executive committee directed the
guideline development process, methodological issues, panel
composition, structure of the final document, and activities of the
topic-specific writing committees.
Each writing committee was led by a member of the expert
panel who agreed to review evidence and develop guidelines on
1 or more of the 20 topics listed earlier. The lead editor for each
writing committee nominated individuals to assist with evidence
review and guideline development, with the goal of including
expertise from all relevant disciplines. Nominations for writing
committee membership were reviewed and approved by the
executive committee.
Funding for both the evidence reviews and guideline develop-
ment was provided through an unrestricted educational grant
from Bristol-Myers Squibb, which had no other role in the
evidence review or guideline development process or content.
Information about potential conflicts of interest were collected
from each member of the expert panel or writing committee at
the time of their nomination in accordance with the policy of the
ACCP. Information on conflicts of interest for each panelist is
listed in the guideline.
Process
Each writing committee was charged with identi-
fying the important management issues for which
guidelines were needed. Each writing committee
received a comprehensive list of existing systematic
reviews and meta-analyses as well as guidelines
published by other groups. In addition, for the five
key topics described earlier, new systematic reviews
were undertaken and were led by either the Duke
Center for Clinical Health Policy Research or the
Health Outcomes Research Group, Department of
Epidemiology and Biostatistics, and Department of
Medicine, Memorial Sloan-Kettering Cancer Cen-
ter. For all other topics, writing committees were
responsible for identifying and interpreting studies
that were not otherwise covered in existing syntheses
or guidelines.
To develop guidelines, the panel used informal
group consensus techniques. The guidelines devel-
oped by the writing committees on each topic were
distributed to the entire expert panel, and comments
were solicited in advance of a meeting. During the
meeting, proposed recommendations were re-
viewed, discussed, and voted on by the entire panel.
Approval required consensus, which was defined as
an overwhelming majority approval. Differences of
opinion were accommodated by revising the pro-
posed recommendation, the rationale, or the grade
until consensus could be reached. The evidence
supporting each recommendation was summarized,
and recommendations were graded as described
Table 1Organizations and Associations Represented
by Guideline Development Panel Members
Panel Member Organizations
Alliance for Lung Cancer Advocacy, Support, and Education
American Association for Bronchology
American Cancer Society
American College of Physicians
American College of Surgeons Oncology Group
American Society of Clinical Oncology
American Society for Therapeutic Radiology and Oncology
American Thoracic Society
Association of Community Cancer Centers
Canadian Thoracic Society
National Comprehensive Cancer Network
Oncology Nurses Society
Society of Thoracic Surgeons
below. The assessments of level of evidence, net
benefit, and grade of recommendation were re-
viewed by the executive committee.
Values
The panel considered data on functional status,
quality and length of life, tolerability of treatment,
and relief of symptoms in formulating guideline
recommendations. Cost was not explicitly considered
in the guideline development process. Data on these
outcomes were informally weighted, without the use
of explicit decision analysis or other modeling. The
values placed on types of outcomes varied with
clinical scenarios. For example, in some situations we
considered life expectancy, such as the effects of
early detection. In other situations we weighed
quality of life more heavily, such as in palliative care
and in interpreting small increases in life expectancy
with chemotherapy for stage IV disease.
Grading Recommendations
Recommendations were graded then using an adap-
tation of the scale used by the US Preventive Services
Task Force (USPSTF), one of the most commonly
cited approaches (Table 2).
2
The USPSTF approach
provided a framework for linking evidence to a concep-
tual model of how a clinical service affects health
outcomes. Evidence is reviewed at the following three
levels: (1) the individual studies; (2) the body of
evidence concerning a single linkage in the concep-
tual model; and (3) the body of evidence relating
interventions with health outcomes. At each level,
the scheme offers general guidelines to assign one of
the following three grades of evidence: good, fair, or
poor. Good or fair quality evidence must include
studies of sufficient design and quality to provide an
unbroken chain of evidence-supported linkages that
connect the intervention with health outcomes. In
general, good evidence included prospective, con-
trolled, randomized clinical trials, and poor evidence
included case series and clinical experience. Trials
with fair quality of evidence, for instance, historically
controlled trials or retrospective analyses, were
somewhere in between. In addition to the strength
of the study design, however, study quality also was
considered. The USPSTF approach considers well-
recognized criteria in rating the quality of individual
studies for a variety of different types of study design
(eg, diagnostic accuracy studies and case-control
studies). The thresholds for distinguishing good vs
fair and fair vs poor evidence are not explicit but are
Table 2Grades of Recommendation and Estimates of Net Benefit*
Grades Description
Recommendation
A The panel strongly recommends that clinicians routinely provide [the service] to eligible patients. An A
recommendation indicates good evidence that [the service] improves important health outcomes and that
benefits substantially outweigh harms.
B The panel recommends that clinicians routinely provide [the service] to eligible patients. A B
recommendation indicates at least fair evidence that [the service] improves important health outcomes and
concludes that benefits outweigh harms.
C The panel recommends that clinicians routinely provide [the service] to eligible patients. A C
recommendation indicates that there was consensus among the panel to recommend [the service] but that
the evidence that [the service] is effective is lacking, of poor quality, or conflicting, or the balance of
benefits and harms cannot be reliably determined from available evidence.
D The panel recommends against clinicians routinely providing [the service]. A D recommendation indicates
at least fair evidence that [the service] is ineffective or that harm outweighs benefit.
I The panel concludes that the evidence is insufficient to recommend for or against [the service]. An I
recommendation indicates that evidence that [the service] is effective is lacking, of poor quality, or
conflicting, and the balance of benefits and harms cannot be determined, and that the panel lacked a
consensus to recommend it.
Net benefit
Substantial Benefit greatly outweighs harm
Moderate Benefit outweighs harm
Small/weak Benefit outweighs harm to a minimally clinically important degree
None/negative Harms equal or outweigh benefit, less than clinically important
*Adapted from Harris et al.
2
The grade of the strength of recommendations is based on both the quality of the evidence and the net benefit of the service (ie, test, procedure,
etc.).
These levels of net benefit are based on clinical assessment. Estimated net benefit may be downgraded based on uncertainty in estimates of
benefits and harms.
left to the judgment of panelists, reviewers, and
members of the executive committee.
The grade of recommendation also depends on an
assessment of the magnitude of benefit, harm, and
net benefit, which is rated as substantial, moderate,
small, or none/negative. Similar to the determination
of quality of evidence, the determination of magni-
tude of benefit was a clinical judgment of the panel
members (Table 2).
In assigning a grade to each recommendation, we
first assigned a level of evidence, then estimated the
magnitude of the net benefit, and used these two
factors to provide the grade for the recommendation,
which was coded as a letter (A, strongly recom-
mended; to D, recommend against) [Table 3]. An I
recommendation indicated situations in which the
evidence is insufficient to determine net benefit.
Our grading scheme is a modification of the
USPSTF grades to allow recommendations for a
service when (1) evidence is poor, (2) the assessment
of the net benefit is moderate to high, and (3) there
is consensus among the expert panel to recommend
it. This change was necessary because, unlike pre-
ventive services (ie, the routine offering of tests or
treatments to well people) in which the burden of
proof is high, clinical decisions about the treatment
of patients with lung cancer often must be based
on an interpretation of the available evidence, even if
it is of poor quality. Our adaptation distinguished
between interventions with poor evidence for which
there is consensus (grade C) and interventions with
poor evidence for which there is not consensus
(grade I).
Validation
After extensive review within the expert panel and
executive committee, the guidelines were reviewed
and approved by the ACCP Health and Science
Policy Committee and then by the ACCP Board of
Regents. The guidelines have not been field-tested.
References
1 AGREE Collaborative Group. Appraisal of guidelines for
research & evaluation (AGREE) instrument: second draft;
August 2000. Available at: http://www.agreecollaboration.org.
Accessed December 16, 2002
2 Harris RP, Helfand M, Woolf SH, et al. Current methods of
the US Preventive Services Task Force: a review of the
process. Am J Prev Med 2001; 20(suppl):2135
Table 3Grade of Recommendation Based on Quality
of Evidence and Estimate of Net Benefit
Quality of
Evidence
Net Benefit
Substantial Moderate Small/weak None/Negative
Good A B C D
Fair B B C D
Poor C C C or I* I
*Depending on whether panelists reached consensus or not, this
grade could be C or I, respectively.
2003;123;3-6 Chest
and Sydney Parker
Douglas C. McCrory, Gene L. Colice, Sandra Zelman Lewis, W. Michael Alberts
Development
Overview of Methodology for Lung Cancer Evidence Review and Guideline
& Services
Updated Information
Reprints
2003;123;7-20 Chest
Kolimaga and Douglas C. McCrory
Linda H. Harpole, Michael J. Kelley, Gilbert Schreiber, Eric M. Toloza, Jane
Cancer
Assessment of the Scope and Quality of Clinical Practice Guidelines in Lung
ISSN: 0012-3692.
Assessment of the Scope and Quality of
Clinical Practice Guidelines in Lung
Cancer*
Linda H. Harpole, MD, MPH; Michael J. Kelley, MD;
Gilbert Schreiber, MD, FCCP; Eric M. Toloza, MD, PhD; Jane Kolimaga, MA;
and Douglas C. McCrory, MD, MHS
Study objectives: To provide an evidence-based background for developing the American College
of Chest Physicians (ACCP) lung cancer guidelines, a systematic review of the literature was
performed to identify published lung cancer guidelines and evaluate their quality.
Design, setting, and participants: A systematic search was performed for relevant literature from
MEDLINE, Cancerlit, CINAHL, HealthStar, the Cochrane Library, and the National Guidelines
Clearinghouse published from January 1989 to July 2001.
Measurement and results: From 369 citations, 51 relevant guidelines were identified. Each
guideline was evaluated by at least four reviewers using the Appraisal of Guidelines for Research
and Evaluation (AGREE) instrument and was coded for clinical topics covered. The recommen-
dations included in each guideline also were abstracted. Of the 51 guidelines evaluated, 27 (53%)
were evidence-based. Clinical topics identified by the ACCP for their guideline effort each were
represented by at least one existing guideline. Of the 880 clinical recommendations abstracted
from the guidelines, only 253 (29%) were evidence-based. The AGREE instrument rates
guidelines along six domains. As a group, the guidelines performed well in the scope and purpose
domain, with only six guidelines (12%) scoring < 50%. For the remaining domains, however, the
guidelines did not perform as well, as follows: for stakeholder involvement, 41 guidelines (80%)
scored < 50%; for rigor of development, 29 guidelines (57%) scored < 50%; for clarity and
presentation, 17 guidelines (33%) scored < 50%; for applicability, 46 guidelines (90%) scored
< 50%; and for editorial independence, 47 guidelines (92%) scored < 50%. After considering the
domain scores, the reviewers recommended only 19 of the guidelines (37%).
Conclusions: All major clinical lung cancer topics are covered by at least one guideline, but no
single guideline addresses all areas. Furthermore, although existing guidelines may accurately
reflect clinical practice, most performed poorly when evaluated for quality. Future guideline
efforts that address each item of the AGREE instrument would add substantially to the literature.
(CHEST 2003; 123:7S20S)
Key words: evidence-based medicine; lung neoplasms; practice guidelines
Abbreviation: AGREE Appraisal of Guidelines for Research and Education
L
ung cancer is the leading cause of cancer death for
both men and women in the United States. In
2001, an estimated 169,500 new cases of lung cancer
will have been diagnosed, and an estimated 157,400
deaths will have been attributed to the disease.
1
Cur-
rent 5-year lung cancer survival rates are estimated at
14%.
1
Clinical practice guidelines are thought to be
capable of improving quality, appropriateness, and
cost-effectiveness of care.
2
A potential mechanism for
improving outcomes in patients with lung cancer, then,
would be to ensure that those patients are receiving
evidence-based guideline care.
Clinical practice guidelines have been defined as
systematically developed statements to assist practi-
tioner and patient decisions about appropriate health
care for specific clinical circumstances.
2
However, the
recent increase in the production of clinical practice
guidelines has been accompanied by growing concern
about the variations in guideline recommendations
3,4
and quality.
57
*From the Departments of Medicine (Drs. Harpole, Kelley,
Schreiber, and McCrory) and Surgery (Dr. Toloza) and Center
for Clinical Health Policy Research (Ms. Kolimaga), Duke Uni-
versity Medical Center, Durham, NC.
Correspondence to: Linda H. Harpole, MD, MPH, Duke Center
for Clinical Health Policy Research, 2200 W Main St, Suite 220,
Durham, NC 27705; e-mail: harpo003@mc.duke.edu
In 2001, a multidisciplinary panel was convened by
the American College of Chest Physicians to develop
evidence-based clinical practice guidelines for lung
cancer diagnosis and treatment. To avoid potential
duplication of effort, the first step taken was to
identify and determine the quality of already pub-
lished guidelines in this area.
Identification of Guidelines
Relevant guidelines were identified through computerized
searches of MEDLINE, Cancerlit, CINAHL, HealthStar, the
Cochrane Library, and the National Guidelines Clearinghouse,
by reviewing the reference lists of review articles and included
guidelines, and by consulting experts in the field. The search
strategy used the MeSH terms lung neoplasms (exploded) and
bronchial neoplasms, and required further indexing with publi-
cation type guideline or practice guideline, MeSH heading
guidelines, or textword guideline or guidelines. Investigators
reviewed English-language studies that had been published since
1989. Identified citations were screened for inclusion based on
the following Institute of Medicine
2
definition of a guideline:
systematically developed statements to assist practitioner and
patient decisions about appropriate health care for specific
clinical circumstances. Single-author overviews, secondary pub-
lications of practice guidelines, editorials, and letters to the editor
were specifically excluded.
Appraisal Instrument
The methodological quality of existing clinical practice guide-
lines was evaluated using the Appraisal of Guidelines for Re-
search and Education (AGREE) instrument,
8
an international,
rigorously developed, and validated instrument that compares
well with other instruments designed for this purpose.
9
This
instrument allowed for the assessment of several components
that are integral to guideline development, as follows: (1) scope
and purpose; (2) stakeholder involvement; (3) rigor of de-
velopment; (4) clarity and presentation; (5) applicability; and
(6) editorial independence. Five reviewers (LH, DM, ET, MK,
and GS) used the AGREE instrument to evaluate the scientific
quality of the lung cancer guidelines. A minimum of four
reviewers completed the AGREE instrument for each guideline
and also determined whether the guideline was evidence-based
or consensus-based.
Each guideline was coded for the following topics covered:
prevention; screening and early detection; initial evaluation;
diagnosis; clinical staging; pathologic/surgical staging; treatment-
early stage; treatment-stage I; treatment-stage II; treatment-stage
IIIA/potentially resectable; treatment-stage IIIB/nonresectable;
treatment-stage IV; treatment-Pancoast tumor, T4, and those
requiring special consideration; treatment-small cell lung cancer;
treatment-solitary pulmonary nodule; follow-up/surveillance; pal-
liative care; palliative treatment; and practice organization.
The text of the recommendations included in each guideline
also was abstracted. Each recommendation was coded for topic,
subtopic, and type of evidence utilized when formulating the
statement (ie, A, strong evidence; B, weak evidence; and C,
consensus). Since different guidelines used different scales to
rate the strength and quality of evidence supporting a particular
recommendation, we often had to map published grades to our
scale. In general, statements were graded as strong evidence (A)
if they were supported by randomized controlled trials, weak
evidence (B) if they were supported by evidence other than that
from randomized controlled trials, and consensus (C) if they were
not supported by clear study data. For guidelines that did not
grade recommendations, we categorized statements based on the
above schema.
Evaluation of Guidelines
The 23-item AGREE instrument is divided into the following
six domains (see Appendix): scope and purpose (three items);
stakeholder involvement (four items); rigor of development
(seven items); clarity and presentation (four items); applicability
(three items); and editorial independence (two items). The score
for each domain is obtained by summing up all the scores of the
individual items in a domain and then standardizing as follows:
obtained score - minimum possible score
maximum possible score - minimum possible score
The maximum score for each domain would be the number of
questions multiplied by the number of reviewers multiplied times
4 (ie, the score for strongly agree). The minimum possible score
for a domain would be the number of questions multiplied times
the number of reviewers multiplied times 1 (ie, the score for
strongly disagree).
The final component of the AGREE instrument involves a
recommendation regarding the use of the guidelines in practice
as strongly recommended, recommended (with provisos or
alterations), would not recommend, or unsure. On this item,
the investigators reached consensus for each guideline. For ease
of interpretation, we considered strongly recommended and
recommended with provisos or alterations as a response of
recommended, and would not recommend or unsure as a
response of would not recommend. The AGREE instrument
instructs the raters to make a judgment as to the quality of the
guideline, taking each of the appraisal criteria into consideration.
In our ratings, we took into account the date of the guideline and
considered whether we would recommend the document as a
useful tool that could be adapted locally by a health-care provider
who was considering implementing the guideline in a health-care
practice or system. We placed relatively more weight on the
quality of development than on whether the recommendations
matched our particular clinical practice or were feasible in our
particular practice environments.
Prior to evaluating the guidelines included in the review, we
first each rated a superseded guideline,
10
compared ratings
among reviewers, discussed discrepancies, and reached consen-
sus about the interpretation of each question.
We used the statistic as a measure of the agreement among
reviewers.
11
However, before performing any calculations, the
response categories were dichotomized into strongly agree/agree
vs strongly disagree/disagree, as we thought that an analysis of
agreement at this level was sufficient. The statistic was then
applied to each of the 23 items of the AGREE instrument. The
simple proportion of agreement also was calculated.
Results
We screened 308 citations that had been identified
through computerized and other database searches,
and an additional 61 citations that had been identi-
fied through clinical experts, reference lists, web site
searches, and other sources. A total of 81 candidate
guidelines were selected, and, after reviewing the
full text of each reference relative to the Institute of
Medicine definition for a clinical practice guideline,
a total of 51 guidelines were selected for this review
(Table 1).
1262
Of the 51 guidelines, 47% were consensus-based,
and the remaining 53% were evidence-based. Except
for two guidelines, all had been written in the past
5 years. All clinical topics defined by the American
College of Chest Physicians were represented by at
least one guideline, with a range of 8 to 118 recom-
mendations per topic (Table 2). Some recommenda-
tions were pertinent to more than one clinical topic.
When this occurred, the recommendation was refer-
enced to all relevant topics. A significant degree of
overlap occurred across the clinical topics. This is
illustrated by the category solitary pulmonary nod-
ule. Recommendations that specifically discussed
the solitary pulmonary nodule were categorized un-
der this topic; however, some recommendations
under the topics clinical staging, diagnosis, and
small cell lung cancer also were relevant to the
solitary pulmonary nodule category.
Of the total of 880 clinical recommendations, the
majority (71%) were consensus-based. There was a
notable dearth of evidence-based recommendations
for the following: diagnosis; initial evaluation and
preparation; follow-up/surveillance; treatment of the
solitary pulmonary nodule, early stage, and special
cases (eg, Pancoast tumor, T4); and palliative care
and treatment.
The quality of the guidelines is represented by the
AGREE domain scores in Table 3.
AGREE Results
Scope and Purpose: The score for this domain
represents the degree to which the overall objectives
of the guideline, the clinical questions covered, and
the patients to whom the guideline was meant to
apply were specifically described. Overall, the mean
score was 72% (range, 29 to 97%), indicating that, on
average, 72% of the criteria for scope and purpose
were met. Most guidelines performed well in this
domain, with only six guidelines (12%) scoring
50%.
Stakeholder Involvement: This domain evaluates
the degree to which the guideline represents the
views of its intended users. Included are questions
regarding the composition of the guideline develop-
ment group (specifically, whether individuals from
all relevant professional groups were represented),
whether patients experiences and expectations in-
formed the development of the guideline, whether
the target users of the guideline were well-defined,
and whether the guideline was piloted among end-
users. Overall, the mean score for this domain was
35% (range, 3 to 70%), with 41 guidelines (80%)
scoring 50%. Only 6% of guidelines included
individuals from all relevant professional groups in
the development stage, and none was piloted among
end-users.
Rigor of Development: This domain specifically
evaluates whether systematic methods were used to
search for evidence, whether the criteria for select-
ing the evidence and the methods used to formulate
the recommendations were clearly described,
whether there was an explicit link between the
recommendations and the supporting evidence,
whether health benefits, side effects, and risks were
considered when formulating the recommendations,
whether the guideline was externally reviewed by
experts prior to publication, and whether a proce-
dure for updating the guideline was provided. Over-
all, the mean score for this domain was 52% (range,
2 to 95%), with 57% of guidelines scoring 50%.
Specifically, only 16 guidelines (31%) described sys-
tematic methods for searching and selecting the
evidence, 18 guidelines (35%) considered health
benefits, side effects, and risks when formulating the
recommendations, and 20 guidelines (39%) de-
scribed the methods used to formulate the recom-
mendations. Moreover, only 18 guidelines (35%)
were externally reviewed prior to publication.
Clarity and Presentation: This domain describes
the clarity of the guidelines. Specifically, it describes
whether the recommendations were specific and
unambiguous, whether the different management
options were clearly presented, whether key recom-
mendations were easily identifiable, and whether the
guideline was supported with tools for application.
Overall, the mean score for this domain was 57%
(range, 15 to 90%). Only two guidelines (4%) in-
cluded tools for application. Seventeen guidelines
(33%) scored 50% for this domain.
Applicability: This domain evaluates issues that
are pertinent to guideline implementation. More
specifically, it considers organizational barriers, cost
implications, and monitoring criteria. The score on
this domain was the lowest of all, with a mean score
of 20% (range, 0 to 98%). Only five guidelines (10%)
scored at least 50%. Two guidelines provided review
criteria for monitoring purposes, and six discussed
potential organizational barriers. No guideline dis-
cussed cost implications.
Editorial Independence: This domain addresses
Table 1Lung Cancer Clinical Practice Guidelines*
Authors Date Title
Evidence-Based or
Consensus-Based? Topics Covered
AATS, STS, STSA, and WTSA 1993 Practice guidelines in
cardiothoracic surgery
12
Consensus-based Treatment-special case
ACS 2001 ACS guidelines for the early
detection of cancer: update of
early detection guidelines for
prostate, colorectal, and
endometrial cancers; also update
2001testing for early lung
cancer detection
13
Consensus-based Screening and early detection
ACR 2000 ACR appropriateness criteria:
nonaggressive, nonsurgical
treatment of inoperable
NSCLC
14
Consensus-based Treatment-stages I, IIIA, IIIB, IV
non-small cell lung cancer,
nonsurgical, aggressive therapy
15
Consensus-based Treatment-stages IIIb, IV
work-up of the solitary
pulmonary nodule
16
Consensus-based Treatment-solitary pulmonary nodule
follow-up and retreatment of
brain metastases
17
Consensus-based Treatment-stage IV
follow-up of non-small cell
lung cancer
18
Consensus-based Follow-up/surveillance
multiple brain metastases
19
neoadjuvant therapy for
marginally respectable (clinical
N2) non-small cell lung cancer
20
Consensus-based Treatment-stage IIIA
postoperative radiotherapy in
non-small cell lung cancer
21
Consensus-based Treatment-stages I, II, IIIA
preirradiation evaluation and
management of brain
metastases
22
solitary brain metastasis
23
Consensus-based Treatment-small cell lung cancer,
stage IV
staging of bronchogenic
carcinoma, non-small cell
lung carcinoma
24
Consensus-based Clinical staging
staging of non-small cell lung
carcinoma
25
Consensus-based Clinical staging
AES, Clinical Research
Committee
1993 Consensus guidelines for high
dose rate remote brachytherapy
in cervical, endometrial, and
endobronchial tumors
26
Consensus-based Palliative treatment
ASCO 1997 Clinical practice guidelines for
the treatment of unresectable
27
Evidence-based Clinical staging; diagnosis; follow-up/
surveillance; pathologic/surgical
staging, prevention; treatment-
stages IIIb, IV
ATS 2000 Management of malignant
pleural effusions
28
Consensus-based Palliative treatment; treatment-stage IV
ATS 1989 Guidelines for percutaneous
transthoracic needle biopsy
29
Consensus-based Diagnosis; treatment-solitary pulmonary
nodule
ATS and ERS 1997 Pretreatment evaluation of non-
small cell lung cancer
30
Consensus-based Clinical staging; diagnosis; initial
evaluation/preparation; screening and
early detection
Table 1Continued*
Authors Date Title
Evidence-Based or
ACCC 2000 Oncology patient management
guidelines: small cell lung
cancer
31
Consensus-based Clinical staging; diagnosis; follow-up/
surveillance; initial
evaluation/preparation; pathologic/
surgical staging; treatment-small cell
lung cancer; stages I, II, IIIA, IIIB,
and IV
Biesalski et al 1997 Consensus statement on lung
cancer
32
Consensus-based Prevention
BTS and SCSGBI Working Party 2001 BTS guidelines: guidelines on
the selection of patients with
lung cancer for surgery
33
Evidence-based Clinical staging; follow-up/surveillance;
initial evaluation/preparation;
pathologic/surgical staging; practice
organization; treatment-small cell lung
cancer, Pancoast tumor, and stages I,
II, IIIA, IIIB, and IV
BTS Standards of Care Committee
Lung Cancer Working Party
1998 BTS recommendations to
respiratory physicians for
organizing the care of patients
with lung cancer
34
surveillance; initial
evaluation/preparation; palliative care;
practice organization; screening and
early detection; treatment-small cell
lung cancer, Pancoast, tumor and
stages I, II, IIIA, IIIB, and IV
CCOPGI 2001 The role of combination
chemotherapy in the initial
management of limited-stage
35
Evidence-based Treatment-small cell lung cancer
CCOPGI 2001 The role of single-agent
docetaxel as a second-line
treatment for advanced non-
36
Evidence-based Treatment-stage IV
CCOPGI 2000 Altered fractionation of radical
radiation therapy in the
management of unresected non-
37
Evidence-based Treatment-stage IIIB
CCOPGI 2000 Chemotherapy in stage IV
(metastatic) non-small cell lung
cancer
38
Evidence-based Treatment-stage IV
CCOPGI 2000 Postoperative adjuvant
chemotherapy and/or radiation
therapy in stage II or IIIA
completely resected non-small
cell lung cancer
39
Evidence-based Treatment-stages II, IIIA
CCOPGI 2000 Prophylactic cranial irradiation
in small cell lung cancer
40
CCOPGI 2000 The role of thoracic
radiotherapy as an adjunct to
standard chemotherapy in
limited-stage small cell lung
cancer
41
CCOPGI 2000 Unresected stage III non-
42
Evidence-based Treatment-stage IIIB
CCOPGI 2000 Use of gemcitabine in non-
43
Evidence-based Treatment-stages IIIB, IV
CCOPGI 2000 Use of preoperative
chemotherapy with or without
postoperative radiotherapy in
technically resectable stage IIIA
44
Evidence-based Treatment-stage IIIA
CCOPGI 2000 Use of vinorelbine in non-small
cell lung cancer
45
Evidence-based Treatment-stages IIIB, IV
Table 1Continued*
Authors Date Title
Evidence-Based or
CGG 1998 Guidance on commissioning cancer
services: improving outcomes in
lung cancer; the manual
46
Evidence-based Clinical staging; diagnosis; palliative care
and treatment; pathologic/surgical
staging; practice organization;
prevention; screening and early
detection; treatment-small cell lung
cancer, stages I, II, IIIA, IIIB, and IV
Colle` ge des Me decins du Que bec 1999 Clinical practice guidelines:
smoking prevention and
cessation
47
Evidence-based Prevention
ESMO 2001 ESMO minimum clinical
recommendations for diagnosis,
treatment and follow-up of non-
small-cell lung cancer
48
Evidence-based Palliative care; treatment-stages I, II
ESMO 2001 ESMO minimum clinical
treatment and follow-up of
small-cell lung cancer
49
HCHP Adult Screening and
Prevention Task Force
1989 Screening for lung cancer
50
Evidence-based Screening and early detection
NCI 2001 Lung cancer (PDQ): prevention
51
NCI 2001 Lung cancer (PDQ): screening
52
Evidence-based Screening and early detection
NCI 2001 Non-small cell lung cancer (PDQ):
treatment
53
surveillance; initial evaluation/
preparation; pathologic/surgical
staging; treatment-Pancoast tumor
and stages I, II, IIIA, IIIB, and IV
NCI 2000 Small cell lung cancer (PDQ):
treatment
54
NCCN 2000 NCCN practice guidelines for non-
55
Consensus-based Clinical staging; follow-up/surveillance;
pathologic/surgical staging; treatment-
stages IIIB, IV
NCCN 2000 NCCN practice guidelines for
56
Consensus-based Clinical staging; follow-up/surveillance;
initial evaluation/preparation;
treatment-small cell lung cancer
RCR Clinical Oncology
Information Network
1999 Guidelines on the nonsurgical
management of lung cancer
57
Evidence-based Practice organization; treatment-small
cell lung cancer, stage IIIB
Scottish Intercollegiate Guidelines
Network
1998 Management of lung cancer: a
national clinical guideline
recommended for use in
Scotland
58
Evidence-based Clinical staging; diagnosis; palliative
care; palliative treatment;
pathological/surgical staging; practice
organization; treatment-small cell lung
cancer, early stage, Pancoast tumor,
and stages I, II, IIIA, IIIB, and IV
SSO 1997 Lung cancer surgical practice
guidelines
59
Consensus-based Clinical staging; diagnosis; initial
evaluation/preparation; palliative
treatment; pathologic/surgical staging;
practice organization; screening and
early detection; treatment-small cell
lung cancer, stages I, II, IIIA, IIIB, IV
Timothy et al 1990 Workshop on consensus guidelines
for management of lung cancer
60
Consensus-based Clinical staging; diagnosis; pathological/
surgical staging
US DHHS 2000 Treating tobacco use and
dependence
61
US Preventive Services Task Force 1996 Screening for lung cancer
62
Evidence-based Prevention; screening and early
detection
*ACR American College of Radiology; BTS British Thoracic Society; CCOPGI Cancer Care Ontario Practice Guidelines Initiative; ESMO
European Society for Medical Oncology; NCCN National Comprehensive Cancer Network; ACS American Cancer Society; ASCO American
Society of Clinical Oncology; ATS American Thoracic Society; NCI National Cancer Institute; AATS American Association for Thoracic
Surgery; STS Society of Thoracic Surgeons; STSA Southern Thoracic Surgical Association; WTSA Western Thoracic Surgical Association;
AES American Endocurietherapy Society; ACCC Association of Community Cancer Centers; SCSGBI Society of Cardiothoracic Surgeons of
Great Britain and Ireland; HCHP Harvard Community Health Plan; RCR Royal College of Radiologists; SSO Society of Surgical Oncology;
DHHSI Department of Health and Human Services; ERS European Respiratory Society; CGG Cancer Guidance Group.
Taxotere; Aventis Pharmaceuticals; Bridgewater, NJ.
conflict of interest, specifically whether the guideline
was editorially independent from the funding body
and whether potential conflicts of interest were
reported for the members of the guideline develop-
ment group. The score in this domain was also poor,
with a mean score of 24% (range, 0 to 83%). Four
guidelines (8%) scored 50%. In 48 guidelines
(94%), potential conflicts of interest on the part of
guideline developers were not recorded.
Overall Recommendations: After reviewing all 51
guidelines and completing the AGREE instrument,
the reviewers came to a consensus with respect to an
overall recommendation for each guideline. As de-
scribed in the Materials and Methods section, we
recommended guidelines that we thought would be
useful to health-care providers and that demon-
strated good quality on the AGREE instrument. In
total, we recommended 19 of 51 guidelines (37%).
As noted in Table 3, for some guidelines (specifically,
the British Thoracic Society recommendations to
respiratory physicians for organizing the care of
patients with lung cancer
34
and the Scottish Inter-
collegiate Guidelines Network guideline on the man-
agement of lung cancer
58
), we based our recommen-
dation on the guidelines superior performance on
the AGREE instrument, while recognizing that
many of the guideline statements specific to practice
organization and review criteria would be directly
relevant only within the system for which they were
developed. Nonetheless, these documents serve as
examples of well-constructed and well-communicated
guidelines.
Agreement Among Reviewers
Table 4 demonstrates both the degree of agree-
ment beyond chance ( statistic) and the observed
simple agreement among the reviewers for the 23
items of the AGREE instrument. The values
indicate that overall agreement was poor to fair for
65% of the items and was moderate to substantial for
35% of the items. Observed agreement among re-
viewers was high, with 74% of items having moder-
ate-to-substantial agreement and 26% of items hav-
ing excellent agreement. The degree of agreement
appeared to be consistent across domains and did not
appear to be correlated with domains that were
quantitative vs qualitative in nature.
Discussion
Over the past decade, a significant number of
guidelines have been written and published on lung
cancer diagnosis and treatment. In total, they cover a
wide range of clinical topics with varying degrees of
evidence to support their recommendations.
Although many of the guidelines are classified as
Table 2Number of Guidelines, Recommendations, and Strength of the Evidence for Each Clinical Topic*
Topic Guidelines Recommendations
Evidence Strength
A B C
Clinical staging 15 (29) 74 (8) 1 (1) 14 (19) 59 (80)
Diagnosis 11 (22) 40 (5) 0 (0) 8 (20) 32 (80)
Follow-up/surveillance 9 (18) 29 (30) 0 (0) 2 (7) 27 (93)
Initial evaluation/preparation 7 (14) 26 (3) 0 (0) 5 (19) 21 (81)
Palliative care 3 (6) 8 (1) 1 (12) 2 (25) 5 (63)
Palliative treatment 5 (10) 12 (1) 2 (17) 4 (33) 6 (50)
Pathologic/surgical staging 15 (29) 61 (7) 5 (8) 10 (16) 46 (75)
Practice organization 6 (12) 73 (8) 3 (4) 7 (10) 63 (86)
Prevention 7 (14) 70 (8) 25 (36) 20 (29) 25 (36)
Screening and early detection 7 (14) 14 (2) 4 (29) 1 (7) 9 (64)
Treatment
Small cell lung cancer 14 (27) 118 (13) 18 (15) 12 (10) 88 (75)
Solitary pulmonary nodule 4 (8) 13 (1) 0 (0) 0 (0) 13 (100)
Early stage 2 (4) 11 (1) 4 (36) 0 (0) 7 (64)
Pancoast tumor, T4, special 8 (16) 25 (3) 0 (0) 4 (16) 21 (84)
Stage I 13 (26) 42 (5) 9 (21) 9 (21) 24 (57)
Stage II 14 (27) 56 (6) 9 (16) 15 (27) 32 (57)
Stage IIIA/potentially
resectable
15 (29) 61 (7) 13 (21) 9 (15) 39 (64)
Stage IIIB/nonresectable 16 (31) 62 (7) 12 (19) 6 (10) 44 (71)
Stage IV or IIIB with
malignant pleural effusion
21 (41) 82 (9) 16 (19) 3 (4) 63 (77)
Total 51 880 122 (14) 131 (15) 627 (71)
*Values given as No. (%). A Strong evidence; B weak evidence; C consensus.
Table 3AGREE Domain Scores for Clinical Practice Guidelines on Lung Cancer*
Guideline
Scope and
Purpose
Stakeholder
Involvement
Rigor of
Development
Clarity of
Presentation Applicability
Editorial
Independence
Overall
Recommendation
Practice guidelines in cardiothoracic
surgery
12
31 17 11 48 4 13 Would not
recommend
ACS guidelines for the early detection
of cancer; update of early detection
guidelines for prostate, colorectal,
and endometrial cancers; also
update 2001testing for early lung
cancer detection
13
67 27 38 38 28 17 Would not
recommended
ACR Appropriateness Criteria:
nonaggressive, nonsurgical
treatment of inoperable non-small
cell lung cancer
14
76 38 51 52 16 17 Would not
recommend
ACR Appropriateness Criteria: non-
small cell lung cancer, nonsurgical,
aggressive therapy
15
73 35 48 47 11 13 Would not
recommend
ACR Appropriateness Criteria: work-
up of the solitary pulmonary
nodule
16
76 35 49 48 13 13 Would not
recommend
ACR Appropriateness Criteria: follow-
up and retreatment of brain
metastases
17
69 31 42 48 6 8 Would not
recommend
ACR Appropriateness Criteria: follow-
up of non-small cell lung cancer
18
73 38 49 52 16 17 Would not
recommend
multiple brain metastases
19
72 40 51 46 14 21 Would not
recommend
neoadjuvant therapy for marginally
respectable (clinical N2) non-small
cell lung cancer
20
75 38 51 50 14 21 Would not
recommend
postoperative radiotherapy in non-
small cell lung canxcer
21
73 35 48 48 11 13 Would not
recommend
ACR Appropriateness Criteria: pre-
irradiation evaluation and
management of brain metastases
22
73 37 48 53 16 17 Would not
recommend
ACR Appropriateness Criteria: solitary
brain metastasis
23
78 37 49 52 16 18 Would not
recommend
ACR Appropriateness Criteria: staging
of bronchogenic carcinoma, non-
small cell lung carcinoma
24
73 33 48 48 11 13 Would not
recommend
ACR Appropriateness Criteria: staging
of non-small cell lung carcinoma
25
76 35 48 48 11 13 Would not
recommend
Consensus guidelines for high dose
rate remote brachytherapy in
cervical, endometrial, and
endobronchial tumors
26
44 13 21 21 14 17 Would not
recommend
Clinical practice guidelines for the
treatment of unresectable non-small
cell lung cancer
27
96 70 95 85 49 63 Recommend
Management of malignant pleural
effusions
28
67 15 33 42 17 17 Would not
recommend
Guidelines for percutaneous
transthoracic needle biopsy
29
64 23 8 48 6 17 Would not
recommend
Pretreatment evaluation of non-small-
cell lung cancer
30
20 3 10 22 4 3 Would not
recommend
Oncology patient management
guidelines: small cell lung cancer
31
62 17 17 62 13 13 Would not
recommend
Consensus statement on lung cancer
32
42 6 13 42 3 0 Would not
recommend
Table 3Continued*
Guideline
Scope and
Purpose
Stakeholder
Involvement
Rigor of
Development
Clarity of
Editorial
Independence
Overall
Recommendation
BTS guidelines: guidelines on the 94 54 89 79 33 83 Recommend
selection of patients with lung
cancer for surgery
33
BTS recommendations to respiratory
physicians for organizing the care of
patients with lung cancer
34
84 58 64 63 51 13 Recommend
The role of combination chemotherapy
in the initial management of
limited-stage small cell lung
cancer
35
89 46 76 65 3 33 Recommend
The role of single-agent docetaxel
(Taxotere) as a second-line
treatment for advanced non-small
cell lung cancer
36
83 46 81 73 6 33 Recommend
Altered fractionation of radical
radiation therapy in the
management of unresected non-
94 48 94 67 42 42 Recommend
Chemotherapy in stage IV (metastatic)
38
94 50 88 65 19 38 Recommend
Postoperative adjuvant chemotherapy
and/or radiation therapy in stage II
or IIIA, completely resected non-
39
92 44 91 63 11 46 Recommend
Prophylactic cranial irradiation in small
cell lung cancer
40
92 54 87 71 8 33 Recommend
The role of thoracic radiotherapy as an
adjunct to standard chemotherapy
in limited-stage small cell lung
cancer
41
92 50 91 73 8 42 Recommend
Unresected stage III non-small cell
lung cancer
42
92 44 89 73 28 38 Recommend
Use of gemcitabine in non-small cell
lung cancer
43
78 42 81 56 25 38 Recommend
Use of preoperative chemotherapy
with or without postoperative
radiotherapy in technically
resectable stage IIIA non-small cell
lung cancer
44
92 40 91 65 8 46 Recommend
Use of vinorelbine in non-small cell
lung cancer
45
84 46 86 54 17 38 Recommend
Guidance on commissioning cancer
services: improving outcomes in
lung cancer; the manual
46
87 63 73 75 98 10 Recommend
Clinical practice guidelines: smoking
prevention and cessation
47
78 31 24 75 14 13 Would not
recommend
ESMO minimum clinical
treatment and follow-up of non-
48
61 21 31 52 8 13 Would not
recommend
ESMO minimum clinical
treatment and follow-up of small
cell lung cancer
49
53 21 30 58 22 21 Would not
recommend
Screening for lung cancer
50
72 19 26 65 39 17 Would not
recommend
Lung cancer (PDQ): prevention
51
58 21 44 46 17 17 Would not
recommend
Lung cancer (PDQ): screening
52
44 21 48 52 17 17 Would not
recommend
Non-small cell lung cancer (PDQ):
treatment
53
44 13 13 45 2 10 Would not
recommend
evidence-based, a thorough review of their quality
utilizing the AGREE instrument led the authors to
recommend fewer than half of the guidelines. The
reasons for this are multifactorial. Overall, almost all
the guidelines performed poorly with respect to
applicability and editorial independence. Even those
guidelines that explicitly based their recommenda-
tions on evidence, such as those developed by the
Cancer Care Ontario Practice Guidelines Initia-
tive,
3545
failed to address issues of barriers to im-
plementation, monitoring criteria, and evidence of
pilot testing. Addressing such issues is necessary if
the guideline movement is to continue successfully.
Although few studies have assessed the impact of
guideline development on patient outcomes,
63
it has
been demonstrated that explicit guidelines can im-
prove clinical practice; however, improvement re-
quires rigorous evaluation. Well-developed guide-
lines should include the consideration of potential
barriers to guideline implementation, should supply
monitoring criteria to assess the guidelines impact, and
should provide evidence of pilot testing to ensure that
the guideline can be practically put to clinical use.
Another area where the lung cancer guidelines con-
sistently failed to performwas in the domain of editorial
independence. Poor performance in this domain could
represent true conflicts of interest between funding
sources and guideline development panels; alterna-
tively, it may simply reflect poor reporting on these
topics. The developers of the AGREE instrument
contacted the authors of each guideline they reviewed
to obtain background material that could inform the
reviewers ratings. For some items, this additional
communication may have provided more information
than we were able to obtain from reviewing the guide-
lines and any accompanying material we could obtain
from additional references or the World Wide Web.
For the lung cancer guidelines, documentation regard-
ing the issue of an individuals conflict of interest was
rarely stated. Explicit statements about whether or not
the funding body was independent editorially from the
guideline committee were also infrequent. Therefore,
poor performance in this domain could have been due
to our failure to obtain further information from each
guideline author. However, future guideline efforts
would benefit from clear documentation on this matter
within the text of the guideline document so that
readers will be able to determine for themselves
whether or not a conflict of interest potentially exists.
Table 4Agreement Among Reviewers for AGREE
Instrument Items
Strength of
Agreement Agreement
Items, No.
Statistic
Simple
Agreement
Poor 0.00 1 0
Slight 0.000.20 7 0
Fair 0.210.40 7 0
Moderate 0.410.60 7 3
Substantial 0.610.80 1 14
Excellent 0.80 0 6
Table 3Continued*
Guideline
Scope and
Purpose
Stakeholder
Involvement
Rigor of
Development
Clarity of
Editorial
Independence
Overall
Recommendation
Small cell lung cancer (PDQ):
treatment
54
53 21 49 56 14 17 Would not
recommend
NCCN practice guidelines for non-
55
53 22 21 72 4 7 Would not
recommend
NCCN practice guidelines for small
cell lung cancer
56
56 15 21 65 2 7 Would not
recommend
Guidelines on the nonsurgical
57
92 52 79 90 64 67 Recommend
Management of lung cancer: a national
clinical guideline recommended for
use in Scotland
58
83 63 79 77 67 25 Recommend
Lung cancer surgical practice
guidelines
59
64 21 2 29 0 17 Would not
recommend
Workshop on consensus guidelines for
60
61 31 14 15 14 0 Would not
recommend
Treating tobacco use and
dependence
61
97 63 94 88 61 79 Recommend
Screening for lung cancer
62
78 27 70 69 19 25 Recommend
Means 72 35 52 57 20 24
*Values given as percentage. See Table 1 for abbreviations not used in text.
Recommended based upon AGREE. Recommendations may not be relevant outside of the United Kingdom.
Recommended based upon AGREE. Utility of guidelines for single agent questionable.
One of the key factors regarding the adequacy of
the guidelines pertains to the rigor of development.
Despite the fact that most of the guidelines included
references to published literature, many did not
clearly delineate the literature review methodology
used or the mechanism by which recommendations
were formulated. This step is crucial in determining
whether the recommendations are truly based on the
evidence and in understanding how the evidence is
synthesized.
Patient preferences and experiences should be
factored into decisions regarding clinical care, espe-
cially in diseases such as lung cancer in which
treatments can have significant morbidity and can
impact on quality of life. Almost all of the guidelines
we reviewed would have benefited from more atten-
tion to this issue. This could be accomplished by
ensuring that all guideline committees have patient
representatives and that literature reviews specifi-
cally address quality of life when available. It would
also be helpful if greater efforts were made in the
research community to ensure that quality of life and
patient preferences are incorporated into research
protocols.
Implementation of practice guidelines also re-
quires attention to local practice patterns. For exam-
ple, the Cancer Guidance Group
46
guideline pro-
vides specific implementation strategies and review
criteria for their recommendations. Many of these
strategies and some of the corresponding monitoring
criteria are specific to the United Kingdom. None-
theless, this guideline provides an example of items
that should be incorporated by those undertaking a
guideline effort. Furthermore, those looking to apply
currently available guidelines can learn from their
detailed efforts to address implementation issues and
adapt what is relevant locally.
The recommendations that result from interpreta-
tion of the evidence also can vary among guidelines.
This variation could be a function of local bias,
difference in data interpretation, or a manifestation
of available resources. Nonetheless, one needs to be
careful when considering others guidelines for local
use and needs to ensure that the clinical data are
concordant with the evidence and clinical judgment.
Because of the variability noted in guidelines that
referenced the same studies, it is crucial that a
guideline effort have a clear methodology for going
from the evidence to the recommendations so that
the possibility of bias is minimized.
In this study, many of the conclusions are based on
a review utilizing the AGREE instrument. Although
this instrument is fairly new, it is one of the few
guideline assessment tools to demonstrate validity
and reliability. Furthermore, the areas covered by
the instrument are logical for anyone to consider
when conducting guideline development or evalua-
tion. A guideline that addresses the issues raised by
the AGREE instrument is more likely to be a
rigorously developed guideline. Nevertheless, the
AGREE instrument has some limitations. For one,
the interrater reliability was primarily slight to mod-
erate. Some of the variability may be due to differ-
ences in interpretation of several items where the
instructions were broad. For example, for item 22,
which is contained within the domain of editorial
independence, the statistic was only 0.14. This
slight agreement probably arises from the fact that
this question, which asks whether or not the guide-
line is editorially independent from the funding
body, is open to interpretation, with some reviewers
stating that the criterion was not met unless the
statement was explicitly made in the guideline, while
others interpreted the criterion to be met if the
funding agency was the government, as in several of
the UK and Canadian guidelines.
3445,58
However, in
this case, the simple agreement was still 58%. In
contrast, for item 12, which addresses whether or not
there is an explicit link between the recommenda-
tions and the supporting evidence, the statistic was
0.6. This is not unexpected, given that this item is
relatively straightforward. The issue of low interrater
reliability was observed in the previous version of
this instrument
64
and was accommodated by the use
of multiple reviewers, as well as through the refine-
ment of the instruments questions and instructions.
Moreover, the degree of agreement among raters
was good, with all items having moderate or better
agreement.
Another potential limitation of the AGREE instru-
ment concerns the validity of the responses to the
question on the overall assessment of the guideline.
Although the reviewers were instructed to consider
the domain scores when making a decision about
whether or not to recommend the guideline, no clear
rules were established as to how to weight the
differing domains. However, when reviewing
the assessments compared with the domain scores,
the responses appear to have validity. For each
guideline that was recommended, the overall domain
scores were 50% for at least three domains, with
an average of four domains with a score of 50%.
For guidelines that were not recommended, on
average only 1.3 domains had a score of 50%.
Furthermore, the score on the domain rigor of
development for recommended guidelines was
high. All scores were 50%, with an average score
of 84%. Conversely, for guidelines we did not rec-
ommend, the average score for this domain was only
33%.
In conclusion, a review of current lung cancer
guidelines demonstrates that many of the clinical
topics of interest have been considered by at least
one guideline. None covers all the necessary ele-
ments. Furthermore, although prior guidelines may
accurately reflect clinical practice, few adhere to the
standards set forth by the AGREE instrument. A
guideline effort for lung cancer, which adheres to all
the criteria explicit in the AGREE instrument and
clearly addresses each item in the guideline text,
would add substantially to the literature.
Appendix: AGREE Instrument*
Response categories for each question are as follows:
1. Strongly disagree
2. Disagree
3. Agree
4. Strongly agree
Scope and Purpose
1. The overall objectives of the guideline are specifically
described.
2. The clinical questions covered by the guideline are specif-
ically described.
3. The patients to whom the guideline is meant to apply are
specifically described.
Stakeholder Involvement
4. The guideline development group includes individuals
from all the relevant professional groups.
5. The patients views and preferences have been sought.
6. The target users of the guideline are clearly defined.
7. The guideline has been piloted among end-users.
Rigor of Development
8. Systematic methods were used to search for evidence.
9. The criteria for selecting the evidence are clearly de-
scribed.
10. The methods used for formulating the recommendations
are clearly described.
11. The health benefits, side effects, and risks have been
considered in formulating the recommendations.
12. There is an explicit link between the recommendations
and the supporting evidence.
13. The guideline was externally reviewed by experts prior to
its publication.
14. A procedure for updating the guideline is provided.
Clarity and Presentation
15. The recommendations are specific and unambiguous.
16. The different options for management of the condition are
clearly presented.
17. Key recommendations are easily identifiable.
18. The guideline is supported with tools for application.
Applicability
19. The potential organizational barriers in applying the rec-
ommendations have been discussed.
20. The potential cost implications of applying the recommen-
dations have been considered.
21. The guideline presents key review criteria for monitoring
and/or audit purposes.
Editorial Independence
22. The guideline is editorially independent from the funding
body.
23. Conflicts of interest of the guideline development mem-
bers have been recorded.
*The AGREE Instrument Collaboration was reprinted with the
kind permission of St. Georges Hospital Medical School (Lon-
don, UK; June 2001). Reprinted with amendments in September
2001. (Available at: http://www.agreecollaboration.org.)
ACKNOWLEDGMENT: The authors would like to acknowl-
edge Gregory P. Samsa, PhD, for help with statistical analysis.
References
1 Greenlee RT, Hill-Harmon MB, Murray T, et al. Cancer
statistics, 2001. CA Cancer J Clin 2001; 51:1536
2 Institute of Medicine Committee to Advise the Public Health
Service on Clinical Practice Guidelines. Clinical practice
guidelines: directions for a new program. Field MJ, Lohr KN,
eds. Washington, DC: National Academy Press, 1990
3 Swales JD. Guidelines on guidelines. J Hypertens 1993;
11:899903
4 Thomson R, McElroy H, Sudlow M. Guidelines on anticoag-
ulant treatment in atrial fibrillation in Great Britain: variation
in content and implications for treatment. BMJ 1998; 316:
509513
5 Shaneyfelt TM, Mayo-Smith MF, Rothwangl J. Are guide-
lines following guidelines? The methodological quality of
clinical practice guidelines in the peer-reviewed medical
literature. JAMA 1999; 281:19001905
6 Lacasse Y, Ferreira I, Brooks D, et al. Critical appraisal of
clinical practice guidelines targeting chronic obstructive pul-
monary disease. Arch Intern Med 2001; 161:6974
7 Cluzeau FA, Littlejohns P, Grimshaw JM, et al. Development
and application of a generic methodology to assess the quality
of clinical guidelines. Int J Qual Health Care 1999; 11:2128
8 The AGREE Collaboration. Appraisal of Guidelines For
Research & Evaluation (AGREE) instrument: September
2001. Available at: http://www.agreecollaboration.org. Ac-
cessed December 16, 2002
9 Graham ID, Calder LA, Hebert PC, et al. A comparison of
clinical practice guideline appraisal instruments. Int J Tech-
nol Assess Health Care 2000; 16:10241038
10 Ettinger DS, Cox JD, Ginsberg RJ, et al. NCCN Non-Small-
Cell Lung Cancer Practice Guidelines: the National Compre-
hensive Cancer Network. Oncology (Huntingt) 1996; 10:81
111
11 Fleiss JL. Measuring nominal scale agreement among many
raters. Psychol Bull 1971; 76:378382
12 American Association for Thoracic Surgery, Society of Tho-
racic Surgeons, Southern Thoracic Surgical Association, et al.
Practice guidelines in cardiothoracic surgery. Ann Thorac
Surg 1993; 56:12031213
13 Smith RA, von Eschenbach AC, Wender R, et al. American
Cancer Society guidelines for the early detection of cancer:
update of early detection guidelines for prostate, colorectal,
and endometrial cancers. Also: update 2001; testing for early
lung cancer detection (erratum appears in CA Cancer J Clin
2001; 51:150). CA Cancer J Clin 2001; 51:3875
14 American College of Radiology. ACR Appropriateness Crite-
ria: nonaggressive, nonsurgical treatment of inoperable
non-small cell lung cancer (NSCLC), 2000. Available at:
http://www.acr.org/dyna/?idappropriateness_criteria. Ac-
cessed December 16, 2002
15 American College of Radiology. ACR Appropriateness
Criteria: non-small cell lung cancer, nonsurgical, aggres-
sive therapy, 2000. Available at: http://www.acr.org/dyna/
?idappropriateness_criteria. Accessed December 16,
2002
ria: work-up of the solitary pulmonary nodule, 2000. Avail-
able at: http://www.acr.org/dyna/?idappropriateness_criteria.
Criteria: follow-up and retreatment of brain metastasis, 1999.
Available at: http://www.acr.org/dyna/?idappropriateness_criteria.
ria: follow-up of non-small cell lung cancer, 1999. Avail-
Criteria: multiple brain metastases, 1999. Available at: http://
www.acr.org/dyna/?idappropriateness_criteria. Accessed
December 16, 2002
Criteria: neoadjuvant therapy for marginally resectable
(clinical N2), non-small cell lung cancer, 1999. Available
at: http://www.acr.org/dyna/?idappropriateness_criteria.
Criteria: postoperative radiotherapy in non-small cell
lung cancer, 1999. Available at: http://www.acr.org/dyna/
?idappropriateness_criteria. Accessed December 16, 2002
Criteria: pre-irradiation evaluation and management of
brain metastases, 1999. Available at: http://www.acr.org/
dyna/?idappropriateness_criteria. Accessed December 16,
2002
ria: solitary brain metastasis, 1999. Available at: http://www.
acr.org/dyna/?idappropriateness_criteria. Accessed Decem-
ber 16, 2002
Criteria: staging of bronchogenic carcinoma, non-small
cell lung carcinoma, 1999. Available at: http://www.acr.org/
dyna/?idappropriateness_criteria. Accessed December 16,
2002
ria: staging of non-small cell lung carcinoma, 1999. Avail-
26 Nag S, Abitbol AA, Anderson LL, et al. Consensus guidelines
for high dose rate remote brachytherapy in cervical, endome-
trial, and endobronchial tumors: Clinical Research Commit-
tee, American Endocurietherapy Society. Int J Radiat Oncol
Biol Phys 1993; 27:12411244
27 American Society of Clinical Oncology. Clinical practice
guidelines for the treatment of unresectable non-small-cell
lung cancer. J Clin Oncol 1997; 15:29963018
28 American Thoracic Society. Management of malignant pleu-
ral effusions. Am J Respir Crit Care Med 2000; 162:1987
2001
29 Sokolowski JW Jr, Burgher LW, Jones FL Jr, et al. Guidelines
for percutaneous transthoracic needle biopsy. Am Rev Respir
Dis 1989; 140:255256
30 American Thoracic Society, and European Respiratory Soci-
ety. Pretreatment evaluation of non-small-cell lung cancer.
Am J Respir Crit Care Med 1997; 156:320332
31 Association of Community Cancer Centers. Oncology patient
management guidelines: small cell lung cancer. Rockville,
MD: Association of Community Cancer Centers, 2000
32 Biesalski HK, de Mesquita BB, Chesson A, et al. Consensus
statement on lung cancer: Lung Cancer Panel. Eur J Cancer
Prev 1997; 6:316322
33 British Thoracic Society, and Society of Cardiothoracic Sur-
geons of Great Britain, and Ireland Working Party. BTS
guidelines: guidelines on the selection of patients with lung
cancer for surgery. Thorax 2001; 56:89108
34 British Thoracic Society Standards of Care Committee Lung
Cancer Working Party. BTS recommendations to respiratory
physicians for organising the care of patients with lung cancer.
Thorax 1998; 53(suppl):S1S8
35 Laurie S, Logan DM, Markman BR, et al. The role of
combination chemotherapy in the initial management of
limited-stage small-cell lung cancer: Cancer Care Ontario
Practice Guidelines Initiative; practice guideline No. 7-13-1;
March 2001. Available at: http://www.ccopebc.ca/guidelines/
lung/cpg7_13-1.html. Accessed December 16, 2002
36 Logan D, Laurie S, Markman BR, et al. The role of single-
agent docetaxel (Taxotere) as a second-line treatment for
advanced non-small-cell lung cancer: Cancer Care Ontario
Practice Guidelines Initiative; practice guideline report No.
7-7-2; January 2001. Available at: http://www.ccopebc.ca/
guidelines/lung/cpg7_7-2.html. Accessed December 16, 2002
37 Yu E, Lochrin C, Dixon P, et al. Altered fractionation of
radical radiation therapy in the management of unresectable
non-small cell lung cancer: Cancer Care Ontario Practice
Guidelines Initiative; practice guideline No. 7-12; October
2000. Available at: http://www.ccopebc.ca/guidelines/lung/
cpg7_12.html. Accessed December 16, 2002
38 Lopez PG, Stewart DJ, Newman TE, et al. Chemotherapy in
stage IV (metastatic) non-small cell lung cancer: Cancer Care
Ontario Practice Guidelines Initiative; practice guideline No.
7-2;April 2000.Availableat:http://www.ccopebc.ca/guidelines/
lung/cpg7_2.html. Accessed December 16, 2002
39 Logan DM, Lochrin CA, Darling G, et al. Postoperative
adjuvant chemotherapy and/or radiation therapy in stage II or
IIIA completely resected non-small cell lung cancer: Cancer
Care Ontario Practice Guidelines Initiative; practice guide-
line No. 7-1; April 2000. Available at: http://www.ccopebc.ca/
guidelines/lung/cpg7_1-1.html. Accessed December 16, 2002
40 Kotalik J, Yu E, Markman BR, et al. Prophylactic cranial ir-
radiation in small cell lung cancer: Cancer Care Ontario Practice
Guidelines Initiative; practice guideline report No. 7-13-2; De-
cember 2000. Available at: http://www.ccopebc.ca/guidelines/
41 Okawara G, Gagliardi A, Evans WK, et al. The role of thoracic
radiotherapy as an adjunct to standard chemotherapy in
limited-stage small cell lung cancer: Cancer Care Ontario
Practice Guidelines Initiative; practice guideline No. 7-13-3;
October 2000. Available at: http://www.ccopebc.ca/guidelines/
42 Okawara G, Rusthoven J, Newman T, et al. Unresected Stage
III non-small cell lung cancer: Cancer Care Ontario Practice
Guidelines Initiative; practice guideline No. 7-3; April 2000.
Available at: http://www.ccopebc.ca/guidelines/lung/cpg7_3.
html. Accessed December 16, 2002
43 Evans WK, Kocha W, Gagliardi A, et al. Use of gemcitabine
in non-small-cell lung cancer: Cancer Care Ontario Practice
44 Goss G, Paszat L, Newman T, et al. Use of preoperative
chemotherapy with or without postoperative radiotherapy in
technically resectable stage IIIA non-small cell lung cancer:
Cancer Care Ontario Practice Guidelines Initiative; practice
guideline No. 7-4; April 2000. Available at: http://www.ccopebc.
ca/guidelines/lung/cpg7_4.html. Accessed December 16, 2002
45 Goss GD, Logan DM, Newman TE, et al. Use of vinorelbine
in non-small cell lung cancer: Cancer Care Ontario Practice
46 Cancer Guidance Group. Guidance on commissioning cancer
services; improving outcomes in lung cancer: the manual.
London, UK: National Health Service Executive, 1998
47 Colle` ge des me decins du Que bec. Clinical practice guidelines:
smoking prevention and cessation; May 1999. Available at:
http://www.cmq.org/tabacang.pdf. Accessed December 16, 2002
48 European Society for Medical Oncology. ESMO minimum
clinical recommendations for diagnosis, treatment and fol-
low-up of non-small-cell lung cancer (NSCLC). Ann Oncol
2001; 12:10491050
49 European Society for Medical Oncology. ESMO minimum
clinical recommendations for diagnosis, treatment and fol-
low-up of small-cell lung cancer (SCLC). Ann Oncol 2001;
12:10511052
50 Harvard Community Health Plan Adult Screening and Pre-
vention Task Force. Screening for lung cancer. In: Harvard
Community Health Plan clinical guidelines and algorithims.
Boston, MA: Harvard Community Health Plan, 1989
51 National Cancer Institute. Lung cancer (PDQ): preven-
tion; March 2001. Available at: http://www.cancer.gov/
cancer_information/pdq. Accessed December 16, 2002
52 National Cancer Institute. Lung cancer (PDQ): screening;
April 2001. Available at: http://www.cancer.gov/cancerin-
formation/pdq. Accessed December 16, 2002
53 National Cancer Institute. Non-small cell lung cancer (PDQ):
treatment; March 2001. Available at: http://www.cancer.gov/
cancerinformation/pdq. Accessed December 16, 2002
54 National Cancer Institute. Small cell lung cancer (PDQ):
treatment; November 2000. Available at: http://www.cancer.
gov/cancer_information/pdq. Accessed December 16, 2002
55 National Comprehensive Cancer Network. NCCN practice
guidelines for non-small-cell lung cancer. In: The complete
library of NCCN guidelines in oncology [CD-ROM]. Rock-
ledge, PA: National Comprehensive Cancer Network, 2000
56 National Comprehensive Cancer Network. NCCN practice
guidelines for small-cell lung cancer. In: The complete library
of NCCN guidelines in oncology [CD-ROM]. Rockledge, PA:
National Comprehensive Cancer Network, 2000
57 Royal College of Radiologists Clinical Oncology Information
Network. Guidelines on the non-surgical management of lung
cancer. Clin Oncol (R Coll Radiol) 1999; 11(suppl):S1S53
58 Scottish Intercollegiate Guidelines Network. Management of
lung cancer: a national clinical guideline recommended for
use in Scotland. Edinburgh, Scotland: Scottish Intercollegiate
Guidelines Network, 1998
59 Ginsberg R, Roth J, Fergusson M. Lung cancer surgical
practice guidelines. Society of Surgical Oncology practice
guidelines. Oncology (Huntingt) 1997; 11:889892, 895
60 Timothy AR, Ash D, Corris P, et al. Workshop on consensus
guidelines for management of lung cancer. Clin Oncol (R Coll
Radiol) 1990; 2:97101
61 Fiore MC, Bailey WC, Cohen SJ, et al. Treating tobacco use
and dependence: clinical practice guideline. Rockville, MD:
US Department of Health and Human Services, Public
Health Service, 2000
62 US Preventive Services Task Force. Screening for lung
cancer. In: Guide to clinical preventive services. 2nd ed.
Baltimore, MD: Williams & Wilkins, 1996; 135139
63 Grimshaw JM, Russell IT. Effect of clinical guidelines on
medical practice: a systematic review of rigorous evaluations.
Lancet 1993; 342:13171322
64 Cluzeau F, Littlejohns P, Grimshaw J, et al. Appraising
clinical guidelines and the development of criteria: a pilot
study. J Interprof Care 1995; 9:227235
2003;123;7-20 Chest
Kolimaga and Douglas C. McCrory
Linda H. Harpole, Michael J. Kelley, Gilbert Schreiber, Eric M. Toloza, Jane
Cancer
Assessment of the Scope and Quality of Clinical Practice Guidelines in Lung
& Services
Updated Information
References
BIBL
free at:
This article cites 25 articles, 12 of which you can access for
Citations
otherarticles
This article has been cited by 1 HighWire-hosted articles:
Reprints
2003;123;21-49 Chest
Anthony J. Alberg and Jonathan M. Samet
Epidemiology of Lung Cancer
ISSN: 0012-3692.
Epidemiology of Lung Cancer*
Anthony J. Alberg, PhD, MPH; and Jonathan M. Samet, MD, MS
In the United States, lung cancer remains the leading cause of cancer death in both men and
women even though an extensive list of risk factors has been well-characterized. Far and away the
most important cause of lung cancer is exposure to tobacco smoke through active or passive
smoking. The reductions in smoking prevalence in men that occurred in the late 1960s through
the 1980s will continue to drive the lung cancer mortality rates downward in men during the first
portion of this century. This favorable trend will not persist unless further reductions in smoking
prevalence are achieved. (CHEST 2003; 123:21S49S)
Key words: air pollution; asbestos; cigarette smoking; epidemiology; lung cancer; nutrition; passive smoking;
occupation; radiation; tobacco
Abbreviations: CPS Cancer Prevention Study; ETS environmental tobacco smoke; FTC Federal Trade
Commission; IARC International Agency for Research on Cancer; LET linear energy transfer
E
pidemiologic evidence is the foundation for pri-
mary and secondary disease prevention. Epide-
miologic approaches are used to track the occur-
rence of disease, to characterize natural history, and
to identify determinants of disease. The benefits of
intervention programs, whether based in risk factor
interventions or screening, also are assessed using
epidemiologic approaches. For lung cancer, routine
mortality statistics have confirmed the clinical im-
pression that the disease became more frequent
during the first half of the 20th century. Case-control
and cohort studies,
13
the epidemiologic study de-
signs that are used to evaluate exposure-disease
associations, causally linked smoking to lung cancer
in investigations reported from the 1950s forward. As
we have continued to follow lung cancer incidence
and mortality rates, we have readily shown that their
rise and decline parallel past trends of cigarette
smoking.
4
The epidemiologic evidence and the com-
plementary biological understanding of respiratory
carcinogenesis have unassailably supported the con-
clusion that smoking causes lung cancer. Epidemio-
logic findings are also relevant to patient care, as
skilled clinicians weigh alternative diagnoses de-
pending on the risk factor profiles of patients.
This article provides a summary of the epidemio-
logic evidence on lung cancer, with an emphasis on
issues that are currently relevant to prevention. This
literature is now extraordinarily large, and we have
not attempted to conduct a comprehensive review
and systematic synthesis. Such syntheses have been
periodically carried out by expert review groups,
including the committees assembled to prepare the
US Surgeon Generals reports on smoking and
health, and other federal documents and expert
committees of other governments and organizations,
including the Royal College of Physicians and the
Scientific Committee on Tobacco of the United
Kingdom, and the International Agency for Research
on Cancer (IARC) of the World Health Organiza-
tion. Several relevant reports can be anticipated from
these groups, including the 2002 and 2003 reports of
the US Surgeon General addressing active and pas-
sive smoking, respectively, and a planned new IARC
monograph on tobacco smoking. The general topic
was reviewed in a 1994 monograph.
5
At the end of the 20th century, lung cancer had
become one of the worlds leading causes of prevent-
able death.
6
It was a rare disease at the start of that
century, but exposures to new etiologic agents and
an increasing lifespan combined to make lung cancer
a scourge of the 20th century. While tobacco had
been widely used throughout the world for centuries,
the present pandemic of lung cancer followed the
introduction of manufactured cigarettes with addic-
tive properties, which resulted in a new pattern of
sustained exposure of the lung to inhaled carcino-
gens.
7
German scientists in Nazi Germany con-
ducted some of the earliest research on the links
between smoking and lung cancer.
8
By the early
1950s, epidemiologic studies
911
in Britain and the
*From the Department of Epidemiology, Johns Hopkins Univer-
sity, Bloomberg School of Public Health, Baltimore, MD.
Dr. Alberg is a recipient of a K07 Award (No. CA73790) from the
National Cancer Institute.
Correspondence to: Anthony J. Alberg, PhD, MPH, Department
of Epidemiology, Johns Hopkins University, Bloomberg School of
Public Health, 615 North Wolfe St, Baltimore, MD 21205; e-mail:
aalberg@jhsph.edu
United States using the case-control method had
shown that cigarettes were strongly associated with
the risk of lung cancer. This association was corrob-
orated by the pioneering cohort studies of British
physicians, US veterans of the armed forces, and
volunteers recruited by the American Cancer Soci-
ety.
12
By 1964, the evidence was sufficient to support
a conclusion by the US Surgeon General that ciga-
rette smoking caused lung cancer.
12
The Royal Col-
lege of Physicians had reached the same conclusion
2 years previously.
13
Passive smoking, the involun-
tary inhalation of tobacco smoke by nonsmokers, also
has been found to cause lung cancer.
14,15
While its predominant cause is now well-known
(ie, tobacco smoking), there are other causes as well,
some acting in concert with smoking to synergisti-
cally increase risk. The suspicion that radon was a
cause of lung cancer in underground miners, which
was raised early in the century, led to what was
probably the first occupational respiratory carcino-
gen to be identified.
16
Radon in indoor environments
is now considered to be a significant cause of lung
cancer.
17
The list of human occupational causes of
lung cancer also includes arsenic, asbestos, chro-
mates, chloromethyl ethers, nickel, polycyclic aro-
matic hydrocarbons, radon progeny, and other
agents.
18
Outdoor air pollution, which includes com-
bustion-generated carcinogens, is also considered to
contribute to the lung cancer burden in urban
dwellers. Indoor air contains several respiratory car-
cinogens, including radon, asbestos, and cigarette
smoke. In some developing countries, exposure to
fumes from cooking stoves and fires is associated
with lung cancer risk. Beginning in the 1980s, asso-
ciations of diet with lung cancer risk have been
vigorously investigated with the anticipation that
dietary micronutrients might be found that modify
the high lung cancer risk in smokers. The biological
basis for the prevention of cancer through supple-
mentation of micronutrients is addressed in another
article in this supplement.
Even though the epidemiology of lung cancer has
been extensively investigated for 50 years, there
are still active areas of research, some quite relevant
to prevention. Investigation of lung cancer and diet
continues, using both observational and experimen-
tal approaches, and concern remains over the risk of
indoor and outdoor pollutants including, for exam-
ple, radon and diesel emissions. There also has been
a need for research to track the risks of smoking over
time, as the cigarette has evolved in its design
characteristics and yields of tar and nicotine, as
assessed by standard protocol using a machine, have
declined since the 1950s. The histologic characteris-
tics of lung cancer in a number of developed coun-
tries, including the United States, also have changed
in recent decades so that the frequency of adenocar-
cinoma has risen and that of squamous cell carci-
noma has declined.
4
There is also emerging evidence
on the genetic determinants of lung cancer risk. A
current research approach, termed molecular epide-
miology, melds the population and laboratory tools
that are used to address susceptibility to environ-
mental carcinogens. While the evidence from the
traditional epidemiologic approaches conclusively
established the carcinogenicity of tobacco smoke,
molecular epidemiology should characterize the se-
quence of molecular and cellular changes as a non-
malignant cell becomes malignant and the factors
determining susceptibility to tobacco smoke. Bi-
omarkers of exposure, dose, susceptibility, and ge-
netic damage may allow epidemiologic investigations
to uncover specific pathways of human lung carcino-
genesis.
Patterns of Occurrence
Temporal Trends
Because of the high case-fatality rate of lung
cancer, the incidence and mortality rates are nearly
equivalent, and, consequently, routinely collected
vital statistics provide a long record of the occur-
rence of lung cancer. We are presently amid an
epidemic of lung cancer that dates to the mid-20th
century (Fig 1).
1921
Lung cancer was rare until the
disease began a sharp rise around 1930 that culmi-
nated by mid-century with lung cancer becoming the
leading cause of cancer death among men.
22
The
epidemic among women followed that among men,
with a sharp rise in rates from the 1960s to the
present, propelling lung cancer to become the most
frequent cause of female cancer mortality.
22
The
epidemic among women not only occurred later, but
will not peak at as high a level as that among men
because smoking prevalence crested at substantially
higher levels among men than among women.
4,23,24
An examination of the time trends of age-specific
lung cancer mortality rates in the United States
further highlights the differing epidemic patterns in
men compared to women (Fig 2).
20,21,23,25,26
In the
older age groups, the rates continue to increase in
both sexes, but the rates of increase are decelerating
more significantly in men than in women.
23
The rates
of lung cancer are now decreasing in the younger age
groups, decreases that are more pronounced for men
but also now are becoming evident in women.
23,27
As
the younger birth cohorts age, their reduced risk of
lung cancer should thus translate into substantial
reductions in the overall occurrence of lung cancer,
reductions that will probably be more favorable for
men than for women. In a recent analysis of lung
cancer mortality in the United States from 1970 to
1997, Jemal and colleagues
23
found that the rates of
decrease in younger men and women (ie, those born
after 1950) were moderating, even though the de-
cline continues. These investigators suggested that
the moderation could reflect patterns of smoking
initiation. In some other countries, lung cancer rates
continue to rise in all age groups.
Notable shifts have taken place in the incidence
rates of lung cancer by histologic type.
28
After
steadily increasing in occurrence during the period
from 1973 to 1987, adenocarcinoma supplanted
Figure 2. US age-specific mortality rates (white men and women) by 2-year age intervals from 26 to
27 years of age through 48 to 49 years of age, plotted against the birth cohort. Adapted from the
American Cancer Society.
22
Figure 1. Lung cancer mortality rates for the United States from 1930 to 1998, age-standardized to
the 1970 US population. Adapted from Gordon et al,
19
and Mckay et al,
20
and Ries et al.
21
squamous cell carcinoma as the most frequent form
of lung cancer (Table 1).
28
Adenocarcinoma in-
creased markedly in all race-sex subgroups (Table 1).
Race and Ethnicity
The patterns of occurrence of lung cancer by race
and ethnicity make lung cancer a relevant disease for
those concerned with the health of women and minor-
ities. Whereas lung cancer incidence rates are similar
among African-American and white women, lung can-
cer occurs about 50% more frequently among African-
American men than among white men.
21
The marked
reduction in cigarette smoking that has occurred
among African-American youths
29
forecasts a possible
reversal of this trend, and, if this trend persists, declines
in the incidence of lung cancer among African-Ameri-
cans can be expected.
Lung cancer mortality rates among Hispanics,
Native Americans, and Asians/Pacific Islanders are
significantly lower than rates among African-
Americans and non-Hispanic whites.
30
Nevertheless,
lung cancer occurs sufficiently often among these
groups to pose a considerable public health burden.
Geographic Patterns
Lung cancer is the most commonly diagnosed cancer
worldwide,
30
but its geographic distribution shows
marked regional variation.
31
The global variation in
age-standardized incidence rates is greater than four-
fold among men, and a fivefold among women (Fig
3).
32
Because of differences in cancer registration be-
tween countries, caution is needed in interpreting these
data. However, this marked variation in rates cannot be
explained on the basis of diagnostic practices and data
quality alone. Lung cancer tends to be most common in
developed countries, particularly in North America and
Europe, and less common in developing countries,
particularly in Africa and South America.
33
The low
rates of lung cancer in Africa
34
are comparable to
United States rates in 1930, when rates of lung cancer
were under 5 cases per 100,000 for both sexes.
19
In
contrast, African-Americans in the United States, an
epicenter of the disease, now experience lung cancer
incidence rates that are among the highest in the world.
As the lung cancer epidemic begins to subside in the
developed countries, it is on the rise in the developing
world.
35
Within countries, lung cancer incidence among
men invariably outpaces that in women, by well over
100% in most nations. The international rankings of
lung cancer incidence for men and women from the
same countries tend to differ only slightly, so that the
highest rates of lung cancer occur in the same
regions of the world for both sexes.
Substantial geographic variation in lung cancer mor-
tality rates also has been observed within countries.
Trends in its regional distribution can provide clues
Table 1Age-Adjusted Incidence Rates per 100,000 Population of Lung Cancer by Histologic Subtype and
Time Period
Group Carcinoma Subtype 19731977* 19781982* 19831987* 19921998
Total Total 39.5 46.8 51.4 54.0
Squamous cell 13.4 15.1 15.3 11.5
Adenocarcinoma 10.5 14.2 16.7 18.9
Small cell 5.9 8.2 9.4 7.9
Large cell 0.0 3.9 4.9 3.8
White men Squamous cell carcinoma 24.3 26.8 25.5 17.2
Small cell 9.5 12.5 13.1 9.7
Large cell 0.0 5.9 7.2 4.6
White women Squamous cell carcinoma 4.0 5.5 6.6 6.8
Small cell 3.4 5.5 7.1 7.5
Large cell 0.0 2.2 3.1 2.8
African-American men Squamous cell carcinoma 43.9 46.3 48.5 30.8
Small cell 9.5 13.3 14.0 10.1
Large cell 0.0 8.0 10.8 8.1
African-American women Squamous cell carcinoma 5.6 6.8 9.5 9.3
Small cell 3.6 3.9 6.0 6.0
Large cell 0.0 2.0 3.0 3.1
*Adapted from Travis et al.
28
Calculated from online SEER data using Cancer Query Systems software. (National Cancer Institute; Bethesda MD; available at http://
seer.cancer.gov/canques/.)
about the determinants of lung cancer. In the past,
rates tended to be highest in urban areas, which led to
conjecture that air pollution might be a cause of the
lung cancer epidemic.
36
Later on, several hypothe-
ses
37,38
were prompted by patterns observed in a
systematic review of US lung cancer mortality rates for
the period 1950 to 1969,
39
particularly the rates among
men. For example, high lung cancer rates in coastal
areas were postulated to reflect employment in ship-
yards with attendant asbestos exposure. This hypothesis
then was tested in a series of population-based, case-
control studies, which showed that employment in the
shipbuilding industry was indeed associated with an
excess risk of lung cancer.
40
Another shift then took
place in the distribution of lung cancer within the
United States, with lung cancer mortality rates among
white men becoming highest in the South and lower in
the Northeast.
41
This fluidity in the geographic varia-
tion underscores the value of regularly monitoring lung
cancer mortality patterns.
The Etiology of Lung Cancer
Overview
Although the causes of lung cancer are almost
exclusively environmental, it is likely that there is
substantial individual variation in the susceptibility to
Figure 3. Age-adjusted lung cancer death rates from 1986 to 1988 per 100,000 population for men and
women. Adapted from The National Cancer Institute.
22
respiratory carcinogens. The risk of the disease can
be conceptualized as reflecting the joint conse-
quences of the interrelationship between (1) expo-
sure to etiologic (or protective) agents and (2) the
individual susceptibility to these agents. The term
environment in its broadest sense may influence the
risk of disease through direct exposures or indirectly
by affecting the likelihood of exposure to exogenous
agents. Given the multifactorial etiology of lung
cancer, synergistic interactions among risk factors
may have substantial consequences for lung cancer
risk. These interactions typically have been consid-
ered on an agent-by-agent basis, as for example the
synergistic effect of cigarette smoking on the lung
cancer risk from asbestos exposure.
42
Our emerging
understanding of cancer genetics indicates the addi-
tional relevance of gene-environment interactions.
When the environment is considered more holis-
tically, an even broader set of interactions can be
proposed. For example, socioeconomic status is as-
sociated with a constellation of the following inter-
acting determinants of lung cancer risk: smoking;
diet; and exposures to inhaled agents in the work-
place and general environment. Lower socioeco-
nomic status is associated with an unfavorable profile
for all of these factors. Risk factor patterns at the
individual and societal levels identify persons and
populations at high risk and can be used to target
prevention approaches toward those who are at
greatest risk.
Given the many risk factors that have been
identified for lung cancer, a practical question is
the relative contribution of these factors to the
overall burden of lung cancer. The population
attributable risk approach takes into account the
magnitude of the relative risk that is associated
with an exposure along with the likelihood of
exposure in the general population. Because of the
interactions between exposures, the combined
population attributable risk for diseases such as
lung cancer that have a well-characterized set of
important risk factors can exceed 100%. As re-
viewed below, population attributable risk esti-
mates for lung cancer indicate that in the United
States, active smoking is responsible for 90% of
lung cancer cases, occupational exposures to car-
cinogens account for approximately 9 to 15% of
lung cancer cases, radon causes 10% of lung
cancer cases,
17
and outdoor air pollution accounts
for perhaps 1 to 2% of lung cancer cases.
43
The
precise contribution of nutritional factors remains
to be determined, but dietary factors have been
hypothesized to account for approximately 20%
(range, 10 to 30%) of the lung cancer burden.
44
Environmental and Occupational Agents
Smoking
Overview: A single etiologic agent, cigarette smok-
ing, is by far the leading cause of lung cancer,
accounting for approximately 90% of lung cancer
cases in the United States and other countries where
cigarette smoking is common.
45
Compared to never-
smokers, smokers have about a 20-fold increase in
lung cancer risk at present. Few exposures to envi-
ronmental agents convey such risks for any disease.
In general, trends of lung cancer occurrence closely
reflect patterns of smoking, but rates of occurrence
lag smoking rates by about 20 years. Analyses using
statistical modeling techniques show a tight associa-
tion between national mortality rates and smoking.
46
The unequivocal causal association of cigarette
smoking with lung cancer is one of the most thor-
oughly documented causal relationships in biomed-
ical research.
7,47
The burden of lung cancer that is attributable to
smoking has been extensively documented. Using an
attributable risk approach, the Centers for Disease
Control and Prevention has documented the num-
bers of deaths caused in the United States by
smoking-related lung cancer. For 1990, that number
was 117,000.
48
Peto et al
45
have used a different
attributable risk method to quantify the burden of
smoking-related deaths from lung cancer in the
major developed countries. For 1990, the US total
was 127,000 deaths, which was the highest in the
world, with country-specific estimates ranging down
to 150 deaths for Tajikistan. The total number of
smoking-related deaths for the developed countries
was 457,371.
45
Peto and colleagues forecast a stag-
gering future burden for China, which now has one
third of the worlds smokers.
49
The numbers are
predicted to reach several millions by mid-century.
Cigar smoking is also an established cause of lung
cancer.
50
The lung cancer risks associated with cigar
smoking are substantial, but are less than the risks
observed for cigarette smoking due to differences in
smoking frequency and depth of inhalation. The
same pattern holds true for pipe smoking.
51
Quantitative Risks: The risk of lung cancer among
cigarette smokers increases with the duration of
smoking and the number of cigarettes smoked per
day
48,52
(Table 2). This observation has been made
repeatedly in cohort and case-control studies. Risk
models have been derived to quantitatively estimate
how lung cancer risk varies with the number of
cigarettes smoked, the duration of smoking, and age.
Such models are useful for estimating the future
burden of lung cancer under various scenarios of
tobacco control. In one widely cited analysis, Doll
and Peto
52
proposed a quantitative model for lung
cancer risk based on data from the cohort study of
British physicians. This model predicted a stronger
effect of duration of smoking than of amount smoked
per day. Thus, a tripling of the number of cigarettes
smoked per day was estimated to triple the risk,
whereas a tripling of the duration of smoking was
estimated to increase the lung cancer risk 100-fold.
53
These quantitative dimensions of the dose-response
relationship between smoking and lung cancer have
implications concerning the now widespread occur-
rence of smoking among youths. Those persons
starting to smoke at younger ages have a greater
likelihood of becoming heavier smokers and remain-
ing smokers.
54
The exponential effect of the duration
of smoking on lung cancer risk markedly increases
the lifetime risk for those who become regular
smokers in childhood. Thus, those who initiate smok-
ing earlier in life are most likely to develop lung
cancer and are most likely to do so at younger ages.
Prevention approaches that delay the age of onset of
smoking in a population could have a substantial
impact on the incidence of lung cancer by shortening
the duration of smoking. In considering the likeli-
hood of lung cancer in a particular patient, clinicians
should give more weight to the duration of smoking
and less weight to actual age.
Smoking Cessation: Cigarette smokers can benefit
at any age by quitting smoking. The likelihood of
developing lung cancer decreases among those who
quit smoking compared to those who continue to
smoke
55
(Table 3
56
). As the period of abstinence
from smoking cigarettes increases, the risk of lung
cancer decreases
55
(Table 3). However, even for
periods of abstinence of 40 years, the risk of lung
cancer among former smokers remains elevated
compared to never-smokers
55,56
(Table 3). The ben-
efits derived from smoking cessation also depend on
the duration of smoking. For a given period of
abstinence, the decrease in risk increases as the
duration of smoking decreases.
55
In general, stud-
ies
57
have shown comparable reductions in risk
following smoking cessation, regardless of sex, type
of tobacco smoked, and histologic type of lung
cancer.
The Changing Cigarette: The composition of cig-
arettes has evolved considerably since the 1950s. The
marketplace has shifted from mainly unfiltered cig-
arettes to predominantly filtered cigarettes. The
filters in use in the United States are predominantly
cellulose acetate, while charcoal filters are used
extensively in Japan and some other countries.
58
In
the mid-1960s, ventilation holes were added to the
filter, which dilute the smoke with air drawn through
them. However, smokers can use their fingers to
readily block the holes, which are left unblocked by
the machines that are used to test cigarettes. There
also have been substantial changes in the design of
the cigarette and in the tobacco used. Reconstituted
tobacco has been used increasingly since the 1960s,
and there have been changes to the cigarette papers
and additives used.
58
A concomitant shift toward lowered levels of tar
and nicotine, as measured by a smoking machine,
has occurred (Fig 4). Cigarette tar is the label
given to the condensable residue of cigarette
smoke (ie, the total particulate matter from ciga-
rette smoke that is deposited on the filter ma-
chines less the moisture and nicotine). Tar is a
complex mixture that includes many chemicals
that are cancer initiators and/or promoters.
59
Tar
and nicotine yields are measured with a smoking
machine according to a standardized protocol
established by the Federal Trade Commission
(FTC) that specifies such details as puff volume,
the frequency of puffing, and the length to which
the cigarette is to be smoked.
60
Table 2Age-Specific Lung Cancer Mortality Rates
per 100,000 Population Among Men and Women of
Comparable Smoking Levels in the CPS-II*
Age Group,
yr
Never-
Smokers
Smoked 20
Cigarettes per Day
Smoked 40
Cigarettes per Day
30 Years 40 Years 30 Years 40 Years
Men
5059 2.5 124.6 236.8 188.4 336.5
6069 11.9 224.3 486.8 572.8 606.6
Women
5059 5.9 93.2 150.2 152.2 159.5
6069 9.8 200.8 264.4 257.7 552.8
*Data from Thun et al.
86
Table 3Risk of Lung Cancer among Ex-Smokers
Relative to Never-Smokers According to Length of
Time Since Smoking Cessation and Number of
Cigarettes Formerly Smoked among a Cohort of
US Veterans*
Years Since
Smoked
Cigarettes Smoked per Day
Total
Ex-smokers 19 1020 2139 40
5 7.6 12.5 20.6 26.9 16.1
59 3.6 5.1 11.5 13.6 7.8
1019 2.2 4.3 6.8 7.8 5.1
2029 1.7 3.3 3.4 5.9 3.3
3039 0.5 2.1 2.8 4.5 2.0
40 1.1 1.6 1.8 2.3 1.5
*Relative risk compared to referent category of never-smokers (1.0).
Table adapted from Hrubec and McLaughlin.
56
Studies using biomarkers of exposure and dose to
tobacco smoke components show little relationship
of levels of these markers with tar or nicotine yield as
measured by the FTC protocol.
60
These studies have
been conducted in both the population context and
during smoking in the laboratory setting. For exam-
ple, Coultas and colleagues
61
collected saliva for the
analysis of cotinine levels, and end-tidal breath sam-
ples for the measurement of carbon monoxide levels
in a population sample of Hispanic persons from
New Mexico who were included in a respiratory
health survey. After taking account of the numbers of
cigarettes smoked, the levels of biomarkers were not
associated with the yields of tar and nicotine of the
current brand that was being smoked. Djordjevic
and colleagues
62
evaluated smoking patterns and
biomarkers in the laboratory setting, contrasting
smokers of medium-yield and low-yield cigarettes.
The smokers had greater puff volumes and frequen-
cies than are specified in the FTC protocol and had
substantially greater intakes of tar and nicotine than
those implied by the brand listings. The lack of
association of tar and nicotine yields with biomarker
levels partially reflects compensatory changes in
smoking patterns for those switching from higher
yield to lower yield products. The compensation
includes the blocking of the ventilation holes, more
frequent and deeper puffs, and an increase in the
number of cigarettes smoked.
63
The gradual reduction in tar yield over recent
decades would be expected to have reduced smok-
ers exposures to carcinogens if the FTC test proto-
col were predictive of carcinogen doses that are
delivered to the lung.
59
However, questions remain
as to whether the FTC test method is informative
with regard to lung cancer risk, or to the risks of
smoking-caused diseases more generally.
63,64
Epide-
miologic studies have been carried out to assess
whether the seemingly substantial changes in tar and
nicotine yield, as measured by the FTC protocol,
have resulted in parallel changes in the risk of
smoking. Epidemiologic studies have been the key
source of information, because they can provide
direct evidence on the risks of smoking cigarettes as
they are actually smoked during use, including any
compensatory behavior.
For lung cancer, and for other diseases, two
lines of epidemiologic data have been available on
the changes in smoking products. The first line of
data comes from case-control studies that com-
pared the smoking history profiles of persons
developing lung cancer over the several-year pe-
riod of case accrual with those of concomitantly
selected control subjects, and the second line of
data comes from cohort studies that have tracked
the risk of lung cancer over time, as the products
being smoked changed. These studies have pro-
vided temporally cross-sectional assessments, but
have not provided insights into the consequences
of smoking one product or another across the full
time period of smoking. In fact, smokers change
the products that they smoke and the product
characteristics continue to change over time as
well. Consequently, researchers cannot compare
the risks of smoking one type of cigarette across
decades with that of smoking another. Because of
these methodological issues, the epidemiologic
data need to be interpreted cautiously.
63
Some epidemiologic data, coming primarily from
case-control studies, suggest that filtered cigarettes
and lower tar yields slightly reduce the risk of lung
cancer associated with cigarette smoking compared
to unfiltered cigarettes (Table 4)
5262,6575
or to
higher tar yields (Table 5).
69,7480
The initial evidence came primarily from case-
control studies that compared risks in persons who
had used filter-tip cigarettes to persons who had
smoked nonfilter cigarettes exclusively.
65,66
This
comparison could be made among smokers in the
1960s, as there was still a substantial proportion of
smokers who had not used filter cigarettes at all.
For example, Bross and Gibson
65
compared the
Figure 4. Sales-weighted average tar and nicotine deliveries,
1954 to 1993. Adapted from US Department of Health and
Human Services.
59
lung cancer risks for those persons smoking filter
cigarettes with those smoking nonfilter cigarettes
among patients seen at Roswell Park Memorial
Cancer Institute in Buffalo. Persons were classi-
fied as filter cigarette smokers if they had used
these products for at least 10 years. These initial
studies indicated that filter cigarettes provided
some reduction of lung cancer risk. Subsequent
case-control studies that have contrasted the use
of either filter or lower yield cigarettes across the
Table 5A Summary of Studies of Lung Cancer Risk According to Tar Exposures or Levels in Cigarettes Smoked
Study/Year Location
Results*
1 2 3 4 5
Case-control studies
Joly et al
76
/1983 Havana, Cuba, hospital-based 1.0 1.75 (1.12, 2.74) 1.62 (0.99, 2.66)
Lubin et al
69
/1984 Western Europe multisite, hospital-based 1.0 1.2 1.5 1.8 1.7 Male
1.0 5.9 4.0 7.7 Female
Wilcox et al
77
/1988 New Jersey, population-based 1.0 1.62 1.96 1.89
Kaufman et al
78
/1989 United States and Canada, multisite, hospital-
based
1.0 1.9 (1.0, 3.7) 3.1 (1.3, 7.1)
Prospective studies
Hammond et al
79
/1976 United States (American Cancer Society CPS-I) 1.0 1.21 (1.02, 1.45) 1.36 (1.15, 1.61)
Higenbottam et al
80
/1982 England (Whitehall Study) 1.0 1.15 (0.77, 1.74) 1.48 (1.0, 2.18)
Garfinkel and Stellman
74
/
1988
US Women (American Cancer Society CPS-II) 1.0 1.17 1.36 1.59 1.85
Sidney et al
75
/1993 United States (Kaiser Permanente) 1.0 0.98 (0.66, 1.47) 0.87 (0.56, 1.37)
*The exposure categories of low to high are for summary purposes only and do not correspond to identical dosage categories.
Referent category (low).
Values given as odds ratio (95% confidence limits).
Values given as relative risk (95% confidence limits).
Table 4A Summary of Studies of Lung Cancer Risk According to Type of Cigarette Smoked
(Non-filtered vs Filtered Cigarettes)*
Study/Year Location Results Comments
Case-control studies
Bross and Gibson
65
/1968 Roswell Park, hospital-based 1.69 (1.12, 2.57)
Wynder et al
66
/1970 New York, hospital-based
(Memorial Sloan-Kettering)
1.62 (0.99, 2.65)
Wynder and Stellman
67
/ US multisite, hospital-based Men: 1.9 (0.92, 1.55) Nonfilter usage vs filter usage 10 yr
1979 Women: 1.29 (0.67,
2.47)
Wynder and Kabat
68
/1988 US multisite, hospital-based Men: 1.45 (0.79, 2.70) Lifetime nonfilter usage vs lifetime filter
Women: 1.56 (0.74,
3.33)
usage
Lubin et al
69
/1984 Western Europe, multisite, hospital-
based
Men: 1.6 (1.3, 1.8) Mixed usage
1.8 (1.5, 2.1) Lifetime nonfilter usage
Women: 1.8 (1.3, 2.6) Mixed usage
2.5 (1.2, 5.2) Lifetime nonfilter usage
Pathak et al
70
/1986 New Mexico, population-based Whites: 1.25 Lifetime nonfilter usage
Hispanics: 25.0
Jockel et al
71
/1992 Germany, hospital-based 2.4 (1.2, 4.8) All men; exclusively or partly nonfilter usage
last 20 yr vs filter usage last 20 yr
Prospective studies
Hawthorne and Fry
72
/
1978
West Scotland 1.20 (0.77, 1.89)
Rimington
73
/1981 England 1.67 (1.14, 2.45) Age-adjusted relative risk 1.54
Garfinkel and Stellman
74
/
1988
United States (American Cancer
Society, CPS II)
1.5 (1.3, 1.8) All women; filter for 40% of smoking
relative to lifetime filter usage
Sidney et al
75
/1993 Kaiser Permanente 1.24 (0.78, 1.89)
*The referent category of filtered cigarette smokers is equal to 1.0.
Values given as odds ratio (95% confidence limits).
Values given as relative risk (95% confidence limits).
cumulative smoking history with nonfilter or
higher yield cigarettes have had generally similar
findings.
64
In a 1976 publication, Hammond and colleagues
79
compared the mortality risks from lung cancer and
other diseases by the tar yield of products smoked by
participants in the American Cancer Society Cancer
Prevention Study (CPS) I participants. The fol-
low-up interval spanned the years 1960 to 1972.
Smokers were placed into the following three cate-
gories of cigarette products smoked: low yield,
17.6 mg per cigarette; high yield, 25.8 to 35.7 mg
per cigarette; and medium yield, intermediate. The
standardized mortality rate for lung cancer in low-
yield and medium-yield smokers was approximately
80% of the rate in high-yield smokers. A further
analysis of tar yield using the same data set con-
firmed that the risk for lung cancer death increased
with tar yield.
8183
This study only provided a comparison of risks
across the 12 years of follow-up from 1960 to 1972.
Further insights have been gained by comparing
the risks in the two CPS studies of the American
Cancer Society. This comparison addressed
whether the risks have changed, comparing smok-
ers who developed the disease between 1960 and
1972 with a similar group who developed the
disease during the initial follow-up of CPS II, from
1980 to 1986.
84,85
If the risk for lung cancer that is
associated with smoking is decreasing over time,
the expectation would be that risks for smokers
would be less in CPS II than in CPS I. In fact, the
opposite was observed, with increasing lung can-
cer mortality in male and female smokers in CPS
II compared with CPS I (Fig 5).
86
While differ-
ences in smoking patterns may partially explain the
increased lung cancer mortality, comparing CPS I
and CPS II, male smokers in CPS II had substan-
tially higher lung cancer mortality rates than their
counterparts in CPS I.
86
In an analysis with similar findings, Doll and
colleagues
87
compared the risks of death from lung
cancer and other causes during the first and second
20-year periods of the 40-year follow-up of the
British physician cohort. Lung cancer mortality in-
creased among smokers in the second 20-year period
(ie, 1971 to 1991), even though the products smoked
during that time period would have had a substan-
tially lower tar and nicotine yield than those smoked
during the first 20-year period (ie, 1951 to 1971). For
the first 20 years, the annual lung cancer mortality
rate among current smokers was 264 per 100,000,
and for the second 20 years, it was 314 per 100,000.
Several expert panels have recently reviewed the
findings. Stratton et al
88
for the Institute of Medicine
carried out a comprehensive review on various harm-
reduction strategies for reducing the disease burden
caused by smoking, including the smoking of lower
yield cigarettes. There are also new products in
various phases of development that are intended to
deliver nicotine without the direct combustion of
tobacco. The Institute of Medicine report concluded
that smoking lower yield products had not been
shown to benefit the health of smokers.
This topic was also addressed in the National
Cancer Institute monograph, Risks Associated with
Smoking Cigarettes with Low Machine-Measured
Yields of Tar and Nicotine.
63
This monograph pro-
vides a comprehensive review of the topic as well as
new analyses of the CPS I data. The report proposes
that the net consequences of having lower yield
products needs consideration, along with any change
in the risk of disease to smokers. If some persons
start smoking or continue to smoke because they
view the smoking of lower yield cigarettes as having
a more acceptable level of risk, then the success of
tobacco control interventions may be lessened. The
report also found that compensatory changes in
smoking patterns reduce any theoretical benefit of
lower yield products. Overall, the report concluded
that changes in cigarette design and manufacturing
over the last 50 years had not benefited public
health.
Passive Smoking: Passive smokers inhale a com-
plex mixture of smoke that is now widely referred to
as environmental tobacco smoke (ETS). Passive
smoking was first considered as a possible risk factor
for lung cancer in 1981 when two studies were
published that described increased lung cancer risk
among never-smoking women who were married to
smokers. Hirayama
89
reported the findings from a
cohort study in Japan, which showed that among
nonsmoking women, those whose husbands smoked
cigarettes were at higher risk for lung cancer than
those whose husbands were nonsmokers. A case-
control study in Athens reported by Trichopoulos
and colleagues
90
shortly thereafter replicated this
finding. Additional evidence rapidly accrued so that
by 1986 two important summary reports were pub-
lished. The National Research Council reviewed the
epidemiologic evidence and concluded that non-
smoking spouses who were married to cigarette
smokers were about 30% more likely to develop lung
cancer than nonsmoking spouses who were married
to nonsmokers, and that this relationship was biolog-
ically plausible.
91
Almost one fourth of lung cancer
cases among never-smokers were estimated to be
attributed to exposure to passive smoking.
91
The
1986 report of the Surgeon General also judged
passive smoking to be a cause of lung cancer,
14
an
inference corroborated by the 1992 review
15
of the
evidence and risk assessment by the US Environ-
mental Protection Agency, which classified ETS as a
known human (class A) carcinogen. Estimates indi-
cate that passive smoking accounts for approximately
3,000 lung cancer deaths per year in the United
States.
15
Since these conclusions were reached, several
major studies have been carried out to further
characterize the association of passive smoking with
lung cancer, while taking into account some of the
limitations of earlier studies, particularly small sam-
ple sizes, exposure misclassification, and omission of
some potentially confounding factors.
92,93
Passive smoking is more weakly associated with
lung cancer than is active smoking, as would be
expected given the generally lower doses of carcin-
ogens that are received passively by the lung of the
nonsmoker compared to the doses received by the
active smoker. Because of broad societal implica-
tions, the conclusion that this association is causal
Figure 5. Age-specific death rates from lung cancer among current cigarette smokers and lifelong
never-smokers based on smoking status at enrollment in the CPS I or CPS II, according to attained age.
Adapted from Thun et al.
86
has been controversial.
94,95
Questions have been
raised about the methodology of the epidemiologic
studies, including confounding and the misclassifica-
tion of exposures to environmental tobacco smoking.
Review groups have concluded
14,15,91
that the asso-
ciation between ETS and lung cancer cannot be
attributed to the methodological limitations of epi-
demiologic data.
Studies have been directed at the specific venues
where nonsmokers are exposed to tobacco smoke,
including the home, workplaces, and public places.
Much of the literature has focused on the increased
risk associated with being married to a smoker, an
exposure variable that can be readily ascertained.
Meta-analyses have been conducted periodically to
summarize the evidence from the epidemiologic
studies. A 1997 meta-analysis by Law and col-
leagues
96
found an approximately 20% increased risk
associated with marriage to a smoker. This excess
risk appeared to be due to exposure to passive
smoking since it could not be explained by confound-
ing or misclassification.
Reynolds
97
summarized the evidence on work-
place exposure to secondhand smoke and lung can-
cer risk. In general, the estimates have indicated the
existence of an increased risk associated with expo-
sure, although they are imprecise due to a limited
numbers of cases. There is also coherence between
model-based estimates derived by combining risk
estimates from studies of exposure to smoking by
husbands and monitoring data from workplaces.
98
The studies of passive smoking provide further evi-
dence documenting the dose-response relationship be-
tween cigarette smoke and lung cancer. The doses
extend to far lower levels than those of active smoking,
and increased risk is observed, suggesting that there is
no threshold for tobacco carcinogenesis.
14
Lung Cancer Histopathology: Lung cancer occurs
in multiple histologic types as classified by conven-
tional light microscopy. The four major types include
squamous cell carcinoma, adenocarcinoma, large cell
carcinoma, and small cell undifferentiated carci-
noma. Together, these four types of lung cancer
account for 90% of lung cancer cases in the
United States.
99
Despite extensive research, the
mechanisms leading to these different types of lung
cancer remain uncertain. Hypotheses have focused
on the cells of origin of lung cancers and on the
pathways of differentiation of malignant cells.
17,99
Smoking has been shown to cause each of the major
histologic types of lung cancer, although the dose-
response relationship with the number of cigarettes
smoked varies across the types and is steepest for
small cell undifferentiated carcinoma.
57,100,101
There
are a few suggestive links of histologic type with
occupational agents. Small cell lung cancer has been
reported to be in excess in workers who have been
exposed to chloromethyl ethers and in underground
miners who have been exposed to radon progeny.
17,99
In the initial decades of the smoking-caused epi-
demic of lung cancer, squamous cell carcinoma was
the most frequent type of lung cancer that was
observed in the population among smokers, and
small cell carcinoma was the next most frequent. In
the late 1970s, the first evidence of a shift toward a
predominance of adenocarcinoma was noted,
99,102,103
and now adenocarcinoma of the lung is the most
common histologic type of lung cancer.
4,28
The decline
in lung cancer rates has been more rapid for squamous
cell and small cell carcinomas than for adenocarci-
noma, which is just beginning to show a lower inci-
dence rate (Fig 6).
4
In women, the Surveillance, Epi-
demiology, and End Results programdata from1973 to
1996 indicate that the incidence rates of squamous cell,
small cell, and large cell carcinomas have at least
reached a plateau while the rate for adenocarcinoma is
still rising.
While the changing patterns of diagnosis and
classification of lung cancers could have led to these
changes over time, most observers have set aside an
artifactual change.
99,103
Beginning in the 1970s, new
techniques for the diagnosis of lung cancer became
available, including the fiberoptic bronchoscope and
thin-needle aspiration.
104
Improved stains for detect-
ing mucin, the hallmark of adenocarcinoma, also
were introduced. Using data from the Connecticut
Tumor Registry, Thun et al
104
showed that the rise in
adenocarcinoma antedated these diagnostic innova-
tions.
Hypotheses concerning the shift in histopathology
have focused on the potential role of changes in the
characteristics of cigarettes and the consequent
changes in the doses of carcinogens inhaled.
60,101,105
Puff volume likely has increased over recent decades
with the possibility that patterns of deposition in the
lung have changed, tending toward enhanced depo-
sition of tobacco smoke in the peripheral airways and
alveoli.
105
The levels of nitrate, which enhances the
combustion of tobacco, also have increased in to-
bacco smoke. While more complete combustion
decreases the concentrations of polycyclic aromatic
hydrocarbons, the increased production of nitrogen
oxides contributes to the increased formation of
tobacco-specific nitrosamines. An increase in the
dose of the potent tobacco-specific nitrosamine
NNK has been postulated as one factor leading to
the increase in adenocarcinoma.
105,106
NNK induces
lung carcinomas, predominantly adenomas and ade-
nocarcinomas, in mice, regardless of the route of
administration.
106
Few studies can provide data to test these hypoth-
eses because of the need for longitudinal observation
of lung cancer risk in relation to the characteristics of
the cigarettes smoked over time. Thun et al
104
compared the risks for lung cancers of different
histologic types among participants in CPS I and
CPS II of the American Cancer Society. They found
markedly rising risks for adenocarcinoma of the lung
to be associated with smoking in both men and
women over the approximately 20 years separating
the two studies. Thun et al
104
concluded that the
increase in lung adenocarcinoma since the 1950s is
more consistent with changes in smoking behavior
and cigarette design than with diagnostic advances.
The hypothesis has been advanced that women
may have a greater risk of lung cancer than men at
the same level of smoking. Hypotheses have been
based on possible hormonally related differences in
response to carcinogens. The evidence is limited and
mixed, and the 2001 Report of the Surgeon Gener-
al
107
did not reach a conclusion on this issue.
Diet
Overview: The dietary factors that have received
greatest emphasis in studies of diet and lung cancer,
namely fruits, vegetables, and specific antioxidant
micronutrients, are the focus of this summary. These
are the research areas that also presently appear to
have the greatest implications for lung cancer pre-
vention. Much of the research on diet and lung
cancer has been motivated by the hypothesis that
diets high in antioxidant nutrients may protect
against oxidative DNA damage and thereby protect
against cancer.
108
Within this limited range of topics, we have at-
tempted to be comprehensive in our review. We
omitted studies with lung cancer mortality as the end
point, as different factors may be associated with
survival of lung cancer than disease risk. We limited
our consideration of specific micronutrients to reti-
nol, total carotenoids, -carotene, and vitamin C.
For these postulated relationships, there is no reason
to suspect a threshold effect in the exposure-
response relationship. Rather, a dietary factor that
protects against lung cancer would theoretically be
expected to confer greater protection when present
in greater amounts. Consequently, dose-response
trends are emphasized in this review, with the
presence of a monotonic dose-response relationship
considered to provide strong evidence favoring an
association, and measures of association in the pro-
tective direction but with no dose-response gradient
providing weak evidence in favor of an association.
The overwhelming contribution of cigarette smok-
ing as a cause of lung cancer imposes challenges to
detecting the role that other lifestyle factors, such as
diet, may play in the etiology of lung cancer. Ciga-
rette smoking is so closely associated with less
healthful lifestyles, such as less healthful diets,
109113
that it is often difficult to discern whether the dietary
factors of interest have truly been disentangled from
the effects of smoking. Compounding this problem,
even associations between dietary factors and lung
cancer that truly exist are likely to be very weak in
relation to smoking. Even for a dietary factor such as
vegetable consumption (reviewed below), which is
fairly consistently associated with a lower risk of lung
Figure 6. Surveillance, Epidemiology, and End Results incidence rates of cancer of the lung and bronchus
by histologic type, sex, race, and ethnicity for all ages, 1973 to 1996. Adapted from Wingo et al.
4
cancer, the highest exposure category is usually
associated with, at most, a halving in the risk of lung
cancer. In many instances, the potential residual
confounding effects of smoking could thus plausibly
be strong enough to explain the observed associa-
tions between dietary factors and lung cancer.
114
In
the future, studies that control for cigarette smoking
in the design are best suited to address the persistent
concern about residual confounding by cigarette
smoking. Examples are case-control studies, in which
cases and control subjects are closely matched on
cigarette smoking history, and studies limited to
never-smokers.
Fruit and Vegetable Consumption: In case-control
studies, fruit consumption has been clearly protective
in 4 studies,
115118
suggestive of a protective association
in 4 other studies,
119122
and not protective in 10
studies.
123132
In prospective studies, fruit consumption
has been clearly protective in three studies,
133135
sug-
gestive of a protective association in two other stud-
ies,
136,137
and not protective in two studies.
138,139
The
combined evidence for fruit consumption is far from
consistent but hints at the possibility of a protective
association.
The evidence favoring a protective association is
more consistent for vegetable consumption. In case-
control studies, 12 studies showed a decreased
lung cancer risk as vegetable consumption in-
creased,
115117,123125,127,129,132,140142
while 5 other
studies showed a weak association in this direc-
tion.
93,118,126,143,144
A protective association was ab-
sent in only six studies.
119121,128,131
In prospective
studies, vegetable consumption was clearly protec-
tive in two studies,
134,137
was suggestive of a protec-
tive association in two other studies,
135,138
and was
not protective in two studies.
133,136
Micronutrients: Two different strategies used to
evaluate the relationship of micronutrients to lung
cancer are (1) using data summarized from food-
frequency questionnaires to estimate micronutrient
intake and (2) drawing blood samples from study
participants and assaying the concentrations of mi-
cronutrients in circulation. The former approach
provides a better average measure of micronutrient
exposure, whereas the latter approach has the
advantage of measuring micronutrient concentra-
tions closer to the cellular level where the postulated
biological effect occurs. As mentioned above, we
focused on vitamins A and C, total carotenoids, and
-carotene.
Evidence consistent with higher dietary retinol
intake being associated with a reduced risk of lung
cancer in at least one subgroup has been observed in
only 11 studies,
122,145154
with 17 studies not providing
supportive evidence.
118,121,124126,129,132,133,140,143,155161
Circulating retinol concentrations also have not been
consistently associated with protective associa-
tions.
162172
In contrast to the results for retinol, the results for
total carotenoids, -carotene, and vitamin C are more
supportive of a reduction in lung cancer risk. Higher
dietary intake of total carotenoids were consistent with
a protective association in 18 observational stud-
ies
119,121,122,125,126,132,135,139,143,146,150,153,155,156,161,173175
and were not supportive in 6 studies.
124,129,133,137,157,166
Circulating concentrations of total carotenoids have
been linked to a reduced risk of lung cancer.
163,172
Similar protective associations have also consistently
been observed for a specific carotenoid, -carotene.
Data from 16 case-control and cohort stud-
ies
93,118,122,125,126,129,135,137,139,140,148,154,160,176178
of
dietary -carotene intake were compatible with as-
sociations in the protective direction. In only five
studies
121,124,134,138,152
have null findings been ob-
served. Further bolstering the evidence in favor of a
protective association between -carotene and lung
cancer are prospective studies in which -carotene is
assayed from blood that has been collected from a
population that was initially cancer-free and was
subsequently followed-up for the occurrence of lung
cancer.
163,166,168,169,172,179,180
The preponderance of
evidence from observational studies thus has dem-
onstrated a protective association between carotene
(specifically, -carotene) and lung cancer.
The evidence for the protective association of
vitamin C is less abundant than for total carotenoids
and -carotene but also is suggestive of a protective
association. For vitamin C intake, six case-control or
cohort studies
133,150,155157,178
were at least compati-
ble with an inverse association with lung cancer risk,
whereas three studies were not.
93,124,176
A serologic
study of ascorbic acid concentrations and lung cancer
was also compatible with a protective association.
179
Chemoprevention Trials: The experimental ratio-
nale for trials of -carotene and retinoids is offered in
another article in this supplement. These data indicated
a potential for prevention with these agents, as sup-
ported by observational data as well. However, a pro-
tective association between -carotene and lung cancer
was not found in three randomized, double-blind,
placebo-controlled chemoprevention trials
181183
of
-carotene, which indicated that -carotene supple-
mentation does not protect against lung cancer. In fact,
-carotene supplementation was associated with an
increased risk of lung cancer among the high-risk
populations of heavy smokers in the -Tocopherol
-Carotene Cancer Prevention Study
181
and of smok-
ers and asbestos-exposed workers in the Carotene and
Retinol Efficacy Trial.
183
In summary, people who eat more vegetables are
at lower risk of lung cancer than persons who
consume fewer vegetables. The same may also apply
to fruit consumption, but the evidence is less clear-
cut. The specific constituents of vegetables that
confer protection are not known. The results of the
chemoprevention trials clearly suggest a more com-
plex role than researchers had previously thought.
These trials also emphasize that we do not know
whether the associations present in observational
studies are specific to the micronutrients or whether
the micronutrient measurements are merely serving
as a marker of the intake of other protective sub-
stances or even of more healthful dietary habits in
general.
If the association is attributable to vegetable con-
sumption, we need to determine whether this effect
applies globally to vegetable consumption or to
specific classes of vegetables or specific vegetables.
For example, carrots
115,137,140,144,146
and toma-
toes
118,119,137,140,146
are associated with a reduced risk
of lung cancer, at least for the highest vs the lowest
categories of consumption. Resolving this question
also will help to clarify whether specific biochemical
constituents that are present in vegetables confer
protection against lung cancer or whether the com-
plex mixture that comprises vegetables leads to the
net protective effect. Reports from large-scale pro-
spective cohort studies that considered overall fruit
and vegetable consumption as well as classes of
vegetables (eg, cruciferous, brassica) and also indi-
vidual micronutrients
138,138
exemplify the compre-
hensive approach that affords the greatest opportu-
nity to pinpoint the specific foods and/or nutrients
that may explain the apparent protective association
with vegetable consumption.
Environmental Exposures
Occupational Exposures: Investigations of occupa-
tional groups, who often are heavily exposed over a
long time to workplace agents, have provided a
substantial understanding of the carcinogenicity of a
number of chemicals and physical agents. Among
cancers that are associated with occupational expo-
sures, cancer of the lung is the most common.
184
Estimates derived from case-control studies of the
proportion of lung cancer that is contributed to by
occupational exposures, via independent or shared
causal pathways, have ranged widely, but most point
estimates or ranges have included values from 9 to
15%.
185191
While disagreement persists concerning
specific estimates,
192
the following message is clear:
in industrialized nations the contribution of occupa-
tional exposures to the lung cancer burden is small
compared to that of cigarette smoking, but it is large
compared to the contributions of most other expo-
sure classes. Cigarette smoking potentiates the effect
of many of the known occupational lung carcino-
gens.
42
Lung cancer has been observed to be associated
with many workplace exposures. Workers exposed to
tar and soot (which contains benzo[a]pyrene), such
as coke oven workers,
193,194
in concentrations ex-
ceeding those present in urban air
195
are at increased
risk of lung cancer. Occupational exposures to a
number of metals, including arsenic,
196198
chromi-
um,
199201
and nickel,
202204
are also causes of lung
cancer.
205,206
For many persons in the worker groups
that have been exposed to these agents, there were
substantial increments in risk. However, in devel-
oped countries these hazards largely have been
controlled.
For some other workplace agents, the evidence
has been less clear. The results of numerous case-
control and cohort studies
207
are compatible with a
weak association between exposure to diesel exhaust
and the development of lung cancer. Although inad-
equate control of cigarette smoking limits the infer-
ences that can be drawn from many of these studies,
the exposure to diesel exhaust remains a likely
explanation for these findings.
207
This association
remains a public health concern, as the public is
exposed to diesel exhaust in urban areas, and in some
European countries diesel vehicles are being used
increasingly.
43
The question of whether silica dust is a risk factor
for lung cancer has been controversial.
208210
A
twofold increase in lung cancer risk was estimated
from a meta-analysis
211
of the relationship between
silicosis and lung cancer mortality. Effects of smok-
ing have not been well-controlled in most of the
studies.
211
The evidence on silica exposure, absent
consideration of the presence of silicosis, is less
clear.
212,213
In 1997, IARC did classify crystalline
silica as a human carcinogen
214
; however some still
continue to question its carcinogenicity
213
and the
role of silica exposure vs that of fibrosis in persons
with silicosis.
212
Asbestos: Asbestos, a well-established occupa-
tional carcinogen, refers to several forms of fibrous,
naturally occurring silicate minerals.
215
The epide-
miologic evidence dates to the 1950s, although clin-
ical case series had previously led to the hypothesis
that asbestos causes lung cancer.
216,217
In a retro-
spective cohort study published in 1955, Doll
218
observed that asbestos textile workers at a factory in
the United Kingdom had a 10-fold elevation in lung
cancer risk and that the risk was most heavily
concentrated during the time frame before regula-
tions had been implemented to limit asbestos dust in
factories. A sevenfold excess of lung cancer was
subsequently observed among insulation workers in
the United States.
219,220
The peak incidence oc-
curred 30 to 35 years after the initial exposure to
asbestos (Fig 7).
221
The risk of lung cancer has been
noted to increase with increased exposure to asbes-
tos
222
and to be associated with the principal com-
mercial forms of asbestos.
223
Whether asbestos acts
directly as a carcinogen or through indirect mecha-
nisms such as causing chronic inflammation that
eventually leads to cancer development remains
uncertain.
224,225
Asbestos and cigarette smoking are both inde-
pendent causes of lung cancer, but in combination
they act synergistically to increase the risk of lung
cancer in a manner that is compatible with a
multiplicative effect (Table 6).
226
Cigarette smok-
ing may increase the lung cancer risk associated
with asbestos exposure by enhancing the retention
of asbestos fibers.
227
Radiation
Epidemiologic studies of populations that have
been exposed to high doses of radiation show that
lung cancer is one of the cancers associated with
exposure to ionizing radiation. However, the risks of
low-dose radiation, which are more relevant to con-
temporary workers and the general population, have
proven difficult to characterize.
228
Assessing the
cancer risk that is associated with exposure to low-
dose radiation among humans is methodologically
difficult because the signal-to-noise ratio is highly
unfavorable.
229
The following two types of radiation, which are
classified by the rate of energy transfer to the tissue,
are relevant to lung cancer: low linear energy trans-
fer (LET) radiation (eg, x-rays and gamma rays); and
high-LET radiation (eg, neutrons and radon). High-
LET radiation produces ionization of a relatively
higher density in tissues than does low-LET radia-
tion, so in equivalent doses more biological damage
Figure 7. Frequency distribution of the number of lung cancer cases by time since asbestos exposure
began among a cohort of asbestos insulation workers. Adapted from Selikoff.
221
Table 6Summary of the Joint Relationship Between
Cigarette Smoking and Asbestos Exposure on the Risk
of Lung Cancer Mortality*
Smoking
Mortality
Rates Additive Model
Multiplicative
Model
No Yes No Yes No Yes
No 11.3 58.4 0 47.1 1.0 5.17
Yes 122.6 601.6 111.3 590.3 10.85 53.24
*Mortality rates are per 100,000 population. Table adapted from
Hammond et al.
226
Rate difference compared to referent category of no smoking and no
asbestos exposure. Expected value under model of no statistical
interaction is 158.4.
Rate ratio compared to referent category of no smoking and no
asbestos exposure. Expected value under model of no statistical
interaction is 56.1.
is produced by high-LET than by low-LET radia-
tion.
230
For both types of radiation, the majority of
the epidemiologic evidence comes from cohorts who
have been exposed at levels substantially greater than
those experienced by the general population. Risk
assessment methods then are used to estimate the
risks to the population.
High-LET Radiation (Radon): Radon is an inert
gas that is produced naturally from radium in the
decay series of uranium. Two of the decay products
of radon emit particles that can, by virtue of their
high energy and mass, cause damage to the DNA of
cells of the respiratory epithelium. Epidemiologic
studies of underground miners of uranium and other
ores have established exposure to radon daughters as
a cause of lung cancer.
17,231,232
In the miners who
were exposed to radon in past centuries, very high
lung cancer risks were observed. These risks fell for
more recent workers, but some epidemiologic stud-
ies
17
still show clear evidence of existing cancer risk.
Cigarette smoking and radon decay products syner-
gistically influence lung cancer risk in a manner that
is supra-additive but submultiplicative.
17,232
Radon is of broader societal interest because it is a
ubiquitous indoor air pollutant, entering buildings in
soil gas. On average, indoor exposures to radon for
the general population are much less than those
received by occupational groups such as uranium
miners. For example, even the lowest historical
radon concentration in a uranium mine is roughly 50
to 100 times higher than in the average home.
232
Exposure to radon in indoor air also is assumed to
cause lung cancer, but the magnitude of the risk is
uncertain because of the assumptions underlying the
extrapolation of findings from uranium miners to the
generally lower exposures indoors. These assump-
tions relate to dose, dose-rate, and dosimetry, and
also reflect the lack of information on risks of
exposures of women and children. Strengthening
biological evidence supports the assumption that a
single hit to a cell by an particle causes permanent
cellular change, an assumption leading to a non-
threshold dose-response relationship.
The assumptions made by the Environmental
Protection Agency and the Biological Effects of
Ionizing Radiation IV and VI Committees of the
National Research Council have led to estimates that
approximately 15,000 to 20,000 lung cancer deaths
per year in the United States are caused by radon.
233
Case-control studies concerning indoor exposure to
radon as a risk factor for lung cancer, which were
undertaken to directly assess risks, have produced
findings that are generally consistent with downward
extrapolation of risk models based on the under-
ground miners.
234
When combined with meta-
analysis, there is a significant association between
indoor radon and lung cancer in the general popu-
lation that is quantitatively comparable with risk
models derived from the underground miners. This
coherence lends support to using extrapolation of the
miner data to estimate the risk of indoor radon.
Low-LET Radiation (X-Rays and Gamma Rays):
Epidemiologic data relating low-LET radiation to
lung cancer stem from the following three principal
populations: the atomic bomb survivors in Japan
235
;
patients with diseases such as ankylosing spondyli-
tis
236
or tuberculosis,
237,238
who received multiple
radiation treatments; and occupational groups in
professions who are exposed to radiation.
239
The
single, high-dose exposure of the atomic bomb sur-
vivors was associated with a significant lung cancer
risk.
235
Regardless of their age when the atomic
bombs were dropped, the excess of lung cancer did
not occur until the survivors reached older ages
when cancer usually occurs.
235
The risks associated with exposure to lower doses
of low-LET radiation have been estimated in two
ways. Statistical models have been used to extrapo-
late from the data of the atomic bomb survivors to
lower doses. Tuberculosis patients who received
radiation therapy also have been studied. They were
intermittently exposed to radiation. Such intermit-
tent, low-dose exposures may be most pertinent for
the general population because this exposure pattern
is the most common in technologically advanced
societies. Studies of tuberculosis patients have sug-
gested that if there is any risk of lung cancer
associated with this exposure pattern, it is
small,
237,238
implying that the assumptions on which
the higher risk estimates obtained from the atomic
bomb survivors data may in actual fact not hold.
238
Low-LET radiation therefore appears to be asso-
ciated with higher lung cancer risk when exposure
occurs at a higher dose rate.
238
These results contrast
with those for high-LET radiation, suggesting that
the two types of radiation have different dose-rate
relationships.
238
Air Pollution
During a typical day, the average adult inhales
about 10,000 L air.
240
Consequently, even the car-
cinogens that are present in the air at low concen-
trations are of concern as a risk factor for lung
cancer. Extrapolation of the risks associated with
occupational exposures to the lower concentration of
carcinogens in polluted ambient air leads to the
conclusion that a small proportion of lung cancer
cases could be due to air pollution.
184,205,241
Atmospheric Air Pollution: Carcinogens generated
by combustion of fossil fuels include polycyclic aro-
matic hydrocarbons and metals such as arsenic,
nickel, and chromium.
205
In considering respiratory
carcinogenesis, the constituents of air pollution will
vary by locale and over time depending on the
pollution sources.
242
Consequently, epidemiologic
investigations of air pollution and lung cancer have
been limited by the difficulty of estimating exposure.
Nevertheless, descriptive evidence is consistent with
a role for air pollution in causing lung cancer.
Urbanization and lung cancer mortality are
linked.
38,243245
This association could arise from
differences in the distributions of other lung cancer
risk factors, such as smoking and occupational expo-
sures, by degree of urbanization. Adjustment for
these factors may considerably attenuate the effect
of urban location,
246,247
but an urban effect persists
in a number of studies.
43
Air pollution has been assessed as a risk factor for
lung cancer in both case-control and cohort studies.
These studies have been reviewed in detail else-
where.
43,248
These reviewers have found the evi-
dence wanting because of the potential for exposure
misclassification and for uncontrolled confounding.
Two prospective cohort studies
249,250
that partially
address these weaknesses of earlier studies have
added evidence suggesting that air pollution is
weakly associated with the risk of lung cancer. By
prospectively studying air pollution levels in relation
to the risk of lung cancer, and by controlling for
possible confounders such as age, smoking, and
socioeconomic status at the individual level, these
studies surmount some shortcomings noted in much
previous research.
251
In a study of six US cities,
249
the adjusted risk of lung cancer mortality in the city
with the highest fine particulate concentration was
1.4 times (95% confidence interval, 0.8 to 2.4) higher
than that in the least polluted city. Using data from
CPS II of the American Cancer Society, Pope and
colleagues
250
observed that, compared to the least
polluted areas, residence in areas with high sulfate
concentrations was associated with an increased risk
of lung cancer (adjusted relative risk, 1.4; 95%
confidence interval, 1.1 to 1.7) after adjustment for
occupational exposures and the factors mentioned
above. However, unlike the Six-Cities study, fine
particulate concentration was not associated with
lung cancer risk.
250
By contrast, in the American
Cancer Society CPS I cohort air pollution was not
associated with lung cancer risk. In that study, men
were stratified according to exposures in the work-
place but relied only on proxy, less specific measures
of air pollution.
251
Some case-control studies
252254
have reported indexes of air pollution to be modestly
associated with elevated risks of lung cancer, but
others
121,255
have reported no association. Another
research approach to evaluate the risk of air pollution
has been to investigate populations residing around
point sources of pollution, such as factories and
smelters. Proximity of residence to the pollution
source can be used as a proxy for exposure. Many
industries have been studied using this approach.
Areas surrounding nonferrous smelters, which emit
arsenic, have been of particular interest. An ecologic
study reported by Blot and Fraumeni
37
in 1975
suggested that excess lung cancer occurred in US
counties with copper, lead, or zinc smelting and
refining industries. The results of several subsequent
case-control studies
256258
have lent support to this
hypothesis by showing that the risk of lung cancer
increased the nearer persons lived to nonferrous
smelters, after accounting for personal cigarette
smoking and employment at the smelter. Other
case-control studies did not replicate this find-
ing
259,260
but were also limited by their failure to
account for smoking and employment at the smelter.
Doll and Peto,
184
in their 1981 review of the
causes of cancer, estimated that perhaps 1 to 2% of
lung cancer cases were related to air pollution. Even
in light of recent findings, this appears to remain a
reasonable estimate. To the extent that air pollution
may contribute to the occurrence of lung cancer, and
the overall epidemiologic evidence is equivocal, its
contribution is minimal relative to cigarette smoking.
This is to be expected, given that respiratory doses of
carcinogens from active smoking are significantly
greater than those received from the inhalation of
atmospheric contaminants.
Indoor Air Pollution: An individuals total expo-
sure to air pollution depends on indoor as well as
outdoor exposures. Indoor air quality has large po-
tential health implications because people may
spend substantial amounts of time indoors. Indoor
air pollution may stem from incoming outdoor air or
may originate indoors from tobacco smoking, build-
ing materials, soil gases, household products, and
combustion from heating and cooking.
261
A trade-off
exists between energy efficiency and indoor air
quality, as ventilation allows heated/cooled air to
escape but improves indoor air quality.
262
As discussed earlier, in more developed countries,
two of the most important indoor pollutants that
influence lung cancer risk in never-smokers are
passive smoking
14
and radon.
231
Asbestos exposure
may pose a risk to building occupants, but it is
estimated to be minimal.
215
Of major concern in the
developing world is the indoor air contamination
resulting from the use of unprocessed solid fuels,
notably coal, for cooking and space heating.
263
Mum-
ford and colleagues
264
inferred that smoky coal was
likely to be a major determinant of the geographic
distribution of lung cancer in Xuan Wei, China, a
finding that was corroborated by an animal model.
265
Case-control studies conducted elsewhere in mainland
China
131,258
and Taiwan
149
further implicated coal use
as a risk factor for lung cancer. Another case-control
study in Shanghai,
266
where most homes are unheated,
reported no association between the use of coal and
lung cancer risk. Exposure to coal burning in the
preadult years was associated with lung cancer risk in a
case-control study of women in Los Angeles County.
154
Host Factors
Overview
Genetic susceptibility to lung cancer has long been
postulated. Environmental agents, even cigarette
smoking, cause lung cancer in only a minority of
exposed persons, leading to the hypothesis that suscep-
tibility is inherently determined. Epidemiologic studies
showing that a family history of lung cancer predicts
increased risk further support a genetic basis for lung
cancer susceptibility. This long-postulated hypothesis is
now being actively addressed using the approach of
molecular epidemiology. Full coverage of this topic is
beyond the scope of this article. Aspects of genetic
susceptibility have been reviewed previously.
267,268
Familial aggregation of lung cancer has been
demonstrated primarily in case-control studies (Ta-
ble 7).
266,269278
In these studies, a family history of
lung cancer tended to be associated with increased
risk of lung cancer. Most of the studies controlled for
smoking, which is known to aggregate in families. In
a large study in Louisiana,
279
segregation analysis
suggested that lung cancer inheritance was consis-
tent with a mendelian codominant autosomal gene
determining the early onset of disease. On the other
hand, the largest study
280
of lung cancer in twins
reported to date did not provide evidence indicating
a genetic basis for susceptibility. A follow-up of
15,924 male twin pairs in the United States did not
show greater concordance in monozygotic compared
with dizygotic twins, and death rates from lung
cancer were similar by zygosity group in surviving
twins whose siblings had died of lung cancer.
Nevertheless, the evidence for aggregation and
seeming susceptibility of some smokers has been
viewed as a sufficient rationale for further investiga-
tion of the potential basis for lung cancer, using the
new techniques of cellular and molecular biology.
Research Findings on the Genetic Basis of
Lung Cancer
With the application of the new and powerful tools
of modern molecular and cell biology, research
findings are now characterizing the changes in cells
that are caused by exposure to tobacco smoke and
are providing a framework for understanding the
genetic basis of lung cancer risk. Figure 8, adapted
Table 7A Summary of Case-Control Studies of Lung Cancer Among Persons With a Family History of Lung
Cancer Compared to Those With No Family History of Lung Cancer
Study/Year Location Odds Ratio* Comment
Tokuhata and
Lilienfeld
269
/1963
Buffalo, NY 2.4 (1.1, 5.2) Adjusted for cigarette smoking, age, sex,
generation
Tokuhata
270
/1964 Baltimore, MD 2.6 (1.4, 5.4) Adjusted for age, sex, generation
Ooi et al
271
/1986 10 Louisiana
Parishes
2.7 (1.7, 4.2) Adjusted for smoking, sex
Samet et al
272
/1986 New Mexico Parental: 5.3 (2.2, 12.8) Adjusted for cigarette smoking, age, sex,
ethnicity
Gao et al
266
/1987 Shanghai, P.R.C. Parental: 1.1 (0.6, 2.3) Adjusted for smoking, age
Sibling: 3.0 (0.7, 12.5)
Tsugane et al
273
/1987 Tokyo, Japan 1.0 (0.3, 3.9) Smoking and occupational exposure data
collected, but not adjusted for
Horwitz et al
274
/1988 New Haven, CT 2.3 (0.6, 9.7) Adjusted for smoking
McDuffie
275
/1991 Saskatchewan,
Canada
2.0 (1.2, 3.4) Smoking and occupational exposure data
Osann
276
/1991 Northern
California
1.9 (0.7, 5.6) Adjusted for cigarette smoking education,
matched on age and race
Shaw et al
277
/1991 Gulf Coast of
Texas
1 1st degree relative with cancer: 1.7 (1.2, 2.4) Adjusted for cigarette smoking, passive
smoking, and number of first degree
relatives
2 1st degree relatives with cancer: 2.8 (1.2, 6.6)
Filho et al
278
/1995 Sao Paulo, Brazil 3.2 (1.0, 11.1) Smoking and occupational exposure data
*Referent category of persons with no family history of lung cancer 1.0. Values in parentheses are confidence limits.
from Hecht,
106
offers a general schema for the
process of carcinogenesis by tobacco smoking.
Viewed in the framework set by this type of model,
research findings mirror the predictions of the mul-
tistage model in many respects and are enhancing
the understanding of the mechanisms by which
smoking causes cancers of the lung and other organs.
A rapidly expanding literature, based both in the
laboratory and in the observational approach often
referred to as molecular epidemiology, addresses the
dosimetry and metabolism of tobacco carcinogens at
the cellular and molecular levels, the genetic deter-
minants of susceptibility, and the patterns of genetic
changes in the tissues of smokers and in the cancers
that the tissues develop.
106,267
In a general formulation of the determinants of
cancer risk, the risk depends on carcinogen exposure
and the factors that determine host susceptibility,
including genetic predisposition.
281
For tobacco
smoking and lung and other cancers, the elements of
this paradigm are all topics of inquiry, using the
combination of laboratory-based and population-
based studies that are indicated in the diagram.
Biomarkers are central to the molecular epidemiol-
ogy approach. The term refers to making measure-
ments of indicators of exposure and dose, suscepti-
bility, and response in biological materials, including
tissue samples, blood, urine, and saliva.
282285
As
research evolves within this paradigm, a more com-
plete biological understanding of the specific events
underlying the multistage model, which originally
was proposed on a conceptual basis, can be antici-
pated.
This framework indicates multiple points at which
genetically determined host characteristics might be
important (eg, carcinogen metabolism and activation,
and DNA repair capacity).
There is a rapidly expanding literature on the
molecular and cellular basis of lung cancer. It is
beyond the scope of this review to cover this topic.
Some reviews have been published.
267,286
The evi-
dence has expanded and deepened our understand-
ing of how smoking injures cells and causes cancer.
Many carcinogenic compounds in tobacco smoke
(eg, polycyclic aromatic hydrocarbons) undergo met-
abolic activation by phase I enzymes of the cyto-
chrome p450 system to form reactive intermediates
that bind to DNA and cause genetic injury. Two of
these enzymes have been investigated with regard to
lung cancer risk (CYP1A1 and CYP2D6). For
CYP1A1, the evidence has varied across populations.
The finding that a particular polymorphism was
associated with increased lung cancer risk in Japan
287
was not replicated in Brazil
288
or in the United
States.
289
A different and unique polymorphism
found in African-Americans and Africans was associ-
ated with lung cancer in one case-control study of
African-Americans
290
but not in another.
291
Both phenotype and genotype have been exam-
ined for CYP2D6. This enzyme determines the
phenotype for debrisoquine metabolism, which has
been studied extensively as a risk factor for lung
cancer.
268
The initial case-control studies found that
fast metabolizers had a greater lung cancer risk,
which is consistent with the hypothesized role of the
rate of metabolism in determining lung cancer
risk,
292
although a subsequent and larger study found
no association.
293
More recent studies
267
that have
assessed genotype have generated inconsistent re-
sults.
Glutathione S-transferase is a phase II enzyme that
detoxifies reactive metabolites of polycyclic aromatic
hydrocarbons. There are at least four genetically dis-
tinct classes of the glutathione S-transferases as follows:
, , , and . The results of several studies
294,295
have
shown that individuals with high activity of the gluta-
thione S-transferase -polymorphism have lower risk
of lung cancer. Other studies, however, have not
supported this finding.
296,297
Nakachi and colleagues
287
conducted a case-control study in Japan that jointly
assessed two polymorphisms of the CYP1A1 gene and
the deficient genotype of glutathione S-transferase .
There was an indication of extremely high risk among
lower exposure smokers if both high-risk genotypes
were present.
There are other candidates for determinants of
susceptibility to lung cancer in smokers, including
oncogenes and suppressor genes and DNA repair
capacity.
267
Much research remains to be done to
clarify the association between variations in DNA
Figure 8. Scheme linking nicotine addiction and lung cancer via tobacco smoke carcinogens and their
induction of multiple mutations in critical genes. Adapted from Hecht.
106
repair capacity and lung cancer risk, but the evidence
to date suggests that this is a promising lead. For
example, a polymorphism of the DNA repair gene
XRCC1 was significantly associated with lung cancer
risk.
298
Individuals with a less proficient DNA repair
capacity phenotype, as measured by a mutagen
sensitivity assay, have been shown to have an in-
creased risk of lung cancer.
299,300
Presence of Acquired Lung Disease
In addition to hereditary factors, increased suscep-
tibility to lung cancer may result from previously
incurred lung damage. Such acquired lung diseases
assume the following two major forms: (1) those that
obstruct airflow, such as COPD; and (2) fibrotic
disorders that restrict lung capacity, such as pneu-
moconiosis.
301
Associations between lung cancer and
both types of acquired lung disease have been noted,
but, as mentioned below, this topic is complex and
many issues await resolution. These questions have
been controversial for 60 years.
302
A substantial body of evidence suggests that
COPD or impaired lung function is associated with
the occurrence of lung cancer.
303
Cigarette smoking
is the principal cause of both COPD
304
and lung
cancer, being so strongly causally associated with
both of these illnesses that presuming that statistical
adjustment procedures remove the effect of ciga-
rette smoking may not be well-founded. Thus, clar-
ifying the relevance of COPD to the development of
lung cancer awaits further proof that this association
is not accounted for by cigarette smoking. The
presence of COPD may indicate that the affected
individual has received a greater dose of tobacco
carcinogens than the typical unaffected individual.
Regardless of the mechanism, the presence of
COPD is a clinically useful risk indicator.
Clarifying the possible relationship between pneu-
moconiosis and lung cancer poses particularly vexing
challenges. Even for asbestos exposure, which clearly
has been established as a potent cause of lung
cancer,
223
whether lung cancer results from asbestos
per se and/or asbestosis remains unresolved.
224
As-
bestos is likely to cause lung cancer via multiple
mechanistic pathways.
305,306
For other mineral fi-
bers, the situation is murkier. For example, deter-
mining whether silica exposure or silicosis mediates
the increased lung cancer risk in silica-exposed per-
sons has proven difficult.
307,308
The presence of
silicosis is associated with an increased risk of lung
cancer.
211
Understanding the basis of this association
will entail isolating the independent effects of silica
exposure and lung fibrosis while controlling for
exposure to smoking and other lung carcino-
gens.
209,225
Such differences in the pattern of associations
between pneumoconiosis and lung cancer emphasize
that fibrosis is not a homogeneous exposure but
one that is dependent on the properties of the
specific mineral fiber or other environmental agent.
Properties of the agent, such as its size, shape, and
durability, and the effects of other exposures, such as
cigarette smoking, are important considerations in
assessing its potential harmfulness.
305
Conclusions
The path to preventing lung cancer is charted by
the identification of numerous exposures that are
causally associated with lung cancer. If steps can be
taken to reduce or eliminate the populations expo-
sure to these agents, this would be expected to
reduce the populations risk of lung cancer. Preven-
tive strategies can be pursued in the public policy
arena or in public health interventions that are
directed at individual behavior. Cigarette smoking
provides a useful example to illustrate the multiple
levels that can form the basis of preventive strategies.
In the legislative/regulatory arena, examples of to-
bacco control strategies include legislation that limits
cigarette advertising, that reduces childrens access
to cigarettes, and that prohibits smoking in the
workplace. Litigation against cigarette manufactur-
ers also has proven to be a productive component of
tobacco control strategies, as exemplified by the
settlement of suits between the states and the to-
bacco industry. Behavioral interventions to prevent
children and adolescents from starting to smoke
cigarettes and behavioral/pharmacologic interven-
tions to promote smoking cessation are individual-
level approaches that, if successful, could be ex-
pected to reduce the occurrence of lung cancer.
In developing lung cancer prevention strategies,
certain groups warrant particular attention. Steps
need to be taken toward the goal of reducing the
very high lung cancer incidence rates in African-
American men.
309
Lung cancer is a major health
issue for women. Due to historical cigarette smoking
patterns, the epidemic of lung cancer started later in
women than in men, but, in contrast to the situation
in men, lung cancer incidence rates in women are
still increasing.
310
Although lung cancer remains a
critical public health problem, the decrease in the
overall lung cancer burden that is presently occur-
ring in the United States, as is the case in much of
the developed world, reflects the successes of pre-
ventive strategies. A critical global priority is to
prevent the uptake of cigarette smoking in develop-
ing countries where smoking prevalence is still low in
order to prevent the increase in morbidity and
mortality from lung cancer that is certain to follow an
increase in smoking prevalence.
A consideration of the epidemiology of lung can-
cer consistently reinforces one major theme. The
pandemic of lung cancer is a consequence of the
tragic and widespread addiction to cigarettes
throughout the world. Curtailing the pandemic of
lung cancer will require preventing youths from
starting to smoke cigarettes and effectively promot-
ing smoking cessation among addicted smokers.
There are other causes that also need control, but
fortunately there have been successes in reducing
exposures to occupational carcinogens in countries of
the developed world.
ACKNOWLEDGMENT: The authors thank Charlotte Gerczak
for her assistance in preparing this manuscript.
References
1 White C, Bailar JC III. Retrospective and prospective
methods of studying association in medicine. Am J Public
Health 1956; 46:3544
2 Doll R, Hill AB. The mortality of doctors in relation to their
smoking habits: a preliminary report. BMJ 1954; 1:1451
1455
3 Doll R, Hill AB. Lung cancer and other causes of death in
relation to smoking: a second report on the mortality of
British doctors. BMJ 1956; 2:10711081
4 Wingo PA, Ries LA, Giovino GA, et al. Annual report to the
nation on the status of cancer, 19731996, with a special
section on lung cancer and tobacco smoking. J Natl Cancer
Inst 1999; 91:675690
5 Samet JM. Epidemiology of lung cancer. New York, NY:
Marcel Dekker, 1994
6 Rosen G. A history of public health. Baltimore, MD: The
Johns Hopkins University Press, 1993
7 US Department of Health, and Human Services (US-
DHHS). Reducing the health consequences of smoking: 25
years of progress; a report of the Surgeon General. Wash-
ington, DC: US Government Printing Office, 1989
8 Proctor RN. The Nazi war on cancer. Princeton, NJ: Prince-
ton University Press, 1999
9 Doll R, Hill AB. Smoking and carcinoma of the lung. BMJ
1950; 2:739748
10 Levin ML, Goldstein H, Gerhardt PR. Cancer and tobacco
smoking: a preliminary report. JAMA 1950; 143:336338
11 Wynder EL, Graham EA. Tobacco smoking as a possible
etiologic factor in bronchiogenic carcinoma: a study of six
hundred and eighty-four proved cases. JAMA 1950; 143:
329336
12 US Department of Health Education, and Welfare
(DHEW). Smoking and health: report of the Advisory
Committee to the Surgeon General. DHEW-Public Health
Service Publication No. 1103. Washington, DC: US Govern-
ment Printing Office, 1964
13 Royal College of Physicians of London. Smoking and
Health: summary of a report of the Royal College of
Physicians of London on smoking in relation to cancer of the
lung and other diseases. Publication No. S2-70. London,
UK: Pitman Medical Publishing, LTD, 1962
14 US Department of Health and Human Services (USDHHS).
The health consequences of involuntary smoking: a report of
the Surgeon General. USDHHS Publication No. (CDC) 87
8398. Washington, DC: US Government Printing Office, 1986
15 US Environmental Protection Agency (EPA). Respiratory
health effects of passive smoking: lung cancer and other
disorders. EPA Publication No. 600/006F. Washington, DC:
US Government Printing Office, 1992
16 Samet JM. Radon and lung cancer. J Natl Cancer Inst 1989;
81:745757
17 National Research Council (NRC), Committee on Health
Risks of Exposure to Radon, Board on Radiation Effects
Research. Health effects of exposure to radon (BEIR VI).
Washington, DC: National Academy Press, 1998
18 Coultas DB, Samet JM. Occupational lung cancer. Clin
Chest Med 1992; 13:341354
19 Gordon T, Crittenden M, Haenszel W. Cancer mortality
trends in the United States, 19301955. Natl Cancer Inst
Monogr 1961; 6:131350
20 Mckay FW, Hanson MR, Miller RW. Cancer mortality in
the United States: 19501977. Natl Cancer Inst Monogr
1982; 59:1357
21 Ries LAG, Miller BA, Hankey BF, et al. Cancer statistics
review, 19731988. Bethesda, MD: US Government Print-
ing Office, 1991
22 American Cancer Society. Cancer facts and figures, 1999.
Atlanta, GA: American Cancer Society, 1999
23 Jemal A, Chu KC, Tarone RE. Recent trends in lung cancer
mortality in the United States. J Natl Cancer Inst 2001;
93:277283
24 Shopland DR. Effect of smoking on the incidence and
mortality of lung cancer. In: Johnson BE, Johnson DH, eds.
Lung cancer. New York, NY: John Wiley & Sons, 1995; 114
25 Horm JW, Cicero JB. SEER program: cancer incidence and
mortality in the United States, 19731981. Washington, DC:
26 National Cancer Institute. National Cancer Institute. Sur-
veillance, Epidemiology, and End Results (SEER) Program,
2001. Available at: http://seer.cancer.gov/ Accessed October
25, 2002
27 Devesa SS, Blot WJ, Fraumeni JF Jr. Declining lung cancer
rates among young men and women in the United States: a
cohort analysis. J Natl Cancer Inst 1989; 81:15681571
28 Travis WD, Travis LB, Devesa SS. Lung cancer. Cancer
1995; 75(suppl):191202
29 US Department of Health and Human Services (US-
DHHS), Public Health Service, Centers for Disease Control
and Prevention (CDC). Preventing tobacco use among
young people: a report of the Surgeon General. Washington,
DC: US Government Printing Office, 1994
30 Coleman MP, Esteve J, Damiecki P, et al. Trends in cancer
incidence and mortality. Lyon, France: International Agency
for Research on Cancer, 1993
31 Gilliland FD, Samet JM. Incidence and mortality for lung
cancer: geographic, histologic, and diagnostic trends. Cancer
Surv 1994; 19:175195
32 National Cancer Institute (NCI), Cancer Statistics Branch,
Division of Cancer Prevention, and Control. Cancer rates
and risks. Bethesda, MD: National Institutes of Health, 1996
33 Parkin DM, Pisani P, Lopez AD, et al. At least one in seven
cases of cancer is caused by smoking: global estimates for
1995. Int J Cancer 1994; 59:494504
34 Parkin DM, Muir CS, Whelan SL, et al. Cancer incidence in
five continents. Lyon, France: International Agency for
Research on Cancer, 1992
35 Boffetta P, Parkin DM. Cancer in developing countries. CA
Cancer J Clin 1994; 44:8191
36 Stocks P, Campbell JM. Lung cancer death rates among
non-smokers and pipe and cigarette smokers: an evaluation
in relation to air pollution by benzo[a]pyrene and other
substances. BMJ 1955; 2:923929
37 Blot WJ, Fraumeni JF Jr. Arsenical air pollution and lung
cancer. Lancet 1975; 2:142144
38 Blot WJ, Fraumeni JF Jr. Geographic patterns of lung
cancer: industrial correlations. Am J Epidemiol 1976; 103:
539550
39 Mason TJ, Mckay FWUS. Cancer mortality by county:
19501969. US Department of Health, Education, and
Welfare Publication No. 74615. Washington, DC: US
Government Printing Office, 1973
40 Blot WJ, Fraumeni JF. Cancer among shipyard workers. In:
Peto R, Schneiderman M, eds. Banbury report 9: quantifi-
cation of occupational cancer. Cold Spring Harbor, NY:
Cold Spring Harbor Laboratory, 1981
41 Blot WJ, Fraumeni JF Jr. Changing patterns of lung cancer
in the United States. Am J Epidemiol 1982; 115:664673
42 Saracci R, Boffetta P. Interactions of tobacco smoking and
other causes of lung cancer. In: Samet JM, ed. Epidemiology
of lung cancer. New York, NY: Marcel Dekker, 1994;
465493
43 Samet JM, Cohen AJ. Air pollution and lung cancer. In:
Holgate ST, Samet JM, Koren HS, et al, eds. Air pollution
and health. San Diego, CA: Academic Press, 1999; 841864
44 Willett WC. Diet, nutrition, and avoidable cancer. Environ
Health Perspect 1995; 103(suppl):165170
45 Peto R, Lopez AD, Boreham J, et al. Mortality from smoking
in developed countries 19502000: indirect estimates from
national vital statistics. Oxford, UK: Oxford University Press,
1994
46 Samet JM. Lung cancer. In: Greenwald P, Kramer BS,
Weed DL, eds. Cancer prevention and control. New York,
NY: Marcel Dekker, 1995; 561584
47 Zaridze D, Peto R. Tobacco: a major international health
hazard. Lyon, France: International Agency for Research on
Cancer, 1986
DHHS), Public Health Service, Centers for Disease Con-
trol, and Prevention (CDC). Cigarette smoking: attributable
mortality and years of potential life lost, United States,
19501990. MMWR Morb Mortal Wkly Rep 1993; 42:
3739
49 Yang G, Lixin F, Tan J, et al. Smoking in China: findings of
the 1996 national prevalence survey. JAMA 1999; 282:1247
1253
50 National Institutes of Health (NIH), National Cancer Insti-
tute Smoking. Tobacco control monograph 9: cigars; health
effects and trends. NIH Publication No. 984302. Be-
thesda, MD: US Department of Health and Human Ser-
vices, 1998
51 Boffetta P, Pershagen G, Jockel KH, et al. Cigar and pipe
smoking and lung cancer risk: a multicenter study from
Europe. J Natl Cancer Inst 1999; 91:697701
52 Doll R, Peto R. Cigarette smoking and bronchial carcinoma:
dose and time relationships among regular smokers and
lifelong non-smokers. J Epidemiol Community Health 1978;
32:303313
53 Peto R. Influence of dose and duration of smoking on lung
cancer rates. In: Zaridze D, Peto R, eds. Tobacco: a major
international health hazard; proceedings of an international
meeting Moscow, USSR, June 46, 1985. International
Agency for Research on Cancer Science Publication No. 74,
2333. Lyon, France: World Health Organization, Interna-
tional Agency for Research on Cancer, 1986
DHHS). Smoking and health: a national status report.
Washington, DC: US Government Printing Office, 1987
DHHS). The health benefits of smoking cessation: a report
of the Surgeon General. USDHHS Publication No. 90
8416. Washington, DC: US Government Printing Office,
1990
56 Hrubec Z, McLaughlin JK. Former cigarette smoking and
mortality among US veterans: a 26-year follow-up, 1954
1980. In: Burns DM, Garfinkel L, Samet JM, eds. Changes
in cigarette-related disease risks and their implication for
prevention and control. Bethesda, MD: US Government
Printing Office, 1997; 501530
57 Wu-Williams AH, Samet JM. Lung cancer and cigarette
smoking. In: Samet JM, ed. Epidemiology of lung cancer.
New York, NY: Marcel Dekker, 1994; 71108
58 Hoffmann D, Hoffmann I, El Bayoumy K. The less harmful
cigarette: a controversial issue; a tribute to Ernst L. Wynder.
Chem Res Toxicol 2001; 14:767790
DHHS). The health consequences of smoking: the changing
cigarette; a report of the Surgeon General. Washington, DC:
60 National Cancer Institute (NCI). The FTC cigarette test
method for determining tar, nicotine, and carbon monoxide
yields of US cigarettes. Monograph 7. Bethesda, MD:
National Institutes of Health, 1996
61 Coultas DB, Howard CA, Peake GT, et al. Discrepancies
between self-reported and validated cigarette smoking in a
community survey of New Mexico Hispanics. Am Rev
Respir Dis 1988; 88:810814
62 Djordjevic MV, Stellman SD, Zang E. Doses of nicotine and
lung carcinogens delivered to cigarette smokers. J Natl
Cancer Inst 2000; 92:106111
DHHS), National Cancer Institute. Risks associated with
smoking cigarettes with low machine-measured yields of tar
and nicotine. Bethesda, MD: National Institutes of Health,
2001
64 Samet JM. The changing cigarette and disease risk: current
status of the evidence; the FTC Cigarette Test Method for
Determining Tar, Nicotine, and Carbon Monoxide Yields of
US Cigarettesreport of the National Cancer Institute
Expert Committee. NIH Publication No. 962789, 7792.
Bethesda, MD: US Department of Human Services, Public
Health Service, National Institutes of Heath, National Can-
cer Institute, 1996
65 Bross IDJ, Gibson R. Risks of lung cancer in smokers who
switch to filter cigarettes. Am J Public Health 1968; 58:
13961403
66 Wynder EL, Mabuchi K, Beattie EJ Jr. The epidemiology of
lung cancer: recent trends. JAMA 1970; 213:22212228
67 Wynder EL, Stellman SD. Impact of long-term filter ciga-
rette usage on lung and larynx cancer risk: a case-control
study. J Natl Cancer Inst 1979; 62:471477
68 Wynder EL, Kabat GC. The effect of low-yield cigarette
smoking on lung cancer risk. Cancer 1988; 62:12231230
69 Lubin JH, Blot WJ, Berrino F, et al. Patterns of lung cancer
risk according to type of cigarette smoked. Int J Cancer
1984; 33:569576
70 Pathak DR, Samet JM, Humble CG, et al. Determinants of
lung cancer risk in cigarette smokers in New Mexico. J Natl
Cancer Inst 1986; 76:597604
71 Jockel KH, Ahrens W, Wichmann HE, et al. Occupational
and environmental hazards associated with lung cancer. Int
J Epidemiol 1992; 21:202213
72 Hawthorne VM, Fry JS. Smoking and health: the association
between smoking behaviour, total mortality, and cardiore-
spiratory disease in west central Scotland. J Epidemiol
Community Health 1978; 32:260266
73 Rimington J. The effect of filters on the incidence of lung
cancer in cigarette smokers. Environ Res 1981; 24:162166
74 Garfinkel L, Stellman SD. Smoking and lung cancer in
women: findings in a prospective study. Cancer Res 1988;
48:69516955
75 Sidney S, Tekawa IS, Friedman GD. A prospective study of
cigarette tar yield and lung cancer. Cancer Causes Control
1993; 4:310
76 Joly OG, Lubin JH, Caraballoso M. Dark tobacco and lung
cancer in Cuba. J Natl Cancer Inst 1983; 70:10331039
77 Wilcox HB, Schoenberg JB, Mason TJ, et al. Smoking and
lung cancer: risk as a function of cigarette tar content. Prev
Med 1988; 17:263272
78 Kaufman DW, Palmer JR, Rosenberg L, et al. Tar content of
cigarettes in relation to lung cancer. Am J Epidemiol 1989;
129:703711
79 Hammond EC, Garfinkel L, Seidman H, et al. Tar and
nicotine content of cigarette smoke in relation to death rates.
Environ Res 1976; 12:263274
80 Higenbottam T, Shipley MJ, Rose G. Cigarettes, lung
cancer, and coronary heart disease: the effects of inhalation
and tar yield. J Epidemiol Community Health 1982; 36:113
117
81 Stellman SD, Garfinkel L. Lung cancer risk is proportional
to cigarette tar yield: evidence from a prospective study.
Prev Med 1989; 18:518525
82 Zang EA, Wynder EL. Differences in lung cancer risk
between men and women: examination of the evidence.
J Natl Cancer Inst 1996; 88:183192
83 Ryberg D, Hewer A, Philips DH, et al. Different suscepti-
bility to smoking-induced DNA damage among male and
female cancer patients. Cancer Res 1994; 54:58015803
DHHS), Public Health Service, National Cancer Institute
(NCI). Changes in cigarette-related disease risks and their
implication for prevention and control. Burns DM, Garf-
inkel L, Samet JM, eds. Monograph 8. Smoking and tobacco
control monograph. NIH Publication No. 974213. Be-
thesda, MD: US Government Printing Office, 1997
85 Thun MJ, Day-Lally CA, Calle EE, et al. Excess mortality
among cigarette smokers: changes in a 20-year interval. Am J
Public Health 1995; 85:12231230
86 Thun MJ, Day-Lally C, Myers DG, et al. Trends in tobacco
smoking and mortality from cigarette use in cancer preven-
tion studies I (19591965) and II (19821988). In: Burns
DM, Garfinkel L, Samet JM, eds. Changes in cigarette
related disease risks and their implication for prevention and
control. Bethesda, MD: US Government Printing Office,
1997; 305382
87 Doll R, Peto R, Wheatley K, et al. Mortality in relation to
smoking: 40 years observations on male British doctors.
BMJ 1994; 309:901911
88 Stratton K, Shetty P, Wallace R, et al. Clearing the smoke:
assessing the science base for tobacco harm reduction.
89 Hirayama T. Non-smoking wives of heavy smokers have a
higher risk of lung cancer: a study from Japan. BMJ 1981;
282:183185
90 Trichopoulos D, Kalandidi A, Sparros L, et al. Lung cancer
and passive smoking. Int J Cancer 1981; 27:14
91 National Research Council (NRC), Committee on Passive
Smoking. Environmental tobacco smoke: measuring expo-
sures and assessing health effects. Washington, DC; Na-
tional Academy Press, 1986
92 Fontham ETH, Correa P, Reynolds P, et al. Environmental
tobacco smoke and lung cancer in nonsmoking women: a
multicenter study. JAMA 1994; 271:17521759
93 Boffetta P, Agudo A, Ahrens W, et al. Multicenter case-
control study of exposure to environmental tobacco smoke
and lung cancer in Europe. J Natl Cancer Inst 1998;
90:14401450
94 Adlkofer F, Heller W. Exposure to environmental tobacco
smoke and risk of lung cancer: the epidemiologic evidence.
Eur Respir J 1995; 8:12401241
95 Gross AJ. The risk of lung cancer in nonsmokers in the
United States and its reported association with environmen-
tal tobacco smoke. J Clin Epidemiol 1995; 48:587598
96 Law MR, Morris JK, Wald NJ. Environmental tobacco
smoke exposure and ischaemic heart disease: an evaluation
of the evidence. BMJ 1997; 315:973980
97 Reynolds P. Epidemiologic evidence for workplace ETS as a
risk factor for lung cancer among nonsmokers: specific risk
estimates. Environ Health Perspect 1999; 107(suppl):865
877
98 Brown KG. Lung cancer and environmental tobacco smoke:
exposure-response relationships with application to occupa-
tional risk. Environ Health Perspect 1999; 107(suppl):885
890
99 Churg A. Lung cancer cell type and occupational exposure.
In: Samet JM, ed. Epidemiology of lung cancer. New York,
NY: Marcel Dekker, 1994; 413436
100 Wynder EL, Covey LS. Epidemiologic patterns in lung
cancer by histologic type. Eur J Cancer Clin Oncol 1987;
23:14911496
101 Wynder EL, Muscat JE. The changing epidemiology of
smoking and lung cancer histology. Environ Health Perspect
1995; 103(suppl):143148
102 Vincent RG, Pickren JW, Lane WW, et al. The changing
histopathology of lung cancer: a review of 1682 cases.
Cancer 1977; 39:16471655
103 Charloux A, Quoix E, Wolkove N, et al. The increasing
incidence of lung adenocarcinoma: reality or artefact?: a
review of the epidemiology of lung adenocarcinoma. Int J
Epidemiol 1997; 26:1423
104 Thun MJ, Lally CA, Flannery JT, et al. Cigarette smoking
and changes in the histopathology of lung cancer. J Natl
Cancer Inst 1997; 89:15801586
105 Hoffmann D, Hoffmann I. The changing cigarette, 1950
1995. J Toxicol Environ Health 1997; 50:307364
106 Hecht SS. Tobacco smoke carcinogens and lung cancer.
DHHS). Women and smoking: a report of the Surgeon
General. Rockville, MD: US Department of Health and
Human Services, 2001
108 Peto R, Doll R, Buckley JD, et al. Can dietary beta-carotene
materially reduce human cancer rates? Nature 1981; 290:
201208
109 Cade JE, Margetts BM. Relationship between diet and
smoking: Is the diet of smokers different? J Epidemiol
Community Health 1991; 45:270272
110 Margetts BM, Jackson AA. The determinants of plasma
beta-carotene: interaction between smoking and other life-
style factors. Eur J Clin Nutr 1996; 50:236238
111 Hebert JR, Kabat GC. Differences in dietary intake associ-
ated with smoking status. Eur J Clin Nutr 1990; 44:185193
112 Morabia A, Wynder EL. Dietary habits of smokers, people
who never smoked, and exsmokers. Am J Clin Nutr 1990;
52:933937
113 Zondervan KT, Ocke MC, Smit HA, et al. Do dietary and
supplementary intakes of antioxidants differ with smoking
status? Int J Epidemiol 1996; 25:7079
114 Fontham ET. Protective dietary factors and lung cancer. Int
J Epidemiol 1990; 19(suppl):S32S42
115 Axelsson G, Liljeqvist T, Andersson L, et al. Dietary factors
and lung cancer among men in west Sweden. Int J Epide-
miol 1996; 25:3239
116 Fontham ET, Pickle LW, Haenszel W, et al. Dietary
vitamins A, and C, and lung cancer risk in Louisiana. Cancer
1988; 62:22672273
117 Gao CM, Tajima K, Duroishi T, et al. Protective effects of
raw vegetables and fruit against lung cancer among smokers
and exsmokers: a case-control study in the Tokai area of
Japan. Jpn J Cancer Res 1993; 84:594600
118 Mayne ST, Janerich DT, Greenwald P, et al. Dietary
beta-carotene and lung cancer risk in US nonsmokers. J Natl
Cancer Inst 1994; 86:3338
119 Harris RWC, Key TJA, Silcocks PB, et al. A case-control
study of dietary carotene in men with lung cancer and in
men with other epithelial cancers. Nutr Cancer 1991;
15:6368
120 Hu J, Johnson KC, Mao Y, et al. A case-control study of diet
and lung cancer in northeast China. Int J Cancer 1997;
71:924931
121 Kalandidi A, Katsouyanni K, Voropoulou N, et al. Passive
smoking and diet in the etiology of lung cancer among
non-smokers. Cancer Causes Control 1990; 1:1521
122 Koo LC. Dietary habits and lung cancer risk among Chinese
females in Hong Kong who never smoked. Nutr Cancer
1988; 11:155172
123 Agudo A, Esteve MG, Pallares C, et al. Vegetable and fruit
intake and the risk of lung cancer in women in Barcelona,
Spain. Eur J Cancer 1997; 33:12561261
124 Alavanja MC, Brown CC, Swanson C, et al. Saturated fat
intake and lung cancer risk among nonsmoking women in
Missouri. J Natl Cancer Inst 1993; 85:19061916
125 Brennan P, Fortes C, Butler J, et al. A multicenter case-
control study of diet and lung cancer among non-smokers.
Cancer Causes Control 2000; 11:4958
126 Candelora EC, Stockwell HG, Armstrong AW, et al. Dietary
intake and risk of lung cancer in women who never smoked.
Nutr Cancer 1992; 17:263270
127 Dorgan JF, Ziegler RG, Schoenberg JB, et al. Race and sex
differences in associations of vegetables, fruits, and carote-
noids with lung cancer risk in New Jersey (United States).
Cancer Causes Control 1993; 4:273281
128 Forman MR, Yao SX, Graubard BI, et al. The effect of
dietary intake of fruits and vegetables on the odds ratio of
lung cancer among Yunnan tin miners. Int J Epidemiol
1992; 21:437441
129 Jain M, Burch JD, Howe GR, et al. Dietary factors and risk
of lung cancer: results from a case-control study, Toronto,
19811985. Int J Cancer 1990; 45:187193
130 Swanson CA, Mao BL, Li JY, et al. Dietary determinants of
lung-cancer risk: results from a case-control study in Yunnan
Province, China. Int J Cancer 1992; 50:876880
131 Wu-Williams AH, Dai XD, Blot W, et al. Lung cancer
among women in northeast China. Br J Cancer 1990;
62:982987
132 Ziegler RG, Mason TJ, Stemhagen A, et al. Carotenoid
intake, vegetables, and the risk of lung cancer among white
men in New Jersey. Am J Epidemiol 1986; 123:10801093
133 Chow WH, Schuman LM, McLaughlin JK, et al. A cohort
study of tobacco use, diet, occupation, and lung cancer
mortality. Cancer Causes Control 1992; 3:247254
134 Knekt P, Jarvinen R, Teppo L, et al. Role of various
carotenoids in lung cancer prevention. J Natl Cancer Inst
1999; 91:182184
135 Shibata A, Paganini-Hill A, Ross RK, et al. Intake of
vegetables, fruits, beta-carotene, vitamin C and vitamin
supplements and cancer incidence among the elderly: a
prospective study. Br J Cancer 1992; 66:673679
136 Fraser GE, Beeson WL, Phillips RL. Diet and lung cancer
in California Seventh-day Adventists. Am J Epidemiol 1991;
133:683693
137 Steinmetz KA, Potter JD, Folsom AR. Vegetables, fruit, and
lung cancer in the Iowa Womens Health Study. Cancer Res
1993; 53:536543
138 Feskanich D, Ziegler RG, Michaud DS, et al. Prospective
study of fruit and vegetable consumption and risk of lung
cancer among men and women. J Natl Cancer Inst 2000;
92:18121823
139 Kromhout D. Essential micronutrients in relation to carci-
nogenesis. Am J Clin Nutr 1987; 45:13611367
140 Le Marchand L, Yoshizawa CN, Kolonel LN, et al. Vegeta-
ble consumption and lung cancer risk: a population-based
case-control study in Hawaii. J Natl Cancer Inst 1989;
81:11581164
141 MacLennan R, Da Costa J, Day NE, et al. Risk factors for
lung cancer in Singapore Chinese, a population with high
female incidence rates. Int J Cancer 1977; 20:854860
142 Takezaki T, Hirose K, Inoue M, et al. Dietary factors and
lung cancer risk in Japanese: with special reference to fish
consumption and adenocarcinomas. Br J Cancer 2001;
84:11991206
143 Darby S, Whitley E, Doll R, et al. Diet, smoking and lung
cancer: a case-control study of 1000 cases and 1500 controls
in South-West England. Br J Cancer 2001; 84:728735
144 Pisani P, Berrino F, Macaluso M, et al. Carrots, green
vegetables and lung cancer: a case-control study. Int
J Epidemiol 1986; 15:463468
145 Bjelke E. Dietary vitamin A and human lung cancer. Int
J Cancer 1975; 15:561565
146 Bond GG, Thompson FE, Cook RR. Dietary vitamin A and
lung cancer: results of a case-control study among chemical
workers. Nutr Cancer 1987; 9:109121
147 Byers TE, Vena J, Mettlin C, et al. Dietary vitamin A, and
lung cancer risk: an analysis of histologic subtypes. Am
J Epidemiol 1984; 120:769776
148 Dartigues J-F, Dabis F, Gros N, et al. Dietary vitamin A,
beta-carotene and risk of epidermoid lung cancer in south-
western France. Eur J Epidemiol 1990; 6:261265
149 Ger L-P, Hsu W-L, Chen K-T, et al. Risk factors of lung
cancer by histological category in Taiwan. Anticancer Res
1993; 13:14911500
150 Hinds MW, Kolonel LN, Hankin JH, et al. Dietary vitamin
A, carotene, vitamin C, and risk of lung cancer in Hawaii.
Am J Epidemiol 1984; 119:227237
151 Mettlin C, Graham S, Swanson M. Vitamin A and lung
cancer. J Natl Cancer Inst 1979; 62:14351438
152 Pastorino U, Pisani P, Berrino F, et al. Vitamin A, and
female lung cancer: a case-control study on plasma and diet.
Nutr Cancer 1987; 10:171179
153 Samet JM, Skipper BJ, Humble CG, et al. Lung cancer risk
and vitamin A consumption in New Mexico. Am Rev Respir
Dis 1985; 131:198202
154 Wu AH, Henderson BE, Pike MC, et al. Smoking and other
risk factors for lung cancer in women. J Natl Cancer Inst
1985; 74:747751
155 Bandera EV, Freudenheim JL, Marshall JR, et al. Diet and
alcohol consumption and lung cancer risk in the New York
State Cohort (United States). Cancer Causes Control 1997;
8:828840
156 Byers TE, Graham S, Haughey BP, et al. Diet and lung
cancer risk: findings from the Western New York Diet
Study. Am J Epidemiol 1987; 125:351363
157 Fontham ETH, Pickle LW, Haenszel W, et al. Dietary
vitamins A, and C, and lung cancer risk in Louisiana. Cancer
1988; 62:22672273
158 Gregor A, Lee PN, Roe FJC, et al. Comparison of dietary
histories in lung cancer cases and controls with special
reference to vitamin A. Nutr Cancer 1980; 2:9397
159 Nyberg F, Isaksson I, Harris JR, et al. Misclassification of
smoking status and lung cancer risk from environmental
tobacco smoke in never-smokers. Epidemiology 1997;
8:304309
160 Paganini-Hill A, Ross RK, Gray GE, et al. Vitamin A, and
cancer incidence in a retirement community. Natl Cancer
Inst Monogr 1985; 69:133135
161 Shekelle RB, Lepper M, Liu S, et al. Dietary vitamin A, and
risk of cancer in the Western Electric study. Lancet 1981;
2:11861190
162 Coates RJ, Weiss NS, Daling JR, et al. Serum levels of
selenium and retinol and the subsequent risk of cancer.
Am J Epidemiol 1988; 128:515523
163 Connett JE, Kuller LH, Kjelsberg MO, et al. Relationship
between carotenoids and cancer: the Multiple Risk Factor
Intervention Trial (MRFIT) Study. Cancer 1989; 64:126
134
164 Criqui MH, Bangdiwala S, Goodman DS, et al. Selenium,
retinol, retinol-binding protein, and uric acid associations
with cancer mortality in a population-based prospective
case-control study. Ann Epidemiol 1991; 1:385393
165 Friedman GD, Blaner WS, Goodman DS, et al. Serum
retinol and retinol-binding protein levels do not predict
subsequent lung cancer. Am J Epidemiol 1986; 123:781789
166 Knekt P, Maatela J, Aaran RK, et al. Serum vitamin A, and
subsequent risk of cancer: cancer incidence follow-up of the
Finnish Mobile Clinic Health Survey. Am J Epidemiol 1990;
132:857870
167 Kok FJ, Van Duijn CM, Hofman A, et al. Micronutrients
and the risk of lung cancer [letter]. N Engl J Med 1987;
316:146
168 Menkes MS, Comstock GW, Vuilleumier JP, et al. Serum
beta-carotene, vitamins A and E, selenium, and the risk of
lung cancer. N Engl J Med 1986; 315:12501254
169 Orentreich N, Matias JR, Vogelman JH, et al. The predictive
value of serum beta-carotene for subsequent development
of lung cancer [letter]. Nutr Cancer 1991; 16:167169
170 Wald N, Boreham J, Bailey A. Serum retinol and subsequent
risk of cancer. Br J Cancer 1986; 54:957961
171 Willett WC, Polk BF, Underwood BA, et al. Relation of
serum vitamins A and E and carotenoids to the risk of
cancer. N Engl J Med 1984; 310:430434
172 Yuan JM, Ross RK, Chu XD, et al. Prediagnostic levels of
serum beta-cryptoxanthin and retinol predict smoking-
related lung cancer risk in Shanghai, China. Cancer Epide-
miol Biomarkers Prev 2001; 10:767773
173 Le Marchand L, Hankin JH, Kolonel LN, et al. Intake of
specific carotenoids and lung cancer risk. Cancer Epidemiol
Biomarkers Prev 1993; 2:183187
174 Michaud DS, Feskanich D, Rimm EB, et al. Intake of
specific carotenoids and risk of lung cancer in 2 prospective
US cohorts. Am J Clin Nutr 2000; 72:990997
175 Nyberg F, Agrenius V, Svartengren K, et al. Dietary factors
and risk of lung cancer in never-smokers. Int J Cancer 1998;
78:430436
176 Hu FB, Stampfer MJ, Manson JE, et al. Dietary fat intake
and the risk of coronary heart disease in women. N Engl
J Med 1997; 337:14911499
177 Mettlin C. Milk drinking, other beverage habits and lung
cancer risks. Int J Cancer 1989; 43:608612
178 Voorrips LE, Goldbohm RA, Verhoeven DT, et al. Vegeta-
ble and fruit consumption and lung cancer risk in the
Netherlands Cohort Study on diet and cancer. Cancer
Causes Control 2000; 11:101115
179 Comstock GW, Alberg AJ, Huang HY, et al. The risk of
developing lung cancer associated with antioxidants in the
blood: ascorbic acid, carotenoids, alpha-tocopherol, sele-
nium, and total peroxyl radical absorbing capacity. Cancer
Epidemiol Biomarkers Prev 1997; 6:907916
180 Stahelin HB, Rosel F, Buess E, et al. Cancer, vitamins, and
plasma lipids: prospective Basel study. J Natl Cancer Inst
1984; 73:14631468
181 Alpha-Tocopheral Beta-Carotene Cancer Prevention Study
Group. The effect of vitamin E and beta-carotene on the
incidence of lung cancer and other cancers in male smokers.
N Engl J Med 1994; 330:10291035
182 Hennekens CH, Buring JE, Manson JE, et al. Lack of effect
of long-term supplementation with beta-carotene on the
incidence of malignant neoplasms and cardiovascular dis-
ease. N Engl J Med 1996; 334:11451149
183 Omenn GS, Goodman GE, Thornquist MD, et al. Effects of
a combination of beta-carotene and vitamin A on lung
cancer and cardiovascular disease. N Engl J Med 1996;
334:11501155
184 Doll R, Peto R. The causes of cancer: quantitative estimates
of avoidable risks of cancer in the United States today. J Natl
Cancer Inst 1981; 66:11911308
185 Damber LA, Larsson LG. Occupation and male lung cancer:
a case-control study in northern Sweden. Br J Ind Med
1987; 44:446453
186 Dave SK, Edling C, Jacobsson P, et al. Occupation, smoking,
and lung cancer. Br J Ind Med 1988; 45:790792
187 Kvale G, Bjelke E, Heuch I. Occupational exposure and
lung cancer risk. Int J Cancer 1986; 37:185193
188 Lerchen ML, Wiggins CL, Samet JM. Lung cancer and
occupation in New Mexico. J Natl Cancer Inst 1987; 79:
639645
189 Schoenberg JB, Stemhagen A, Mason TJ, et al. Occupation
and lung cancer risk among New Jersey white males. J Natl
Cancer Inst 1987; 79:1321
190 Vineis P, Thomas T, Hayes RB, et al. Proportion of lung
cancers in males, due to occupation, in different areas of the
USA. Int J Cancer 1988; 42:851856
191 Siemiatycki J. Risk factors for cancer in the workplace.
Boston, MA: CRC Press, 1991
192 Samet JM, Lerchen ML. Proportion of lung cancer caused
by occupation: a critical review. In: Bernard J, Gee L, Keith
W, et al, eds. Occupational lung disease. New York, NY:
Raven Press, 1984: 5567
193 Doll R, Fisher REW, Gammon EJ, et al. Mortality of gas
workers with special reference to cancers of the lung and
bladder, chronic bronchitis, and pneumoconiosis. Br J Ind
Med 1965; 22:112
194 Lloyd JW. Long-term mortality of steelworkers: V. Respira-
tory cancer in coke plant workers. J Occup Med 1971;
13:5368
195 Lawther PJ, Commins BT, Waller RE. A study of the
concentrations of polycyclic aromatic hydrocarbons in gas
works retort houses. Br J Ind Med 1965; 22:1320
196 Lee AM, Fraumeni JF Jr. Arsenic and respiratory cancer in
man: an occupational study. J Natl Cancer Inst 1969;
42:10451052
197 Ott MG, Holder BB, Gordon HL. Respiratory cancer and
occupational exposure to arsenicals. Arch Environ Health
1974; 29:250255
198 Pinto SS, Henderson V, Enterline PE. Mortality experience
of arsenic-exposed workers. Arch Environ Health 1978;
33:325331
199 Alderson MR, Rattan NS, Bidstrup L. Health of workmen in
the chromate-producing industry in Britain. Br J Ind Med
1981; 38:117124
200 Bidstrup PL, Case RAM. Carcinoma of the lung in workmen
in the bichromates-producing industry in Great Britain. Br
J Ind Med 1956; 13:260264
201 Taylor FH. The relationship of mortality and duration of
employment as reflected by a cohort of chromate workers.
Am J Public Health 1966; 56:218229
202 Doll R. Cancer of the lung and nose in nickel workers. Br
J Ind Med 1958; 15:217223
203 Kreyberg L. Lung cancer in workers in a nickel refinery. Br
J Ind Med 1978; 35:109116
204 Pedersen E, Hogetveit AC, Andersen A. Cancer of respira-
tory organs among workers at a nickel refinery in Norway.
Int J Cancer 1973; 12:3241
205 Friberg L, Cederlof R. Late effects of air pollution with
special reference to lung cancer. Environ Health Perspect
1978; 22:4566
206 Sunderman JW Jr. A review of the carcinogenicities of
nickel, chromium and arsenic compounds in man and
animals. Prev Med 1976; 5:279294
207 Cohen AJ, Higgins MWP. Health Effects of diesel exhaust:
epidemiology. In: Diesel exhaust: a critical analysis of emis-
sions, exposure, and health effects; a special report of the
Health Effects Institutes working group. Cambridge, MA:
Health Effects Institute, 1995; 251292
208 Goldsmith DF, Guidotti TL, Johnston DR. Does occupa-
tional exposure to silica cause lung cancer? Am J Ind Med
1982; 3:423440
209 Heppleston AG. Silica, pneumoconiosis, and carcinoma of
the lung. Am J Ind Med 1985; 7:285294
210 McDonald JC. Silica, silicosis, and lung cancer. Br J Ind
Med 1989; 46:289291
211 Smith AH, Lopipero PA, Barroga VR. Meta-analysis of
studies of lung cancer among silicotics. Epidemiology 1995;
6:617624
212 Checkoway H, Franzblau A. Is silicosis required for silica-
associated lung cancer? Am J Ind Med 2000; 37:252259
213 Hessel PA, Gamble JF, Gee JB, et al. Silica, silicosis, and
lung cancer: a response to a recent working group report. J
Occup Environ Med 2000; 42:704720
214 International Agency for Research on Cancer (IARC). IARC
monographs on the evaluation of carcinogenic risks to
humans: silica, some silicates, coal dust and para-aramid
fibrils (vol 68). Lyon, France: World Health Organization,
IARC, 1997
215 Health Effects Institute, Asbestos Research Committee,
Literature Review Panel. Asbestos in public and commercial
buildings: a literature review and a synthesis of current
knowledge. Cambridge, MA: Health Effects Institute, 1991
216 Wedler HW. Asbestosis and pulmonary carcinoma (asbes-
tose und lungenkrebs). Bull Hyg 1944; 19:362
217 Lynch KM, Smith WA. Pulmonary asbestosis: V. A report of
bronchial carcinoma and epithelial metaplasia. Am J Cancer
1939; 36:567573
218 Doll R. Mortality from lung cancer in asbestos workers. Br
J Ind Med 1955; 12:8186
219 Selikoff IJ, Churg J, Hammond EC. Asbestos exposure and
neoplasia. JAMA 1964; 188:2226
220 Selikoff IJ, Hammond EC, Seidman H. Mortality experience
of insulation workers in the United States and Canada. Ann
N Y Acad Sci 1979; 330:91116
221 Selikoff IJ. Latency of asbestos disease among insulation
workers in the United States and Canada. Cancer 1980;
46:27362740
222 Newhouse ML, Berry G. Patterns of mortality in asbestos
factory workers in London. Ann N Y Acad Sci 1979;
330:5360
223 Lemen RA, Dement JM, Wagoner JK. Epidemiology of
asbestos-related diseases. Environ Health Perspect 1980;
34:111
224 Browne K. Is asbestos or asbestosis the cause of the
increased risk of lung cancer in asbestos workers? Br J Ind
Med 1986; 43:145149
225 Samet JM. Does idiopathic pulmonary fibrosis increase lung
cancer risk? Am J Respir Crit Care Med 2000; 161:12
226 Hammond EC, Selikoff IJ, Seidman H. Asbestos exposure,
cigarette smoking and death rates. Ann N Y Acad Sci 1979;
330:473490
227 Churg A, Stevens B. Enhanced retention of asbestos fibers
in the airways of human smokers. Am J Respir Crit Care
Med 1995; 151:14091413
228 Boice JD Jr. Studies of atomic bomb survivors: understand-
ing radiation effects. JAMA 1990; 264:622623
229 MacMahon B. Some recent issues in low-exposure radiation
epidemiology. Environ Health Perspect 1989; 81:131135
230 Hendee WR. Estimation of radiation risks: BEIR V and its
significance for medicine. JAMA 1992; 268:620624
231 National Research Council (NRC), Committee on the Bio-
logical Effects of Ionizing Radiation. Health risks of radon
and other internally deposited alpha-emitters: BEIR IV.
232 Lubin JH, Boice JD Jr, Edling C, et al. Lung cancer in
radon-exposed miners and estimation of risk from indoor
exposure. J Natl Cancer Inst 1995; 87:817827
233 US Environmental Protection Agency (EPA). Technical
support document for the 1992 citizens guide to radon.
234 Lubin JH, Boice JD Jr, Samet JM. Errors in exposure
assessment, statistical power, and the interpretation of resi-
dential radon studies. Radiat Res 1995; 144:329341
235 Shimizu Y, Schull WJ, Kato H. Cancer risk among atomic
bomb survivors. JAMA 1990; 264:601604
236 Darby SC, Doll R, Gill SK, et al. Long-term mortality after
a single treatment course with x-rays in patients treated for
ankylosing spondylitis. Br J Cancer 1987; 55:179190
237 Davis FG, Boice JD Jr, Hrubec Z, et al. Cancer mortality in
radiation-exposed cohort of Massachusetts tuberculosis pa-
tients. Cancer Res 1989; 49:61306136
238 Howe GR. Lung cancer mortality between 1950 and 1987
after exposure to fractionated moderate-dose-rate ionizing
radiation in the Canadian Fluoroscopy Cohort Study and a
comparison with mortality in Atomic Bomb Survivors Study.
Radiat Res 1995; 142:295304
239 Gilbert ES, Cragle DL, Wiggs LD. Updated analyses of
combined mortality data for workers at the Hanford Site,
Oak Ridge National Laboratory, and Rocky Flats Weapons
Plant. Radiat Res 1993; 136:408421
240 National Research Council (NRC), Commission on Life
Sciences, Board on Toxicology, and Environmental Health.
Hazards. Epidemiology and air pollution. Publication No.
1224. Washington, DC: National Academy Press, 1985
241 Doll R. Atmospheric pollution and lung cancer. Environ
Health Perspect 1978; 22:2331
242 Pershagen G. Air pollution and cancer. In: Vainio H, Sorsa
M, McMichael AJ, eds. Complex mixtures and cancer risk.
Lyon, France: International Agency for Research on Cancer,
1990; 240251
243 Buffler PA, Cooper SP, Stinnett S, et al. Air pollution and
lung cancer mortality in Harris County, Texas, 19791981.
Am J Epidemiol 1988; 128:683699
244 Nasca PC, Burnett WS, Greenwald P, et al. Population
density as an indicator of urban-rural differences in cancer
incidence, upstate New York, 19681972. Am J Epidemiol
1980; 112:362375
245 Stocks P. Recent epidemiological studies of lung cancer
mortality, cigarette smoking and air pollution, with discus-
sion of a new hypothesis of causation. Br J Cancer 1966;
20:596623
246 Shannon HS, Hertzman C, Julian JA, et al. Lung cancer and
air pollution in an industrial city: a geographical analysis.
Can J Public Health 1988; 79:255259
247 Weinberg GB, Kuller LH, Redmond CK. The relationship
between the geographic distribution of lung cancer inci-
dence and cigarette smoking in Allegheny County, Pennsyl-
vania. Am J Epidemiol 1982; 115:4058
248 Speizer FE, Samet JM. Air pollution and lung cancer. In:
Samet JM, ed. Epidemiology of lung cancer. New York, NY:
Marcel Dekker, 1994; 131150
249 Dockery DW, Pope CA III, Xu X, et al. An association
between air pollution and mortality in six U.S. cities. N Engl
J Med 1993; 329:17531759
250 Pope CA, III, Thun MJ, Namboodiri MM, et al. Particulate
air pollution as a predictor of mortality in a prospective study
of U.S. adults. Am J Respir Crit Care Med 1995; 151:669
674
251 Speizer FE. Assessment of the epidemiological data relating
lung cancer to air pollution. Environ Health Perspect 1983;
47:3342
252 Barbone F, Bovenzi M, Cavalleri F, et al. Air pollution and
lung cancer in Trieste, Italy. Am J Epidemiol 1995; 141:
11611169
253 Jedrychowski W, Becher H, Wahrendorf J, et al. A case-
control study of lung cancer with special reference to the
effect of air pollution in Poland. J Epidemiol Community
Health 1990; 44:114120
254 Vena JE. Air pollution as a risk factor in lung cancer. Am
J Epidemiol 1982; 116:4256
255 Samet JM, Humble CG, Skipper BE, et al. History of
residence and lung cancer risk in New Mexico. Am
J Epidemiol 1987; 125:800811
256 Brown LM, Pottern LM, Blot WJ. Lung cancer in relation to
environmental pollutants emitted from industrial sources.
Environ Res 1984; 34:250261
257 Pershagen G. Lung cancer mortality among men living near
an arsenic-emitting smelter. Am J Epidemiol 1985; 122:
684694
258 Xu Z-Y, Blot WJ, Xiao HP, et al. Smoking, air pollution and
the high rates of lung cancer in Shenyang, China. J Natl
Cancer Inst 1989; 81:18001806
259 Greaves WW, Rom WN, Lyon JL, et al. Relationship
between lung cancer and distance of residence from non-
ferrous smelter stack effluent. Am J Ind Med 1981; 2:1523
260 Rom WN, Varley G, Lyon JL, et al. Lung cancer mortality
among residents living near the El Paso smelter. Br J Ind
Med 1982; 39:269272
261 Spengler JD. Sources and concentrations of indoor air
pollution. In: Samet JM, Spengler JD, eds. Indoor air
pollution: a health perspective. Baltimore, MD: Johns Hop-
kins University Press, 1991; 3367
262 Harrje DT. Building dynamics and indoor air quality. In:
Samet JM, Spengler JD, eds. Indoor air pollution: a health
perspective. Baltimore, MD: Johns Hopkins University
Press, 1991; 6881
263 Chen BH, Hong CJ, Pandey MR, et al. Indoor air pollution
in developing countries. World Health Stat Q 1990; 43:127
138
264 Mumford JL, He XZ, Chapman RS, et al. Lung cancer and
indoor air pollution in Xuan Wei, China. Science 1987;
235:217220
265 Mumford JL, Helmes CT, Lee XM, et al. Mouse skin
tumorigenicity studies of indoor coal and wood combustion
emissions from homes of residents in Xuan Wei, China with
high lung cancer mortality. Carcinogenesis 1990; 11:397
403
266 Gao YT, Blot WJ, Zheng W, et al. Lung cancer among
Chinese women. Int J Cancer 1987; 40:604609
267 Vineis P, Caporaso N. Tobacco and cancer: epidemiology
and the laboratory. Environ Health Perspect 1995; 103:156
160
268 Economou P, Lechner JF, Samet JM. Familial and genetic
factors in the pathogenesis of lung cancer. In: Samet JM, ed.
Epidemiology of lung cancer. New York, NY: Marcel Dek-
ker, 1994; 353396
269 Tokuhata GK, Lilienfeld AM. Familial aggregation of lung
cancer among hospital patients. Public Health Rep 1963;
78:277283
270 Tokuhata GK. Familial factors in human lung cancer and
smoking. Am J Public Health 1964; 54:2432
271 Ooi WL, Elston RC, Chen VW, et al. Increased familial risk
for lung cancer. J Natl Cancer Inst 1986; 76:217222
272 Samet JM, Humble CG, Pathak DR. Personal and family
history of respiratory disease and lung cancer risk. Am Rev
Respir Dis 1986; 134:466470
273 Tsugane S, Watanabe S, Sugimura H, et al. Smoking,
occupation and family history in lung cancer patients under
fifty years of age. Jpn J Clin Oncol 1987; 17:309317
274 Horwitz RI, Smaldone LF, Viscoli CM. An ecogenetic
hypothesis for lung cancer in women. Arch Intern Med
1988; 148:26092612
275 McDuffie HH. Clustering of cancer in families of patients
with primary lung cancer. J Clin Epidemiol 1991; 44:6976
276 Osann KE. Lung cancer in women: the importance of
smoking, family history of cancer, and medical history of
respiratory disease. Cancer Res 1991; 51:48934897
277 Shaw GL, Falk RT, Pickle LW, et al. Lung cancer risk
associated with cancer in relatives. J Clin Epidemiol 1991;
44:429437
278 Filho YW, Magaldi C, Nakao N, et al. Industrial work and
lung cancer (in Spanish). Rev Saude Publica 1995; 29:166
176
279 Sellers TA, Bailey-Wilson JE, Elston RC, et al. Evidence for
Mendelian inheritance in the pathogenesis of lung cancer.
280 Braun MM, Caporaso NE, Page WF, et al. Genetic compo-
nent of lung cancer: cohort study of twins. Lancet 1994;
344:440443
281 Hussain SP, Harris CC. Molecular epidemiology of human
cancer: contribution of mutation spectra studies of tumor
suppressor genes. Cancer Res 1998; 58:40234037
282 International Agency for Research on Cancer (IARC). IARC
monographs on the evaluation of the carcinogenic risk of
chemicals to humans: tobacco smoking. Monograph No. 38.
Lyon, France: World Health Organization, IARC, 1986
283 International Agency for Research on Cancer (IARC). Can-
cer incidence in five continents. Lyon, France: World
Health Organization, 1992
284 Schulte PA. A conceptual and historical framework for
molecular epidemiology. In: Schulte PA, Perera FP, eds.
Molecular epidemiology. New York, NY: Academic Press,
1993
285 Schulte PA, Perera FP. Molecular epidemiology: principles
and practices. New York, NY: Academic Press, 1993
286 Bergen AW, Caporaso N. Cigarette smoking. J Natl Cancer
Inst 1999; 91:13651375
287 Nakachi K, Imai K, Hayashi S, et al. Polymorphisms of the
CYP1A1 and glutathione S-transferase genes associated with
susceptibility to lung cancer in relation to cigarette dose in a
Japanese population. Cancer Res 1993; 53:29942999
288 Sugimura H, Suzuki I, Hamada GS, et al. Cytochrome P-450
1A1 genotype in lung cancer patients and controls in Rio de
Janeiro, Brazil. Cancer Epidemiol Biomarkers Prev 1994;
3:145148
289 Shields PG, Caporaso NE, Falk RT, et al. Lung cancer, race,
and a CYP1A1 genetic polymorphism. Cancer Epidemiol
290 Taioli E, Crofts F, Trachman J, et al. A specific African-
American CYP1A1 polymorphism is associated with adeno-
carcinoma of the lung. Cancer Res 1995; 55:472473
291 Kelsey KT, Wiencke JK, Spitz MR. A race-specific genetic
polymorphism in the CYP1A1 gene is not associated with
lung cancer in African Americans. Carcinogenesis 1994;
15:11211124
292 Amos CI, Caporaso NE, Weston A. Host factors in lung
cancer risk: a review of interdisciplinary studies. Cancer
293 Shaw GL, Falk RT, Deslauriers J, et al. Debrisoquine
metabolism and lung cancer risk. Cancer Epidemiol Bio-
markers Prev 1995; 4:4148
294 Nazar-Stewart V, Motulsky AG, Eaton DL, et al. The
glutathione S-transferase polymorphism as a marker for
susceptibility to lung carcinoma. Cancer Res 1993;
53(suppl):23132318
295 Seidegard J, Pero RW, Markowitz MM, et al. Isoenzyme(s)
of glutathione transferase (class Mu) as a marker for the
susceptibility to lung cancer: a follow-up study. Carcinogen-
esis 1990; 11:3336
296 Brockmoller J, Kerb R, Drakoulis N, et al. Genotype and
phenotype of glutathione S-transferase class mu isoenzymes
mu and psi in lung cancer patients and controls. Cancer Res
1993; 53:10041011
297 Zhong S, Howie AF, Ketterer B, et al. Glutathione S-
transferase m locus: use of genotyping and phenotyping
assays to assess association with lung cancer susceptibility.
Carcinogenesis 1991; 12:15331537
298 Ratnasinghe D, Yao SX, Tangrea JA, et al. Polymorphisms of
the DNA repair gene XRCC1 and lung cancer risk. Cancer
299 Spitz MR, Hsu TC, Wu X, et al. Mutagen sensitivity as a
biological marker of lung cancer risk in African Americans.
Cancer Epidemiol Biomarkers Prev 1995; 4:99103
300 Wei Q, Cheng L, Amos CI, et al. Repair of tobacco
carcinogen-induced DNA adducts and lung cancer risk: a
molecular epidemiologic study. J Natl Cancer Inst 2000;
92:17641772
301 Cotran RS, Kumar V, Robbins SL. Robbins pathologic basis
of disease. 5th ed. Philadelphia, PA: WB Saunders, 1994
302 Macklin MT, Macklin CC. Does chronic irritation cause
primary carcinoma of the human lung? Arch Pathol 1940;
30:924955
303 Tockman MS. Other host factors and lung cancer suscepti-
bility. In: Samet JM, ed. Epidemiology of lung cancer. New
York, NY: Marcel Dekker, 1994; 397412
DHHS). The health consequences of smoking: chronic
obstructive lung disease; a report of the Surgeon General.
305 Barrett JC, Lamb PW, Wiseman RW. Multiple mechanisms
for the carcinogenic effects of asbestos and other mineral
fibers. Environ Health Perspect 1989; 81:8189
306 Rom WN, Travis WD, Brody AR. Cellular and molecular
basis of the asbestos-related diseases. Am Rev Respir Dis
1991; 143:408422
307 Ng TP. Silica and lung cancer: a continuing controversy. Ann
Acad Med Singapore 1994; 23:752755
308 Pairon JC, Brochard P, Jaurand MC, et al. Silica and lung
cancer: a controversial issue. Eur Respir J 1991; 4:730744
DHHS). Tobacco use among US racial/ethnic minority
groups: African Americans, American Indians and Alaska
Natives, Asian Americans and Pacific Islanders, Hispanics; a
report of the Surgeon General. Washington, DC: USDHHS,
1998
310 Ernster VL. Impact of tobacco on womens health. In:
Samet JM, Yoon S-Y, eds. Women and the tobacco epi-
demic: challenges for the 21st century. Geneva, Switzerland:
World Health Organization, 2001; 116
2003;123;21-49 Chest
Anthony J. Alberg and Jonathan M. Samet
Epidemiology of Lung Cancer
& Services
Updated Information
References
#BIBL
free at:
Citations
#otherarticles
Reprints
2003;123;50-59 Chest
Michael J. Kelley and Douglas C. McCrory
Prevention of Lung Cancer: Summary of Published Evidence
ISSN: 0012-3692.
Prevention of Lung Cancer*
Summary of Published Evidence
Michael J. Kelley, MD; and Douglas C. McCrory, MD, MHS
Study objectives: To describe empiric research related to lung cancer prevention strategies,
including chemoprevention aimed at reducing lung cancer incidence and various smoking
avoidance and cessation interventions aimed at reducing smoking rates.
Design, setting, and participants: Systematic searches of MEDLINE, HealthStar, and Cochrane
Library databases to July 2001 and print bibliographies. For chemoprevention studies, we considered
only randomized controlled trials (RCTs) with lung cancer incidence as an end point. For studies of
smoking avoidance or cessation, we selected systematic reviews and meta-analyses, and searched for
individual RCTs only where high-quality and current reviews and meta-analyses were not available.
Measurement and results: Chemoprevention of lung cancer has been studied in five RCTs of primary
prevention, no RCTs of secondary prevention, and five RCTs of tertiary prevention. None of these
trials has shown evidence for efficacy of any agents tested, including retinol (vitamin A), -carotene,
N-acetylcysteine, and selenium. There is a great deal of evidence about a wide variety of clinician-
based and community-based efforts at smoking avoidance or cessation. Certain approaches have been
shown to be effective (eg, mass media public education campaigns, direct restrictions on smoking,
clinician-based approaches ranging from brief clinician advice to more in-depth sessions, and
life-skills training in schools). Some approaches have intermediate or short-term effectiveness (ie,
youth access restrictions and school-based interventions), and others have been shown to be
ineffective (ie, acupuncture and provider education) or have been insufficiently studied (ie, provider
feedback).
Conclusions: There are no agents that have been proven to be effective for preventing lung cancer.
Several clinician-based and community-based interventions show promise for reducing lung cancer
incidence through smoking avoidance and prevention. (CHEST 2003; 123:50S59S)
Key words: carotenoids; chemoprevention; health education; lung neoplasms; primary prevention; smoking cessation;
vitamin A
Abbreviations: ATBC-Tocopherol -Carotene Lung Cancer Prevention Study; CARET -Carotene and Retinol
Efficacy Trial; CI confidence interval; NSCLCnon-small cell lung cancer; RCT randomized controlled trial;
RR relative risk
C
igarette smoking is causally associated with the
development of cancer of the lung, which is the
leading cause of cancer mortality in the United
States and worldwide. More Americans die of lung
cancer each year than breast, prostate, and colon
cancer combined.
1
The battle to decrease lung cancer
mortality has been waged on the following four fronts:
(1) treatment of disease; (2) early detection; (3) che-
moprevention; and (4) smoking avoidance and cessa-
tion. This article will focus on the latter two fronts.
Chemoprevention is the use of specific natural or
synthetic chemical agents to inhibit the development of
invasive cancer by blocking the DNA damage that
initiates carcinogenesis or by reversing or arresting the
progression of premalignant cells.
2
Chemoprevention
strategies can be applied to the prevention of lung
cancer in those persons with known risk factors (pri-
mary chemoprevention), those persons with disease
precursors (secondary chemoprevention), or those per-
sons with a prior cancer that had been treated with
curative intent (tertiary chemoprevention). As a strong
and prevalent risk factor for lung cancer, tobacco
smoking has been the target for the prevention of lung
cancer and other smoking-related diseases.
We searched for phase III studies of putative chemopreventive
agents used for primary, secondary, or tertiary prevention in
which the primary end point was lung cancer incidence. We
*From the Department of Medicine, Duke University Medical
Center, Durham, NC.
Correspondence to: Michael J. Kelley, MD, Hematology/Oncology
(111G), Durham VAMC, 508 Fulton St, Durham NC 27705;
e-mail: kelleym@duke.edu
conducted computerized searches of the MEDLINE biblio-
graphic database from 1966 to July 2001, the HealthStar data-
base, and the Cochrane Library. We searched using the terms
lung neoplasm, prevention and control and smoking, prevention
and control, along with terms to identify randomized controlled
trials (RCTs), systematic reviews, meta-analyses, and practice
guidelines. In addition, we searched the reference lists of in-
cluded studies, practice guidelines, systematic reviews, and meta-
analyses.
For chemoprevention studies, we considered only RCTs with
lung cancer incidence as an end point. For studies of smoking
avoidance or cessation, we selected systematic reviews and
meta-analyses, and we searched for individual RCTs only where
high-quality and current reviews and meta-analyses were not
available.
Results
Primary Chemoprevention Interventions
Risk factors for the development of lung cancer
include smoking cigarettes or other tobacco prod-
ucts, asbestos exposure, and radon exposure. Eight
publications
310
describing five RCTs of primary
prevention aimed at reducing lung cancer incidence
in subjects with one or more of these risk factors
were identified (Table 1). Four of the studies tar-
geted high-risk groups, while the Physicians Health
Study
6
targeted a group with lower than average risk
of lung cancer.
Although none of the interventions was shown to
be effective at preventing lung cancer, the results
were consistent in showing that -carotene, rather
than reducing lung cancer incidence, was associated
with increased lung cancer incidence. Statistically
significant increases in lung cancer incidence in
smokers receiving -carotene supplements were
shown in the -Tocopherol, -Carotene (ATBC) Lung
Cancer Prevention Study
35
and the -Carotene and
Retinol Efficacy Trial (CARET).
8,9
A similar trend was
observed in a small study comparing -carotene and
retinol (vitamin A) in asbestos workers,
7
although the
difference was not statistically significant.
A closer examination of these studies is instructive
for the design of future studies. The ATBC trial
35
examined the effect of -tocopherol (vitamin E) and
-carotene on the incidence of lung cancer in 29,133
Finnish male smokers using a 2 2 factorial design.
The selection of these agents was based almost
exclusively on epidemiologic studies linking a vege-
table-rich diet (high in -carotene and vitamin E)
with a decrease in the risk of lung cancer.
11
-
carotene and vitamin E have antioxidant properties
in vitro. Subjects were recruited from 1985 to 1993
and took supplements (ie, -tocopherol, 50 mg/d,
and -carotene, 20 mg/d) for 5 to 8 years (mean, 6.1
years) for a total follow-up of 169,751 subject-years.
No effect of -tocopherol (vitamin E) on lung cancer
incidence was observed (relative risk [RR], 0.98;
95% confidence interval [CI], 0.86 to 1.12). How-
ever, in the -carotene arms, the RR of lung cancer
incidence was 1.18 (95% CI, 1.03 to 1.36). This
effect was most pronounced in those who persons
who smoked 20 cigarettes a day and in those with
higher alcohol intake.
5
The CARET study
8,9
was a large two-arm study
that was designed to compare the effects of a
combination of -carotene and retinol to those of
placebo on lung cancer incidence in subjects who
were at high risk for the development of lung cancer.
The rationale for the study was based on the results
of observational epidemiologic studies that had dem-
onstrated a statistically significantly decreased RR of
lung cancer between the extreme quintiles or quar-
tiles of dietary intake and serum levels of -carotene
and vitamin A.
1113
In addition, vitamin A analogs
were demonstrated to have potential utility in pre-
venting cancer in animal models. A pilot study
14
of
1,029 high-risk subjects (ie, those persons with 20
pack-years of cigarette smoking who were currently
smoking or had quit within the previous 6 years)
demonstrated the safety and tolerability of -
carotene, 50 mg daily, retinol, 25,000 IU daily, and
the combination of the two. The completed study
enrolled a total of 18,314 subjects, including current
and former smokers of both sexes and male asbestos
workers (ie, those persons who had worked with
asbestos at least 15 years prior to study enrollment
who had evidence of asbestosis on chest radiogram),
with a total of 73,135 person-years of follow-up and
a mean length of follow-up of 4.0 years. The active
intervention was continued throughout this period of
time. A planned interim analysis demonstrated a
statistically significantly increased RR for the devel-
opment of lung cancer (RR, 1.28; 95% CI, 1.04 to
1.57), death from any cause (RR, 1.17; 95% CI, 1.03
to 1.33), and death from lung cancer (RR, 1.46; 95%
CI, 1.07 to 2.00), resulting in early termination of the
active intervention arm. It has been suggested
15,16
that higher concentrations of -carotene resulting
from supplementation may have pro-oxidant effects,
inducing DNA damage and membrane instability.
The Physicians Health Study
6
was a randomized,
double-blind, placebo-controlled trial of aspirin
(325 mg every other day) and -carotene (50 mg
every other day) conducted using a 2 2 factorial
design among 22,071 male physicians in the United
States. Only 11% of participants were current smok-
ers and 39% were former smokers at study entry.
The study began in 1982, and the aspirin arms were
terminated in 1988 when a significant reduction in
first myocardial infarction incidence was noted. The
primary end points were cardiovascular disease and
cancer incidence. An analysis of the cancer incidence
in patients in the -carotene arms at the end of the
Table 1RCTs Examining the Effect of Primary Chemoprevention Interventions on Lung Cancer Incidence*
Study/Year No. Population Intervention End Point Result RR (95% CI) Comments
ATBC
35
/1994 29,133 Male smokers
(Finland)
-tocopherol, 50 mg/
d, -carotene
Lung cancer
incidence
444 cases 0.98 (0.861.12) Blinded; statistically significant
increase in lung cancer
incidence seen in the
smokers receiving -
carotene
Placebo -carotene 450 cases 1
-carotene, 20 mg/
d, -tocopherol
482 cases 1.18 (1.031.36)
Placebo -tocopherol 412 cases 1
CARET
8,9
/1996 18,314 Smokers (64%);
former smokers
(34%); asbestos
workers; age 45
74 yr; men and
women (except
only male asbestos
workers)
-carotene, 30 mg/d,
and retinol 25,000
IU/d
Lung cancer
incidence
5.92/1,000
person-yr
1.28 (1.041.57) Blinded; statistically significant
increase in relative risk of
lung cancer in those
receiving -carotene alone
or in combination with
retinol
Placebo 4.62/1,000
person-yr
1
Physicians Health
Study
6
/2000
22,071 Male physicians
(US); 11% smokers
-carotene, 50
mg, ASA, 325 mg
every other day
Lung cancer
incidence
85 cases 0.9 (0.71.2) Blinded; cancer incidence in
the -carotene arms
showed no difference in the
incidence of total cancers,
lung cancer, or any other
cancer. No information on
effect ASA as on lung
cancer incidence
Placebo ASA, 325 mg
every other day
93 cases
-carotene, 50
mg, ASA, 325 mg
every other day
All cancer
incidence
1,314 cases 1.0 (0.91.0)
Placebo ASA, 325 mg
every other day
1,353 cases
-carotene, 50
mg, ASA, 325 mg
every other day
Cancer deaths 414 1.0 (0.91.2) Lung cancer incidence
Placebo ASA, 325 mg
every other day
406
Womens Health
Study
10
/1999
39,876 Female health
professionals;
45 yr of age
-carotene, 50 mg
qod, ASA
vitamin E
Lung cancer
incidence
30 cases 1.43 (0.822.49) Blinded; -carotene arm
terminated early due to
results of ATBC and
CARET studies, limiting
ability of study to detect
effect of -carotene on lung
cancer incidence
Placebo ASA
vitamin E
21
-carotene, 50 mg
qod, ASA
vitamin E
All cancer
incidence
378 1.03 (0.891.18)
Placebo ASA
vitamin E
369
-carotene, 50 mg
qod, ASA
vitamin E
Cancer deaths 31 1.11 (0.671.85)
Placebo ASA
vitamin E
28
Western Australia
7
/
1998
1,024 Asbestos workers Retinol, 25,000 IU/d Lung cancer
incidence
4 cases 0.66 (0.192.32) No placebo; not blinded; 56
subjects crossed over (most
from retinol group); no
difference in lung cancer
incidence, but malignant
mesothelioma was
significantly lower in the
retinol group compared
with the -carotene group
-carotene, 30 mg/d 6 cases
Retinol, 25,000 IU/d Mesothelioma
incidence
3 cases 0.24 (0.070.86)
Retinol, 25,000 IU/d All-cause
mortality
21 cases 0.56 (0.330.95)
*ASA acetylsalicylic acid; with or without; qod every other day.
Investigators in the Womens Health Study employed a Cox proportional hazard model to calculate RR for lung cancer associated with -carotene vs placebo
and reported that the results were not statistically significant. However, they did not report the actual values obtained for RR and 95% CI. The values given
here were calculated by the authors of this article using the Cochran-Mantel-Haenszel
2
using the number of lung cancer cases and total number of patients
in each treatment group. More precise methods would require data not published, such as individual patient data or person-years at risk for each treatment
group.
study in 1995 showed no significant difference in the
incidence of total cancers, lung cancer, or any other
cancer. The study has not been analyzed for an effect
of aspirin on lung cancer incidence.
17
The Womens Health Study
10
randomized nearly
40,000 women who were at least 45 years of age in a
2 2 2 factorial design to measure the effect of
-carotene, aspirin, and vitamin E on cancer and
cardiovascular disease incidence. Thirteen percent
of enrolled women were current smokers. Following
the publication of other studies, the -carotene arm
was closed. After supplementation for 2.1 years with
-carotene and an additional median follow-up of
2.0 years (median total follow-up, 4.1 years), there
was no difference in cancer incidence or death with
respect to -carotene supplementation. There were
30 cases of lung cancer in the -carotene arms and
21 cases in the placebo arms (RR, 1.42; 95% CI, 0.88
to 2.49), a result that is not statistically significant.
The early termination of the study limits the ability
to detect a significant effect of the -carotene sup-
plementation on lung cancer incidence or death
from lung cancer.
A South African study
7
of 1,024 asbestos workers
(92% male) randomly assigned subjects in equal
proportions to either -carotene (30 mg/d) or retinol
(25,000 IU/d) from 1990 to 1995. Twenty-one per-
cent of participants were current smokers, and 52%
were former smokers. After a median intervention
and follow-up time of 4.5 years, no difference in lung
cancer incidence was found (RR, 0.66; 95% CI, 0.19
to 2.32), although the small number of lung cancers
(10) severely limited the power of this statement.
The incidence of malignant mesothelioma was sta-
tistically significantly lower in the retinol group
compared to the -carotene group (RR, 0.24; 95%
CI, 0.07 to 0.86).
No ongoing large-scale RCTs for the primary
prevention of lung cancer in high-risk individuals
were identified. The recently begun Selenium and
Vitamin E Cancer Prevention Trial
18
will examine
the effect of selenium and vitamin E on lung cancer
incidence as a secondary end point in low-risk men.
Secondary Chemoprevention Interventions
Persons with identifiable precursor lesions are
appropriate for possible secondary prevention inter-
ventions. In lung cancer, there appears to be an
orderly histologic progression of changes in the
epithelium of the large airways from normal epithe-
lium to metaplasia, increasing degrees of dysplasia,
carcinoma in situ, and invasive squamous cell can-
cer.
19
These histologic changes are more pro-
nounced in the proximal than in the distal airways
and are more common with increased cigarette
usage, reaching their maximum in the airways unin-
volved with lung cancer in patients who died of lung
cancer.
20,21
In addition, the frequency of carcinoma
in situ increases with the number of cigarettes
smoked per day and is highest in patients who died
of lung cancer.
22,23
Bronchial epithelial histologic
changes can be identified in directed or blind endo-
bronchial biopsy samples and as atypia in sputum
samples. In the periphery of the lung, atypical
adenomatous hyperplasia has been proposed as a
precursor lesion of invasive adenocarcinoma,
19
the
most common histologic subtype of non-small cell
lung cancer (NSCLC).
24
Angiogenic squamous hy-
perplasia is a recognized lesion that also may be a
malignant precursor.
25
Neither of these latter mor-
phologic lesions is frequently identified in nonsurgi-
cal biopsy specimens.
We identified no RCTs of secondary prevention in
subjects who had been selected for the presence of
precursor lesions that used lung cancer incidence as
an end point. At least five small randomized studies
of secondary chemoprevention have been performed
with retinoids, -carotene plus retinol, and the com-
bination of vitamin B
12
and folate administered for
up to 6 months.
26
However, all of these studies used
either bronchial epithelial metaplasia or sputum
atypia as a surrogate end point. None used lung
cancer incidence as a primary or secondary end
point.
Tertiary Chemoprevention Interventions
Patients with a previous incidence of cancer of the
upper aerodigestive tract (ie, lung, head and neck,
and esophagus) have the highest rate of development
of lung cancer of any population. Patients who have
undergone resection for early-stage NSCLC develop
a second primary lung cancer at a rate of approxi-
mately 2% per subject per year.
27
This population
has been used to test chemoprevention strategies in
five RCTs
2832
that had primary end points of second
primary cancer (of the lung) [Table 2]. Each of these
studies was limited by the use of a clinicopathologic
definition of the term second primary cancer. Mo-
lecular genetic analysis can assess more definitively
the clonality of metachronous tumors of the upper
aerodigestive tract.
33
Two of the smallest studies suggest beneficial
effects for retinoids (ie, oral isotretinoin)
28
and reti-
nyl palmitate
29
in reducing second primary cancer
incidence. The study by Hong et al
28
was designed
with a primary end point of primary tumor recur-
rence, and the observation of a reduction in second
primary tumors was a secondary analysis. The inter-
pretation of these studies from the perspective of
lung cancer prevention is further complicated by the
fact that a fraction of second primary tumors were of
the head and neck. Furthermore, the results of
subsequent large trials that were undertaken to
confirm these results do not support this effect for
isotretinoin
30
or retinyl palmitate and/or N-acetylcys-
teine.
31
A trial of -carotene (50 mg/d) vs placebo in
patients with prior cancer of the head and neck was
stopped early after the results of the ATBC and
CARET studies were available.
32
Similar to the
primary prevention trials of -carotene, this trial
showed a nonsignificant trend toward increased lung
cancer incidence in patients who were treated with
-carotene.
A randomized trial of selenium in patients who
had undergone resection for early-stage NSCLC was
recently begun in North America.
34
The basis for this
study, as well as that for the ongoing Selenium and
Vitamin E Cancer Prevention Trial
18
mentioned
earlier, includes observational studies showing lower
serum levels of selenium in lung cancer patients
compared with control subjects, antioxidant proper-
ties of selenium as essential cofactor for glutathione
peroxidase, decreased cancer incidence following
selenium supplementation in Linxian, China,
35
and
decreased lung cancer incidence in an RCT of
selenium supplementation to prevent skin cancer.
36
Tin miners in the Yunnan province of China have a
high incidence of lung cancer (ie, 1% per year). A
Table 2RCTs Examining the Effect of Tertiary Chemoprevention Interventions on Lung Cancer Incidence
(Second Primary Tumor)*
Study/Year No. Population Intervention End Point Result
RR (95% CI)
or p Value Comments
MD Anderson
Cancer Center
28
/
1990
103 Head and neck
cancer
Isotretinoin, 50100 mg/m
2
/
d 12 mo
Second primary
tumor
2 cases 0.0005 Small study, but
suggested beneficial
effects from oral
isotretinoin
Placebo 12 cases
Istituto Nazionale
Tumori
29
/1993
307 Stage I NSCLC
status post-
resection
Retinol palmitate, 300,000
IU/d 12 mo
Any second
primary cancer
18 cases 0.09 Small study, but
suggested retinyl
palmitate reduced
second primary cancer
incidence
Placebo 29 cases
IU/d 12 mo
UADTC second
primary tumor
13 cases 0.045
Placebo 25 cases
IU/d 12 mo
Lung cancer
incidence
11 p value not
reported
Placebo 21
Lippman et al
30
/
2001
1,166 Stage I NSCLC
status post-
resection
Isotretinoin, 30 mg/d 3 yr All second
primary tumor
incidence
4.2%/yr 1.08 (0.781.42) Large study; showed no
effect of isotretinoin
on incidence of second
primary cancers
Placebo 3.9%/yr
EUROSCAN
31
/
2000
2,592 Stage IIII head
and neck
cancer (60%)
and lung
(40%) cancer
Retinyl palmitate, 300,000
IU/d 1 yr, then
150,000 IU/d 1 yr
All second
primary tumor
incidence
54 cases 0.174 Large study; showed no
effect of retinyl
palmitate and/or N-
acetylcysteine on
second primary cancer
incidence
NAC, 600 mg/d 2 yr 61 cases
Retinyl palmitate (as above)
plus NAC (as above)
61
Placebo 32
Yale Study
32
/2001 264 Early-stage head
and neck
cancer
-carotene, 50 mg/d Lung cancer
incidence
13 cases 1.44 (0.623.39) Underpowered; stopped
early due to results
from other studies, but
showed trend toward
increase in lung cancer
incidence in patients
treated with -
carotene
Placebo 9
-carotene, 50 mg/d Head and neck
cancer,
recurrent or
second primary
33 0.9 (0.561.45)
Placebo 34
-carotene, 50 mg/d Second primary
tumor
17 1.20 (0.592.45)
Placebo 14 cases
*NAC N-acetylcysteine; UADTC urinary, airways, and digestive tract cancer.
By log rank test.
small (n 40) blinded feasibility study
37
in this
population demonstrated that selenium supplemen-
tation with malt cakes was well-tolerated, increased
serum and tissue levels of selenium, increased glu-
tathione peroxidase levels, decreased lipid peroxide
levels, and improved DNA repair in response to
ultraviolet or benzo-a-pyrene damage. This popula-
tion continues to be studied.
Smoking Prevention (Primary Smoking Prevention)
and Cessation (Secondary Smoking Prevention)
An overwhelming body of evidence links cigarette
smoking to lung cancer in a causative relationship.
The cessation of smoking results in a slow decline in
the risk of cancer development, but this risk remains
elevated compared to never-smokers 15 years
after smoking cessation.
38
The relative benefit of
smoking cessation appears to be greater for those
persons with shorter smoking histories.
38,39
Thus, the
prevention of smoking initiation among nonsmokers
and the encouragement of smoking cessation among
smokers lead to a decline in lung cancer incidence
and mortality, which recently has been seen in states
instituting aggressive antismoking campaigns. State-
wide comprehensive tobacco control programs,
which typically involve some mix of public education,
print media campaigns, prevention of youth access to
tobacco, restriction of advertising, creation of smoke-
free environments, work site antismoking programs,
health professional training on cessation techniques,
and school-based smoking prevention curricula in-
cluding a life-skills training approach, have had the
most success. A distinguishing feature of these com-
prehensive tobacco control programs is their focus
on changing smoking behavior through strategies
that alter the social environment where smoking
occurs. Successful school-based programs have in-
cluded interventions that teach social reinforcement
(ie, dealing with peer pressure), social norms (ie,
increasing self-esteem), and developmental orienta-
tion (ie, development of decision making and inter-
personal skills).
40
To determine whether community-based or clini-
cian-based smoking prevention interventions are ef-
fective in reducing smoking, we sought to identify
existing systematic reviews and meta-analyses to
summarize an enormous body of literature. The
following two major efforts have covered most of the
available evidence on these topics: the US Public
Health Service Tobacco Use and Dependence Clin-
ical Practice Guideline
41
and the Task Force on
Community Preventive Services.
42
Evidence for the
effectiveness of clinician-based interventions has
been described by the Tobacco Use and Depen-
dence Clinical Practice Guideline panel, staff, and
consortium representatives in two reports, the first
published by Agency for Health Care Research and
Quality in 1996,
43
and the second, an update, by the
US Department of Health and Human Services in
2000.
41,44
The Task Force on Community Preventive
Services focused on interventions intended to
achieve tobacco use prevention and control in the
general population but excluded interventions that
targeted only high-risk individuals (eg, cessation
interventions for smokers with coronary artery dis-
ease, cessation programs conducted entirely in hos-
pital settings, or interventions to reduce exposure to
environmental tobacco smoke in homes with asth-
matic children). This task force reviewed interven-
tions at the health-care system level (eg, provider
education or provider performance feedback), but
not interventions delivered by clinicians to patients
in a clinical setting.
Table 3 identifies, for each topic covered, the most
up-to-date and complete systematic review. We de-
scribe the conclusions reached, the number of stud-
ies cited, and the quality of the supporting evidence
according to the scale used by the Task Force on
Community Preventive Services
42
(ie, strong, suf-
ficient, or insufficient). The systematic reviews, in
turn, identify the individual studies cited and, in the
case of Hopkins et al,
42
provide detailed descriptions
of each study in an appendix.
Discussion
The relative rarity of lung cancer incidence in
chemoprevention studies has limited the number of
agents tested in definitive RCTs due to the cost and
prolonged duration of studies. There are good-
quality clinical studies demonstrating that the admin-
istration of -carotene alone or in combination with
retinol to smokers increases the incidence of, and
deaths from, lung cancer, and one study has shown
that the administration of vitamin E to smokers has
no effect on lung cancer incidence. In two ade-
quately powered studies, no effect on second pri-
mary tumor incidence (primarily but not exclusively
lung cancer) was observed with retinoids (ie, retinyl
palmitate and isotretinoin), and in one study, no
effect was seen with the administration of N-
acetylcysteine. A trial of selenium supplementation
to prevent second lung cancers is ongoing.
There is a great deal of evidence about a wide
variety of clinician-based and community-based ef-
forts at smoking avoidance or cessation. Certain
approaches have been shown to be effective (eg,
mass media public education campaigns, direct re-
strictions on smoking, clinician-based approaches
ranging from brief clinician advice to more in-depth
Table 3Clinician-Based and Community-Based Interventions to Reduce Tobacco Exposure*
Topic
Existing Systematic
Review/Year Summary/Conclusions
School-based interventions Lantz et al
45
/2000 Many experimental and quasi-experimental evaluations suggest that some
programs resulted in a significant short-term reduction in smoking,
delay in initiation, or change in attitudes toward tobacco use. Programs
that used a social influences model tended to be the most effective,
especially when enhanced by an extensive community-based education
program. Long-term effects on deterrence of youth smoking are less
certain. The recent Hutchinson Smoking Prevention Project,
46
a
randomized trial testing the social influences approach, found no long-
term effect of an early school-based intervention (grades 310). Only
one rigorously designed trial
47
has shown an effect: a life-skills
training approach with ongoing school consultation aimed at reducing
tobacco, alcohol, and illegal drug use showed a positive effect on long-
term smoking deterrence.
Community interventions
Education to reduce
ETS in the home
Hopkins et al
42
/2001 Ineffective in reducing infant ETS exposure (1 qualifying study, evidence
is insufficient)
Workplace smoking
cessation programs
Eriksen and Gottlieb
48
/
1998
Effective in increasing tobacco use cessation and reducing consumption
of tobacco products. Inconsistent results (52 studies)
Bibeau et al
49
/1988
Statewide
comprehensive
tobacco control
programs
Wakefield and
Chaloupka
50
/2000
Inconsistent evidence of reducing teenage smoking prevalence
(5 programs)
Mass media/public
education
Campaigns (brief
recurring message)
Hopkins et al
42
/2001 Effective in increasing tobacco use cessation and reducing consumption
of tobacco products (15 studies, strong evidence)
Cessation series Hopkins et al
42
/2001 Inconsistent results and inadequate comparison groups (9 studies,
evidence insufficient)
Advertising restrictions None Noted as in progress by the Task Force on Community Preventive
Services. To be covered in future publication
Youth access restrictions Lantz et al
45
/2000 Can lead to a general reduction in illegal sales of cigarettes to minors;
whether this will translate into reduced and sustained reductions in
youth tobacco use is uncertain
Tobacco excise taxes;
increasing unit price for
tobacco products
Hopkins et al
42
/2001 Effective in reducing tobacco use prevalence and consumption among
both adolescents and young adults (8 studies, strong evidence).
Increases tobacco use cessation and reduces consumption (17 studies,
strong evidence)
Direct restrictions on
smoking
Hopkins et al
42
/2001 Effective in reducing exposure to ETS in workplace (9 studies, strong
evidence)
Eriksen and Gottlieb
48
/
1998
Effective in reducing cigarette consumption at work (9 studies);
ineffective at increasing tobacco use cessation (14 studies)
Clinician-based approaches Fiore et al
44
/2000
Identifying tobacco
users
Implementing clinic systems designed to increase the assessment and
documentation of tobacco use status markedly increases the rate at
which clinicians intervene with their patients who smoke (9 studies,
strong evidence); results in higher rates of smoking cessation
(3 studies, sufficient evidence)
Assessment of
willingness to quit
Little consistent evidence that motivation to quit is useful for treatment
matching. Tailored interventions based on specialized assessment (eg,
stages of change) do not consistently produce higher long-term quit
rates than nontailored interventions of equal intensity (sufficient
evidence)
Advice to quit smoking Brief physician advice significantly increases long-term smoking
abstinence rates (7 studies, strong evidence)
Intensity of clinical
interventions
Clinician-to-smoker advice delivered on four or more occasions is
especially effective in increasing abstinence rates. Greater session
length for clinician advice to quit was associated with greater
abstinence rates from 3 to 90 min (35 studies, strong evidence)
sessions, and life-skills training in schools). Some
approaches have intermediate or short-term effec-
tiveness (ie, youth access restrictions, school-based
interventions), and others have been shown to be
ineffective (ie, acupuncture and provider education)
or have been insufficiently studied (ie, provider
feedback). Thus, although there has been little
progress in chemoprevention, several clinician-based
and community-based interventions show promise
for reducing lung cancer incidence through smoking
avoidance and prevention.
References
Table 3Continued*
Topic
Existing Systematic
Review/Year Summary/Conclusions
Type of clinician Treatment delivered by a variety of clinician types increases abstinence
rates (29 studies, strong evidence). Treatments delivered by multiple
types of clinicians are more effective than interventions delivered by a
single type of clinician (37 studies, sufficient evidence)
Format of psychosocial
treatments
Proactive telephone counseling, group counseling, and individual
counseling formats are effective in increasing abstinence rates (58
studies, strong evidence). Use of multiple formats increases abstinence
rates (54 studies, strong evidence)
Type of counseling and
behavioral therapies
Three types of counseling and behavioral therapies result in higher
abstinence rates: (1) practical counseling (problem solving skills/skills
training); (2) social support as part of treatment; and (3) helping
smokers obtain social support outside of treatment (62 studies,
sufficient evidence)
Aversive smoking interventions (rapid smoking, rapid puffing) increase
abstinence rates (38 studiessufficient evidence)
Acupuncture No difference in abstinence rates (5 studiesinsufficient evidence)
Pharmacotherapy Pharmacotherapy is effective in smoking cessation: bupropion SR
(2 studies, strong evidence); nicotine gum (13 studies, strong
evidence); nicotine inhaler (4 studies, strong evidence); nicotine nasal
spray (3 studies, strong evidence); nicotine patch (27 studies, strong
evidence); clonidine (5 studies); nortriptyline (2 studies, sufficient
evidence); combination treatment (3 studies, sufficient evidence)
Antidepressants other than bupropion SR and nortriptyline (no studies,
insufficient evidence)
Anxiolytics/benzodiazepines/beta-blockers (few studies, insufficient
evidence)
Silver acetate (2 studies, insufficient evidence)
Mecamylamine (2 studies, insufficient evidence)
Reduces the likelihood of relapse (multiple studies, strong evidence)
Health-care system level
interventions
Hopkins et al
42
/2001
Provider reminder
systems
Increase provider delivery of advice to quit to tobacco-using patients
Provider education Few studies evaluated effect on patient tobacco use cessation;
inconsistent results on provider delivery of advice to quit (16 studies,
insufficient evidence)
Reminder plus
education
(multicomponent
interventions)
Increase both provider delivery of advice to quit and patient tobacco use
cessation (20 studies, strong evidence)
Provider feedback No studies assessed effect on provider delivery of advice to quit or
patient tobacco use cessation (3 studies, insufficient evidence)
Reducing patient out-of-
pocket cost for
effective cessation
therapies
Increases both use of effective therapy and patient tobacco use cessation
Patient telephone
support
Increases tobacco use cessation when implemented with other
interventions (32 studies, strong evidence)
*Includes interventions aimed at reducing environmental tobacco smoke exposure, decreasing tobacco use initiation, and increasing tobacco use
cessation. ETS environmental tobacco smoke.
2 Kelloff GJ, Sigman CC, Greenwald P. Cancer chemopreven-
tion: progress and promise. Eur J Cancer 1999; 35:17551762
3 The ATBC Cancer Prevention Study Group. The alpha-
tocopherol, beta-carotene lung cancer prevention study: de-
sign, methods, participant characteristics, and compliance.
Ann Epidemiol 1994; 4:110
4 The Alpha-Tocopherol, Beta Carotene Cancer Prevention
Study Group. The effect of vitamin E and beta carotene on
the incidence of lung cancer and other cancers in male
smokers. N Engl J Med 1994; 330:10291035
5 Albanes D, Heinonen OP, Taylor PR, et al. Alpha-
Tocopherol, and beta-carotene supplements and lung cancer
incidence in the Alpha-Tocopherol, Beta-Carotene Cancer
Prevention Study: effects of base-line characteristics and
study compliance. J Natl Cancer Inst 1996; 88:15601570
6 Cook NR, Le IM, Manson JE, et al. Effects of beta-carotene
supplementation on cancer incidence by baseline character-
istics in the Physicians Health Study (United States). Cancer
Causes Control 2000; 11:617626
7 de Klerk NH, Musk AW, Ambrosini GL, et al. Vitamin A, and
cancer prevention: II. Comparison of the effects of retinol
and beta-carotene. Int J Cancer 1998; 75:362367
a combination of beta carotene and vitamin A on lung cancer
and cardiovascular disease. N Engl J Med 1996; 334:1150
1155
9 Omenn GS, Goodman GE, Thornquist MD, et al. Risk
factors for lung cancer and for intervention effects in
CARET, the Beta-Carotene and Retinol Efficacy Trial. J Natl
Cancer Inst 1996; 88:15501559
10 Lee IM, Cook NR, Manson JE, et al. Beta-carotene supple-
mentation and incidence of cancer and cardiovascular dis-
ease: the Womens Health Study. J Natl Cancer Inst 1999;
91:21022106
11 Peto R, Doll R, Buckley JD, et al. Can dietary beta-carotene
materially reduce human cancer rates? Nature 1981; 290:
201208
12 Steinmetz KA, Potter JD. Vegetables, fruit, and cancer: II.
Mechanisms. Cancer Causes Control 1991; 2:427442
13 Omenn GS. What accounts for the association of vegetables
and fruits with lower incidence of cancers and coronary heart
disease? Ann Epidemiol 1995; 5:333335
14 Goodman GE, Omenn GS, Thornquist MD, et al. The
Carotene and Retinol Efficacy Trial (CARET) to prevent lung
cancer in high-risk populations: pilot study with cigarette
smokers. Cancer Epidemiol Biomarkers Prev 1993; 2:389
396
15 Patrick L. Beta-carotene: the controversy continues. Altern
Med Rev 2000; 5:530545
16 Lowe GM, Booth LA, Young AJ, et al. Lycopene and
beta-carotene protect against oxidative damage in HT29 cells
at low concentrations but rapidly lose this capacity at higher
doses. Free Radic Res 1999; 30:141151
17 Sturmer T, Glynn RJ, Lee IM, et al. Aspirin use and
colorectal cancer: post-trial follow-up data from the Physi-
cians Health Study. Ann Intern Med 1998; 128:713720
18 Klein EA, Thompson IM, Lippman SM, et al. SELECT the
next prostate cancer prevention trial: Selenium and Vitamin E
Cancer Prevention Trial. J Urol 2001; 166:13111315
19 Kerr KM. Pulmonary preinvasive neoplasia. J Clin Pathol
2001; 54:257271
20 Auerbach O, Gere JB, Forman JB, et al. Changes in the
bronchial epithelium in relation to smoking and cancer of the
lung. N Engl J Med 1957; 256:97104
21 Auerbach O, Stout AP, Hammond EC, et al. Changes in
bronchial epithelium in relation to cigarette smoking and in
relation to lung cancer. N Engl J Med 1961; 265:253267
22 Jay SJ, Wehr K, Nicholson DP, et al. Diagnostic sensitivity
and specificity of pulmonary cytology: comparison of tech-
niques used in conjunction with flexible fiberoptic bronchos-
copy. Acta Cytol 1980; 24:304312
23 Mao L, Lee JS, Kurie JM, et al. Clonal genetic alterations in
the lungs of current and former smokers. J Natl Cancer Inst
1997; 89:857862
24 Devesa SS, Shaw GL, Blot WJ. Changing patterns of lung
cancer incidence by histological type. Cancer Epidemiol
25 Keith RL, Miller YE, Gemmill RM, et al. Angiogenic squa-
mous dysplasia in bronchi of individuals at high risk for lung
cancer. Clin Cancer Res 2000; 6:16161625
26 Khuri FR, Lippman SM. Lung cancer chemoprevention.
Semin Surg Oncol 2000; 18:100105
27 Rosa UW, Prolla JC, Gastal Ed. Cytology in diagnosis of
cancer affecting the lung: results in 1,000 consecutive pa-
tients. Chest 1973; 63:203207
28 Hong WK, Lippman SM, Itri LM, et al. Prevention of second
primary tumors with isotretinoin in squamous-cell carcinoma
of the head and neck. N Engl J Med 1990; 323:795801
29 Pastorino U, Infante M, Maioli M, et al. Adjuvant treatment
of stage I lung cancer with high-dose vitamin A. J Clin Oncol
1993; 11:12161222
30 Lippman SM, Lee JJ, Karp DD, et al. Randomized phase III
intergroup trial of isotretinoin to prevent second primary
tumors in stage I non-small-cell lung cancer. J Natl Cancer
Inst 2001; 93:605618
31 van Zandwijk N, Dalesio O, Pastorino U, et al. EUROSCAN,
a randomized trial of vitamin A and N-acetylcysteine in
patients with head and neck cancer or lung cancer: for the
European Organization for Research and Treatment of Can-
cer Head and Neck and Lung Cancer Cooperative Groups.
32 Mayne ST, Cartmel B, Baum M, et al. Randomized trial of
supplemental beta-carotene to prevent second head and neck
cancer. Cancer Res 2001; 61:14571463
33 Yang HK, Linnoila RI, Conrad NK, et al. TP53, and RAS
mutations in metachronous tumors from patients with cancer
of the upper aerodigestive tract. Int J Cancer 1995; 64:229
233
34 National Cancer Institute. Phase III randomized chemopre-
vention study of selenium in participant swith previously
resected stage I non-small cell lung cancer (summary last
modified September 2001); protocol IDs E-5597, NCI-P00-
0176. Available at: http://www.cancer.gov/cancer_informa-
tion/pdq/doc_cit.aspx?citationid200_19869. Accessed De-
cember 16, 2002
35 Blot WJ, Li JY, Taylor PR, et al. Nutrition intervention trials
in Linxian, China: supplementation with specific vitamin/
mineral combinations, cancer incidence, and disease-specific
mortality in the general population. J Natl Cancer Inst 1993;
85:14831492
36 Clark LC, Combs GF Jr, Turnbull BW, et al. Effects of
selenium supplementation for cancer prevention in patients
with carcinoma of the skin: a randomized controlled trial;
Nutritional Prevention of Cancer Study Group. JAMA 1996;
276:19571963
37 Hilton LW. The Robert Tiffany Lectureship: vital signs at the
millennium; becoming more than we are. Cancer Nurs 1999;
22:616
38 Halpern MT, Gillespie BW, Warner KE. Patterns of absolute
risk of lung cancer mortality in former smokers. J Natl Cancer
Inst 1993; 85:457464
39 Sobue T, Yamaguchi N, Suzuki T, et al. Lung cancer inci-
dence rate for male ex-smokers according to age at cessation
of smoking. Jpn J Cancer Res 1993; 84:601607
40 Bruvold WH. A meta-analysis of adolescent smoking preven-
tion programs. Am J Public Health 1993; 83:872880
41 The Tobacco Use and Dependence Clinical Practice Guide-
line Panel, Staff, and Consortium Representatives. A clinical
practice guideline for treating tobacco use and dependence: a
US Public Health Service report. JAMA 2000; 283:3244
3254
42 Hopkins DP, Briss PA, Ricard CJ, et al. Reviews of evidence
regarding interventions to reduce tobacco use and exposure
to environmental tobacco smoke. Am J Prev Med 2001;
20:1666
43 Fiore MC, Bailey WC, Cohen SJ, et al. Smoking cessation:
clinical practice guideline No. 18. Agency for Health Care
Policy and Research Publication No. 96-0692. Rockville, MD:
US Department of Health and Human Services, Public
Health Service, Agency for Health Care Policy and Research,
1996
44 Fiore MC, Bailey WC, Cohen SJ, et al. Treating tobacco use
and dependence. Agency for Health Care Policy and Research
Publication No. 00-0032. Rockville, MD: US Department of
Health and Human Services, Public Health Service, 2000.
45 Lantz PM, Jacobson PD, Warner KE, et al. Investing in youth
tobacco control: a review of smoking prevention and control
strategies. Tob Control 2000; 9:4763
46 Peterson AV Jr, Kealey KA, Mann SL, et al. Hutchinson
Smoking Prevention Project: long-term randomized trial in
school-based tobacco use preventionresults on smoking.
47 Botvin GJ, Baker E, Dusenbury L, et al. Long-term follow-up
results of a randomized drug abuse prevention trial in a white
middle class population. JAMA 1995; 273:11061112
48 Eriksen MP, Gottlieb NH. A review of the health impact of
smoking control at the workplace. Am J Health Promot 1998;
13:83104
49 Bibeau DL, Mullen KD, McLeroy KR, et al. Evaluations of
workplace smoking cessation programs: a critique. Am J Prev
Med 1988; 4:8795
50 Wakefield M, Chaloupka F. Effectiveness of comprehensive
tobacco control programmes in reducing teenage smoking in
the USA. Tob Control 2000; 9:177186
2003;123;50-59 Chest
Michael J. Kelley and Douglas C. McCrory
Prevention of Lung Cancer: Summary of Published Evidence
& Services
Updated Information
References
#BIBL
free at:
Reprints
2003;123;60-71 Chest
Konstantin H. Dragnev, Diane Stover and Ethan Dmitrovsky
Lung Cancer Prevention: The Guidelines
ISSN: 0012-3692.
Lung Cancer Prevention*
The Guidelines
Konstantin H. Dragnev, MD; Diane Stover, MD, FCCP; and
Ethan Dmitrovsky, MD
Lung carcinogenesis is a chronic and multi-step process resulting in malignant lung tumors. This
progression from normal to neoplastic pulmonary cells or tissues could be arrested or reversed
through pharmacologic treatments, which are known as cancer chemoprevention. These thera-
peutic interventions should reduce or avoid the clinical consequences of lung cancer by treating
early neoplastic lesions before the development of clinically evident signs or symptoms of
malignancy. Preclinical, clinical, and epidemiologic findings relating to different classes of
candidate chemopreventive agents provide strong support for lung cancer prevention as an
attractive therapeutic strategy. Smoking prevention and smoking cessation represent an essential
approach to reduce the societal impact of tobacco carcinogenesis. However, even if all the goals
of the national antismoking efforts were met, there still would be a large population of former
smokers who would be at increased risk for lung cancers. Lung cancer also can occur in those
persons who never have smoked. This article focuses on what is now known about pharmacologic
strategies for lung cancer prevention. Randomized clinical trials using -carotene, retinol,
isotretinoin or N-acetyl-cysteine did not show benefit for primary and tertiary lung cancer
prevention. There is also evidence that the use of -carotene and isotretinoin for lung cancer
chemoprevention in high-risk individuals may increase the risk for lung cancer, especially in
individuals who continue to smoke. There is a need for relevant in vitro models to identify
pathways that activate chemopreventive effects in the lung. An improved understanding of cancer
prevention mechanisms should aid in the design of clinical trials and in the validation of candidate
chemopreventive targets as well as the discovery of new targets. Until such studies are completed,
no agent or combination of agents should be used for lung cancer prevention outside of a clinical
trial. (CHEST 2003; 123:60S71S)
Key words: chemoprevention; epidermal growth factor receptor; G1 cyclin; lung cancer; retinoids
Abbreviations: COX cyclooxygenase; EGFR epidermal growth factor receptor; 4HPR fenretinide; LOH loss
of heterozygosity; NNK N-nitrosamine-4-(methylnitrosamino)-1-(3 pyridyl)-1-butanone; RA all-trans retinoic acid;
Rb retinoblastoma gene product; RXR retinoid X receptor
L
ung carcinogenesis is a chronic process involving
multiple genetic, cellular, and tissue alterations.
This results from mutagenic damage to growth-
regulating genes and their products that ultimately
leads to the development of invasive or metastatic
cancer.
1
The transformation steps from normal
through preneoplasia to overt malignancy occur as a
consequence of the following: (1) initiation, in which
DNA damage occurs; (2) promotion, in which ge-
netic and epigenetic changes confer additional
genomic damage; and (3) progression to locally
invasive or metastatic disease. Carcinogen exposure
forms fields of altered cells long before invasive
malignant disease is detected clinically, as was first
hypothesized by Slaughter and coworkers in 1953.
2
The concept of field cancerization provides a basis
for understanding the clonality of preneoplastic cells.
Some of these genetically altered cells acquire a
malignant phenotype, while others do not.
There are many interventions that might be con-
sidered as strategies for reducing lung-specific can-
cer risks including smoking prevention and cessation,
lifestyle as well as dietary or nutritional changes,
*From the Norris Cotton Cancer Center and Dartmouth Medical
School (Drs. Dragnev and Dmitrovsky), Lebanon, NH; and the
Pulmonary Section (Dr. Stover), Department of Medicine, Me-
morial Sloan-Kettering Cancer Center, New York, NY.
This work was supported in part by grants RO1-CA87546 (ED)
and RO1-CA62275 (ED) from the National Institutes of Health,
and by grant RPG-90-019-10-DDC (ED) from the American
Cancer Society. Dr. Dragnev was supported in part by the
American Society of Clinical Oncology (ASCO) Young Investiga-
tor Award. This work was supported in part by the Oracle Giving
Fund and by the Samuel Waxman Cancer Research Foundation.
Correspondence to: Konstantin H. Dragnev, MD, Hematology/
Oncology Section, Department of Medicine, Dartmouth-Hitch-
cock Medical Center, Lebanon, NH 03756; e-mail: dragnev@
dartmouth.edu
effective screening of identified high-risk individuals,
among others. Of these strategies, only smoking
prevention and cessation has been shown to reduce
lung cancer risk. Although the focus of this article is
on the chemoprevention of lung cancer, it is primary
prevention (ie, smoking prevention) that should be a
major focus within our society including local com-
munities, schools from kindergarten through college,
and among persons in the medical profession. Strat-
egies that have been the most successful in prevent-
ing children from starting to smoke include all-grade
inclusive school programs that emphasize a life skills
training approach, the use of brief recurring antismok-
ing messages that point out the positive aspects of being
nicotine-free, and the enforcement of high excise taxes
on tobacco products. For current smokers, there is
strong evidence that brief recurring physician advice
significantly increases long-term smoking abstinence
rates. Clinician-based approaches should always in-
clude the routine identification of tobacco users, which
in turn increases the rate of clinician intervention with
patients who smoke.
Recommendations
1. For all individuals, smoking prevention should
be strongly encouraged to decrease the risk of
lung cancer. Level of evidence, good; benefit,
substantial; grade of recommendation, A
2. For all individuals, school-based and community-
based interventions that are aimed at reducing
tobacco exposure should be recommended, in-
cluding a life skills training approach that is
aimed at reducing tobacco, alcohol, and illicit
drug use, campaigns with brief recurring anti-
smoking messages, high tobacco excise taxes, and
restrictions on smoking in the workplace. Level of
evidence, good; benefit, substantial; grade of rec-
ommendation, A
3. Smokers should be identified as smoking ces-
sation reduces the risk of lung cancer. Level of
evidence, good; benefit, substantial; grade of
recommendation, A
4. Current smokers should be advised to quit
smoking, and, when appropriate, clinicians
should prescribe and monitor pharmacother-
apy. Individuals who smoke and want to quit
also should have access to psychosocial treat-
ment and behavioral modification therapies as
indicated. There is sufficient-to-strong evi-
dence that indicates these practices will help to
increase long-term smoking abstinence rates.
Level of evidence, good; benefit, substantial;
grade of recommendation, A
Lung Cancer Chemoprevention Agents
Complex cytogenetic or molecular genetic abnor-
malities can occur in lung cancers or in nonmalignant
lung tissues that are isolated from patients with lung
cancer.
37
It has been hypothesized
7
that multiple
changes are required for the progression of lung
carcinogenesis. The gain and loss of specific growth
regulatory species or DNA loci often are found in
lung carcinogenesis.
812
Alterations involve domi-
nant genetic events that occur through the activation
of oncogenes and recessive genetic changes that
occur through the deletion or inactivation of tumor
suppressor genes.
13
Some studies
12,14
reveal that loss
of heterozygosity (LOH) on chromosomes 3p, 9p, or
17p often is detected not only in lung cancers, but
also in preneoplastic lesions and even in adjacent
histologically normal lung tissues. Bronchial biopsy
specimens from former smokers also exhibit genetic
damage. LOH at chromosomes 9p and 17p is re-
ported in former smokers. LOH at chromosome 3p
is also frequent but might be a reversible alteration.
4
These oncogenic changes could be required for the
development or maintenance of lung cancer and
might represent attractive targets for lung cancer
prevention. Conceivably, these would be useful as
surrogate markers with which to monitor clinical
response to lung cancer chemoprevention agents.
Squamous cell lung carcinomas often arise in
association with areas of metaplasia, dysplasia, or
carcinoma in situ. Because of this association, Sac-
comanno and coworkers
3
proposed in 1974 a model
that emphasized the progressive changes of squa-
mous cell metaplasia (ie, hyperplasia), metaplasia
with atypia (ie, dysplasia) that was mild, moderate, or
marked, or carcinoma in situ that preceded the
development of invasive squamous cell carcinoma. It
is now possible to highlight specific genetic or
chromosomal alterations and changes in gene ex-
pression that are associated with these preneoplastic
changes in the lung, as is summarized in Figure 1.
Related models exist for the development of lung
adenocarcinoma. It is not yet known which individ-
ual or cassette of carcinogenic changes is rate-
limiting in the maintenance or progression of pre-
neoplastic lesions in the lung. These could be distinct
for each carcinogenic agent or lung cancer his-
topathologic subtype.
The chronic and multistep nature of lung carcino-
genesis makes attractive a chemopreventive ap-
proach that would arrest or reverse one or more of
these carcinogenic changes. The cancer chemopre-
vention concept that was first proposed by Sporn et al
15
stresses interventions at early stages of carcinogene-
sis, even before malignancies become clinically evi-
dent. If chemoprevention were effective in the lung,
this would avoid many clinical consequences of lung
cancer and would reduce the need for the treatment
of disseminated lung cancers that are often resistant
to combination chemotherapy. Clinical validation for
the cancer prevention concept was provided by a
randomized trial using the selective estrogen recep-
tor modulator (SERM) tamoxifen in women who are
at high risk for breast cancer development.
16
In
women who were randomized to receive tamoxifen
as compared to control subjects, there was a highly
statistically significant reduction in the risk of inva-
sive and noninvasive breast cancers.
16
These findings
provide an impetus to identify other pharmacologic
agents that would reduce the risk of primary or
second lung cancers.
Despite the fact that the best way to prevent most
lung cancers is not to smoke, there remains an
alarming number of current smokers and former
smokers worldwide. Chemoprevention offers an ap-
pealing therapeutic approach assuming that certain
nontoxic agents can be identified. An improved
understanding of the basic biology of lung carcino-
genesis remains essential for the development of
effective lung cancer chemoprevention. The defined
steps in lung carcinogenesis could be targeted by
antiproliferative, differentiation-inducing, proapo-
ptotic, or antiangiogenic agents.
17
This article fo-
cuses on those chemopreventive agents that exert
their pharmacologic actions through specific path-
ways that can be targeted in the lung.
Pharmacologic strategies for chemoprevention
should target multiple steps in carcinogenesis.
18
Ef-
fective chemopreventive agents could block DNA
damage occurring as an initiating step in carcinogen-
esis or could suppress the growth or progression of
preneoplastic cells that already have acquired
genomic DNA damage. Agents also might act at the
promotion or progression steps of carcinogenesis.
Telomerase activation, cell-stromal interactions, as
well as new blood vessel formation (ie, neoangiogen-
esis) likely have important roles in the development
of invasive lung cancers. An empirical approach to
cancer chemoprevention has been replaced by tar-
geted therapeutic strategies that emphasize mecha-
nistic actions of candidate preventive agents in clin-
ical chemopreventive trials. These strategies build on
the basic scientific understanding of the relevant
pathways that are involved in cancer chemopreven-
tion. Many candidate lung cancer therapeutic or
preventive agents exist with diverse structures and
mechanisms of action.
1719
Several of these agents
are already available for testing in lung cancer
prevention trials. A partial list of these agents ap-
pears in Table 1. In addition to these agents, other
small-molecule agonists and antagonists are under-
going preclinical testing and will soon be available
for testing in lung cancer chemoprevention trials.
In vitro carcinogen-induced or engineered animal
models should be useful to validate whether a ther-
apeutic target is required for the maintenance or
progression of a preneoplastic or neoplastic lung
lesion.
20
The presence of the differential overexpres-
sion of growth regulatory species in preneoplastic or
malignant lung tissues vs normal lung tissues pro-
vides one basis for targeting these species in lung
carcinogenesis.
2124
Clinical pharmacologic data al-
Figure 1. Steps involved in squamous cell lung carcinogenesis.
This figure depicts the multiple steps involved in the develop-
ment of squamous cell cancers of the lung. Focal atypia can arise
in preneoplastic lesions. Some preneoplastic lesions are viewed as
reversible (solid arrow), while others may not be reversible
(arrow without a solid line). The outlined box has highlighted
preneoplastic targets for lung cancer chemoprevention. Also
displayed are some representative alterations (genetic, chromo-
somal, or gene expression) that are frequent in bronchial preneo-
plasia. The sites where these occur during lung carcinogenesis are
shown by their relative positions. The text provides a detailed
discussion of these alterations.
Table 1A Partial List of Candidate Lung Cancer
Chemoprevention Agents*
COX-2 inhibitors
EGFR inhibitors
Rexinoids
Lipoxygenase inhibitors
Angiogenesis inhibitors
Cell cycle inhibitors
Demethylation agents
mTOR inhibitors
Prostacyclin analogs
Triterpenoids
PPAR agonists
Ras inhibitors
Budesonide
Proteasome inhibitors
PKC inhibitors
Histone deacetylase inhibitors
Epigallocatechin gallate
Selenium
*PKC protein kinase C
ready exist for some chemopreventive agents in the
advanced-disease setting. From these clinical data,
therapeutic activity and toxicity profiles would al-
ready be available. Findings could be extrapolated to
the use of the same agent in lung cancer prevention.
An attractive target for clinical chemoprevention
would have pharmacologic data from preclinical
testing in in vitro and animal models. Potential
targets include receptor and nonreceptor kinases,
growth factor receptors or their ligands, and en-
zymes (especially as part of signal transduction path-
ways), among other targets. For clinical chemopre-
vention in the lung or at other sites, several desirable
features exist for the selection of an optimal agent for
study. These include having a validated target, based
on appropriate preclinical and clinical pharmacologic
data for this target, completed phase I safety and
dose-response data, objective responses in phase I
and II trials, and convenient schedules and routes of
drug administration. In the next sections, several
candidate lung cancer chemopreventive agents will
be discussed.
Retinoids
Antiproliferative, differentiation-inducing, as well
as proapoptotic agents could target carcinogenic
changes. The retinoids are a class of prevention
agents that could exert many of these potential
clinical chemopreventive effects. Retinoids are nat-
ural and synthetic derivatives of vitamin A that have
diverse chemical structures, pharmacologic proper-
ties, nuclear receptor affinities, and associated toxic-
ity profiles.
17,18
A strong rationale for use of the
retinoids in lung cancer prevention stems from
results obtained from experimental animal models,
epidemiologic studies, and clinical trials.
25
Wolbach
and Howe
26
first identified in 1925 vitamin A-
dependent pathways that are required for epithelial
cell homeostasis. They discovered that vitamin A
deficiency in rodents caused squamous metaplasia in
the trachea as well as at other epithelial sites. This
was reversed by the correction of the vitamin A
deficiency. These metaplastic changes were similar
to those that arose in smokers, implicating a role for
vitamin-A dependent signals in suppressing lung
carcinogenesis. Further evidence for an association
between vitamin A and cancer incidence came from
epidemiologic data establishing an inverse relation-
ship between vitamin A levels and the incidence of
cancer at specific epithelial sites.
25
These and other
findings provided a basis for the use of retinoids in
lung cancer prevention.
Additional support for a retinoid role in cancer
chemoprevention derived from clinical trials that
were conducted using retinoids that resulted in the
successful treatment of certain premalignant condi-
tions such as oral leukoplakia,
27
cervical dysplasia,
28
and xeroderma pigmentosum.
29
Other clinical trials
revealed retinoid activity in reducing some second
primary cancers. These include trials that demon-
strated a reduction in second aerodigestive tract
cancers in patients having prior head and neck
carcinomas,
30
lung carcinomas,
31
or hepatocellular
carcinomas.
32
In contrast to these promising trials, a
randomized Intergroup trial that was conducted with
subjects who had been treated with 13-cis-retinoic
acid following resection of stage I lung cancers did
not show clinical benefit in smokers, although a
reduction in second cancers was observed in subjects
who never had smoked.
33
This indicated the poten-
tial for negative clinical actions when chemopreven-
tive agents are administered in active smokers. Like-
wise, this underscores the need to understand how
retinoids may be clinically useful in chemopreven-
tion for former smokers or never-smokers.
Epidemiologic evidence for an inverse relation-
ship between serum vitamin A or -carotene levels
and specific cancer incidences led to cancer preven-
tion trials using -carotene. Chemopreventive treat-
ment with -carotene had appeal because of its role
as a vitamin A precursor with antioxidant activity
and its reduced clinical toxicity compared to 13-cis-
retinoic acid. Treatment with -carotene also was
hypothesized to restore physiologic carotenoid lev-
els. However, randomized trials using -carotene for
primary lung cancer prevention in high-risk individ-
uals did not result in a reduction of lung cancers and
appeared harmful in this setting, especially when
subjects continued to smoke or to consume etha-
nol.
3436
These trials are discussed in detail in an-
other chapter. Yet, it is important to emphasize here
that some of the lessons learned from these trials
included the differences between physiologic levels
and pharmacologic treatments with chemopreven-
tive agents, the potential for negative interactions in
smokers who are treated with these agents, and the
clinical need to combine lung cancer prevention
agents with smoking cessation.
It is also important to note that nonclassic retin-
oids such as retinoid X receptor (RXR) agonists (ie,
rexinoids) have yet to be studied in lung cancer
prevention. A rationale for their use comes from a
study reporting clinical activity of bexarotene when
used with combination chemotherapy for the treat-
ment of patients with advanced-stage lung cancer.
37
Rexinoids and retinoids activate distinct nuclear
receptors but can engage a similar chemopreventive
pathway in the lung, as discussed below.
38,39
Other
nonclassic retinoids exist that promote apoptosis,
inhibit AP1, or signal other biological effects, as
reviewed.
17,18
These pharmacologic agents also could
become available for testing in lung cancer chemo-
prevention trials.
Recommendations
5. Individuals who are at risk for lung cancer and
were treated with -carotene, retinol, isotreti-
noin, or N-acetyl-cysteine for lung cancer pre-
vention did not experience clinical benefit.
There is also evidence that the use of -
carotene and isotretinoin for lung cancer che-
moprevention in high-risk individuals may in-
crease the risk for lung cancer, especially in
individuals who continue to smoke. These
agents should not be used outside of a clinical
trial for primary, secondary, or tertiary lung
cancer prevention. Level of evidence, good;
benefit, none/negative; grade of recommenda-
tion, D
Mechanisms of Retinoid Actions
The retinoids exert their biological effects through
nuclear retinoid receptors that are associated with
inhibitory corepressors or stimulatory coactiva-
tors.
4042
These ligand-dependent interactions result
in the activation of target genes that signal retinoid
growth-suppressive effects or differentiation effects.
There are three retinoic acid nuclear receptors
(RAR, RAR, and RAR) as well as three retinoid
X nuclear receptors (RXR, RXR, and RXR),
along with multiple isoforms.
43
These share homol-
ogy with other members of the steroid receptor
superfamily of nuclear receptors, which include the
glucocorticoid receptor, vitamin D receptor, and
estrogen receptor, among others. Orphan nuclear
receptors exist, and their physiologic ligands remain
to be discovered. Retinoid receptor-dependent path-
ways have been identified through the use of homol-
ogous recombination in defined cell contexts
4446
and through the use of retinoid receptor-specific
agonists.
47,48
The expression of specific retinoid re-
ceptors is linked to retinoid responses in certain
preneoplastic and malignant diseases. For example,
the induction of RAR expression is associated with
clinical responses in oral leukoplakia.
49
Retinoids bind to the ligand binding domain of
specific retinoid nuclear receptors. These nuclear
receptors also contain DNA-binding domains that
recognize defined DNA response elements in the
genome.
40,41
Following these ligand-receptor and
receptor-DNA interactions, the transcription of di-
rect target genes that signal retinoid biological ef-
fects are activated or repressed. Retinoid nuclear
receptors can undergo heterodimerization or ho-
modimerization during retinoid receptor signal-
ing.
5052
These receptors associate with coregulator
proteins that are known as inhibitory corepressors or
stimulating coactivators.
42
Protein-protein interac-
tions between retinoid receptors and their coregula-
tors provide an important level of regulation to the
retinoid-signaling pathway since these can affect the
basal transcriptional machinery
53
through chromatin
remodeling that is related to changes in the state of
acetylation. Coregulators represent additional phar-
macologic targets for lung cancer prevention.
Other retinoid binding proteins exist, including
the cytosolic retinoic acid binding protein (CRABP)
and the cytosolic retinol binding protein (CRBP).
These regulate the intracellular binding of retinoids
and appear to contribute to retinoid metabolism and
signaling pathways. These cytosolic receptors may
serve as intracellular storage sites for the retinoids or
may facilitate retinoid transport from the cytoplasm
into the nucleus. Alternatively, these proteins could
sequester retinoids in the cytoplasm and perhaps
reduce extracellular stores, as reviewed.
54
Cytosolic
retinoid receptors might contribute to clinical retin-
oid resistance, as in patients with acute promyelo-
cytic leukemia, although pharmacologic mechanisms
also could confer this resistance.
55,56
Pharmacologic agonists and antagonists have been
engineered to affect specific components of the
retinoid signaling pathway. For instance, all-
trans-retinoic acid (RA) is an agonist for the RAR but
not the RXR pathway, while 9-cis-retinoic acid is
bifunctional, activating the RAR and RXR path-
ways.
57,58
Bexarotene is an RXR-selective agonist
that has been approved for clinical use by the US
Food and Drug Administration. Some retinoids tar-
get the AP-1 transcription factor,
59
while others,
such as N-(4-hydroxyphenyl)retinamide (fenretinide
4HPR), act through receptor-independent mecha-
nisms that can generate reactive oxygen species
60
and preferentially signal apoptosis even in cells that
harbor retinoid receptor defects that confer RA
resistance.
61
4HPR triggers apoptosis even in RA-
resistant tumor cells that also have acquired chemo-
therapy resistance.
61
Clinical benefits have been
reported with 4HPR in the prevention of second
breast cancers.
62
Whether 4HPR would have chemo-
preventive effects in the lung remains to be deter-
mined.
To date, lung cancer chemoprevention clinical
trials have emphasized the use of classic retinoids
that activate the RAR pathway. Since RAR repres-
sion is frequent in several epithelial cancers includ-
ing lung cancers, this could affect clinical chemopre-
ventive responses to classic retinoids.
49,63
The
mechanisms that are responsible for RAR repres-
sion are under active study. Preclinical evidence
points to a role for methylation-induced silencing of
this nuclear receptor, among other possible mecha-
nisms. Perhaps demethylation agents that target
RAR sequences could be used in conjunction with
an active retinoid to overcome RAR repression and
to engage clinical chemopreventive effects. Clinical
cancer chemoprevention trials in the lung should
consider the use of retinoids that do not activate the
classic retinoid signaling pathway, thus bypassing
RAR repression. While transcriptional mechanisms
linked to retinoid actions have been under intensive
study, posttranscriptional mechanisms also are rec-
ognized as being important. These likely contribute
to retinoid chemopreventive effects.
38
A posttransla-
tional mechanism linked to retinoid chemopreven-
tion was identified through studies conducted in an
in vitro model, as discussed below.
38,39
Retinoid Lung Cancer Chemoprevention
Mechanism
The optimal retinoids to be used in lung cancer
chemoprevention trials need to be determined. If
clinical outcome is the only end point used for
clinical lung cancer prevention activity, then
progress in this field will not be rapid. One approach
with which to assess the activities of candidate
chemoprevention agents is to examine the effects in
relevant preclinical models of lung carcinogenesis.
These models include in vitro cellular models,
20
experimental animal models of carcinogen-induced
lung tumors,
64
as well as genetically engineered mice
that develop lung cancer.
65
The mechanistic insights
derived from these models should aid in the conduct
of clinical chemoprevention trials. This chapter fo-
cuses on the use of cellular models of lung cancer
prevention.
Epithelial cell transformation can be prevented in
vitro by retinoid treatment.
20
The BEAS-2B immor-
talized human bronchial epithelial cell line was
adapted to investigate tobacco-related carcinogenic
transformation mechanisms. These cells were trans-
formed after independent exposure to the following
carcinogens: cigarette smoke condensate; or N-
nitrosamine-4-(methylnitrosamino)-1-(3 pyridyl)-1-
butanone (NNK).
20
RA treatment prevented
cigarette smoke condensate-dependent or NNK-
dependent carcinogenic transformation of these
bronchial epithelial cells. This chemopreventive ac-
tivity in human bronchial epithelial cells was linked
to the triggering of G1 cell cycle arrest, concomitant
growth suppression, and a decline in the expression
of G1 cyclin proteins.
20,38,39
The signaling of this G1
arrest involved a posttranslational mechanism. Inhib-
itors of the proteasome-dependent degradation
pathway blocked retinoid repression of G1 cyclin
expression.
38,39
It was hypothesized that the delay at
G1 signaled by retinoid treatment permitted the
repair of genomic DNA damage caused by carcino-
gens. This may be a common retinoid mechanism
that signaled growth suppression in that it was also
activated during tumor cell differentiation.
66
When normal, immortalized, or carcinogen-
transformed human bronchial epithelial cells were
treated with receptor-selective retinoids, RAR-
dependent and RXR-dependent agonists signaled
growth suppression, while RAR and RAR agonists
did not.
39
Carotenoids were unable to repress cyclin
D1 protein expression or to activate this proteasome-
dependent degradation pathway.
39
RAR and
RXR agonists, unlike other examined retinoid
receptor-selective agonists, induced the proteasome-
dependent proteolysis pathway that previously was
shown to be activated by RA treatment. Whether this
chemopreventive mechanism is activated in vivo
during a retinoid clinical trial is the subject of current
work. Independent evidence for an important role for
retinoid induction of a ubiquitin-dependent degrada-
tion pathway was found from microarray analyses.
67
Retinoid target genes were studied during leukemic
cell differentiation. One cluster of species that was
induced by RA treatment was that of the proteasomal
degradation pathway. These and other findings
66
were
consistent with the view that the promotion of ubiq-
uitination by retinoid treatment is involved directly in
the triggering of G1 arrest and the progression of a
tumor cell differentiation program.
One prediction from these bronchial epithelial cell
chemopreventive studies was that either cyclin D1 or
cyclin E would be expressed aberrantly in bronchial
preneoplasia. Bronchial preneoplastic tissues were
examined for G1 cyclin expression. It was hypothe-
sized that the expression of these species would be
deregulated early during lung carcinogenesis. The
aberrant expression of cyclin D1 or cyclin E was
found to be frequent in bronchial preneoplasia and
in associated squamous cell carcinomas.
8
The aber-
rant expression of these cyclins was more frequent
than that detected for either p53 or the retinoblas-
toma gene product (Rb).
8
These findings implicated
the deregulated expression of G1 cyclins as candi-
date biomarkers for lung cancer prevention trials.
Perhaps treatment with an effective lung cancer
chemoprevention agent would repress aberrant G1
cyclin expression in lung tissues that were at high risk
for malignant conversion. Future clinical trials will
determine whether this is the case.
Cyclooxygenase-2 Inhibitors
Inducible cyclooxygenase (COX)-2 is involved in
the synthesis of prostaglandins from arachidonic acid
and often is activated during inflammation (Fig 2).
Evidence exists for COX-2 as a therapeutic target for
cancer chemoprevention, as has been reviewed pre-
viously.
17,18
Notably, celecoxib, a selective COX-2
inhibitor, decreased the number of polyps in familial
polyposis coli (FAP), a cancer predisposition syn-
drome.
68
Several lines of evidence have provided a
rationale for the use of a COX-2 inhibitor in lung
carcinogenesis.
17,18
COX-2 regulated synthesis of
prostaglandins that promoted tumorigenesis and
COX-2 inhibition reduced NNK-mediated lung ad-
enomas in the A/J mouse. Preclinical studies indi-
cated that COX-2 overexpression inhibited apopto-
sis, and COX-2 also was linked to the regulation of
angiogenesis. The differential expression of COX-2
was observed in lung cancers, and this had a negative
prognostic impact in stage I lung cancers.
69
Epide-
miologic evidence also was consistent with a role for
COX-2 inhibition in lung cancer therapy or preven-
tion.
17,18
Several COX-2 inhibitors are already avail-
able for cancer chemoprevention trials, as shown in
Figure 3. COX-2 inhibitors have targeted inducible
COX-2, which results in pharmacologic selectivity
that overcomes clinical toxicities from the inhibition
of constitutively expressed COX-1. Diverse biological
effects could result fromthis inhibition of prostaglandin
synthesis. Some prostaglandins exert anticarcinogenic
actions. Others could have an opposite effect. Future
work should identify those regulated prostaglandins
that directly mediate chemopreventive effects. These
could represent other pharmacologic targets for lung
cancer prevention.
Epidermal Growth Factor Receptor Inhibition
Preclinical and clinical data provided a strong
rationale for targeting the epidermal growth factor
receptor (EGFR) pathway in lung cancer preven-
tion, as shown in Figure 4. Differential overexpres-
sion of the EGFR and several of its ligands was
reported in malignant and preneoplastic compared
to normal lung tissues.
70,71
In vitro studies
71
demon-
strated promotion of mitogenesis of pulmonary cells
following treatment with the epidermal growth fac-
Figure 2. Arachidonic acid generates biologically active prosta-
glandins that can play an important role in cancer chemopreven-
tion. Two enzymes are involved in this biosynthesis, constitutive
COX-1 and inducible COX-2. Celecoxib and rofecoxib preferen-
tially inhibit COX-2, and this inhibition is associated with cancer
chemopreventive activity.
Figure 3. The structure of candidate lung cancer chemopreven-
tion agents. The depicted pharmacologic agents include several
that are already available for clinical chemopreventive trials.
These include COX-2 inhibitors (ie, celecoxib and rofecoxib),
inhibitors of the tyrosine kinase associated with the EGFR, such
as ZD1839 (Iressa; AstraZeneca; Wilmington, DE) and OSI-774
(erlotinib, Tarceva; Genentech; South San Francisco, CA), as
well as the RXR agonist bexarotene (Targretin; Ligand Pharma-
ceuticals; San Diego, CA).
Figure 4. Schematic representation of the cellular EGFR
pathway. Potential therapeutic targets of this pathway are de-
picted. The EGFR is located at the cellular surface, where it is
activated by interactions with its ligands (F) secreted in an
autocrine or paracrine manner. This leads to receptor autophos-
phorylation and to mitogenic signals as well as other signals that
are transmitted to the nucleus through second messenger path-
ways. In the nucleus, these effects often result in an induction of
AP-1 components (shown here as 1Fos and 1Jun). These
ligand-receptor interactions can be pharmacologically targeted by
using anti-EGFR antibodies or small molecule inhibitors of the
EGFR-associated tyrosine kinase ().
tor. In xenograft models, pharmacologic inhibitors of
the EGFR (blocking antibodies or inhibitors of the
EGFR-associated tyrosine kinase) were shown to
oppose this mitogenic stimulation pathway.
70
Direct
evidence for EGFR as a target for lung cancer
prevention comes from finding that the exposure of
human bronchial epithelial cells to the carcinogen
NNK increased EGFR expression.
71
Notably, treat-
ment with the chemopreventive agent RA prevented
this induction.
71
Other evidence for an important
role for the EGFR in lung cancer prevention comes
from clinical findings of activity of EGFR-specific
inhibitors in the treatment of advanced-stage cancers
including chemotherapy-refractory lung cancer
cases.
70,72,73
The favorable toxicity profile observed
in the advanced-disease setting with these agents
indicated that targeting the EGFR should be con-
sidered for lung cancer chemoprevention trials.
Other Agents
Other candidate cancer chemoprevention agents
include those acting through specific nuclear recep-
tors such as the vitamin D receptor (ie, deltanoids),
peroxisome proliferator-activated receptor-, or
other nuclear receptors. One promising class of
pharmacologic agents for study is the synthetic trit-
erpenoids, which are derivatives of natural products,
the cyclosqualenoids.
74
Triterpenoids have exhibited
potent differentiation-inducing, antiproliferative and
anti-inflammatory activities. Pertinent to their poten-
tial role in cancer chemoprevention, one of the
synthesized triterpenoids, CDDO, suppressed the
induction of the inflammatory enzymes inducible
nitric oxide synthase (iNOS) and COX-2.
74
Whether
these findings will be extended to the setting of
chemoprevention of lung tumors or other tumors is
the subject of ongoing work. Other candidate lung
cancer chemoprevention agents are displayed in
Table 1.
Combination Therapy
Clinical cancer chemoprevention trials have fea-
tures that are distinct from other therapeutic trials.
To exert the desired clinical effects, chemopreven-
tion agents would be administered on a long-term
basis and should have few, if any, associated clinical
toxicities. For individuals who are at increased risk
for lung cancer, primary cancer prevention with
chemopreventive agents, when coupled with smok-
ing cessation and other interventions, would be an
attractive strategy to reduce cancer risk. Even in
individuals who were at high risk for a primary lung
cancer, a chemoprevention agent would not be
clinically adopted when side effects frequently occur.
In contrast, subjects who already have undergone
resection for a primary lung cancer may accept some
side effects of chemoprevention agents, if this would
reduce the risk of a second cancer.
One way to limit the clinical toxicities of cancer
chemopreventive agents would be through combina-
tion therapy. Agents targeting different chemopre-
ventive pathways could each be administered at
dosages lower than when those agents are used as
single agents. This should reduce the clinical toxici-
ties of each agent. For each chemopreventive agent
that is used in a combination regimen, a valid target
would be required based on preclinical and clinical
activities. A combination regimen also must be asso-
ciated with a tolerable toxicity profile as well as a safe
and convenient schedule of long-term administra-
tion. Ideally, synergistic or additive effects should be
observed in in vitro and animal models. Animal
model testing could establish that a combination
regimen is safe. If available, the clinical evidence for
drug synergy in the treatment of patients with
advanced-stage lung cancer might provide a basis for
the use of a regimen in lung cancer chemopreven-
tion. An example of combination chemopreventive
regimens to consider would involve the use of a
rexinoid with an EGFR inhibitor.
71
Alternatively, a
retinoid might be used in conjunction with an agent
that would modify chromatin structure, such as a
histone deacetylase inhibitor.
75,76
Other potential
combination regimens exist.
Recommendation
6. For individuals at risk for lung cancer and for
patients with a history of lung cancer there are
not yet sufficient data to recommend the use of
any agent either alone or in combination for
primary, secondary, or tertiary lung cancer
chemoprevention outside of a clinical trial.
Level of evidence, poor; benefit, none/negative;
grade of recommendation, I
Innovative Delivery Approaches
Chemopreventive treatments typically have in-
volved oral administration of a pharmacologic agent.
Yet, aerosolized delivery of chemopreventive agents
is a promising alternative delivery strategy.
77
Since
pulmonary tissues are directly targeted, this delivery
approach should reduce or prevent systemic toxici-
ties. Proof-of-principle animal studies
7880
high-
lighted the use of aerosolized retinoids or steroids in
lung cancer chemoprevention. These preclinical
studies have supported a similar strategy in clinical
trials. Aerosolized delivery could target small or large
airways. Drug formulations need to be optimized to
deliver to the desired airways. Technical advances
that will become clinically available include the use
of programmable delivery systems. Other potential
approaches for some chemopreventive agents might
include intranasal delivery or sustained release from
intramuscular injections.
Clinical Considerations: Biomarkers and Surrogate
End Points
Cancer chemoprevention trials are of a large size
and require long clinical follow-up. For this reason,
biomarkers or surrogate end points have been pro-
posed to assess chemopreventive responses even
before clinical outcomes become known. These bio-
markers could be indicative of carcinogen-induced
changes in affected pulmonary cells or tissues. Ex-
amples in the lung include genomic instability that
leads to aneuploidy or LOH, cell cycle alterations
that affect cellular proliferation, as well as changes in
transcription that perhaps are due to methylation-
dependent or acetylation-dependent changes in the
genome, among other alterations. Examples of ge-
netic alterations that occur in lung carcinogenesis
include those affecting oncogenes (eg, ras, myc,
EGFR, and others) or tumor suppressor genes (eg,
p53 and Rb). Conceivably, these not only represent
targets for cancer chemoprevention but also may
serve as surrogate end points for the response to
cancer chemopreventive agents. If validated biomar-
kers are identified, then clinical trials could be
designed around changes in these markers rather
than evaluating lung cancer mortality as in major
trials to date. Trials designed around surrogate end
points should be smaller in their sample sizes and
shorter in their duration than trials designed around
mortality.
Lung cancer chemoprevention trials that have
similar eligibility criteria and study designs would
permit comparisons to be made between studies
conducted at different centers. It would be helpful to
have improved models with which to assess lung
cancer risk. A possible inclusion criterion for lung
cancer chemoprevention trials might include the
presence of persistent dysplasia found in repeat
biopsy or cytology specimens. Subjects in random-
ized, placebo-controlled chemoprevention trials
should be stratified for smoking history. Perhaps
chemopreventive agents would exert different ac-
tions in current, former, and never-smokers. Every
effort should be made to encourage smoking cessa-
tion, but this may not always occur in the study
population. Lung cancer chemoprevention trials
should carefully monitor for adverse events, espe-
cially for potential adverse events in current smok-
ers. It is desirable to enroll former smokers and
never-smokers in lung cancer chemoprevention trials.
There is some concern about the inclusion of current
smokers in these trials, given the potential for neg-
ative interactions with chemopreventive agents. Ev-
idence for the activity of pharmacologic agents in
patients with advanced-stage lung cancers would
provide a rationale for their use in lung cancer
chemopreventive trials.
Proof of principle lung cancer chemoprevention
trials are warranted. These would validate the de-
sired therapeutic effects on a target. Short-term
mechanistic trials could be undertaken in which
pharmacologic effects on a target would be assessed
by comparing changes in posttreatment biopsy spec-
imens compared to those obtained before treatment.
The biochemical effects of the agent on the target
could be assessed in harvested tissues. Pharmaco-
kinetic studies should be performed with candidate
chemopreventive agents. These trials might yield
early objective tumor responses that would offer
another reason for the use of a pharmacologic agent
in lung cancer chemoprevention. A validated agent
could be used in a combination chemopreventive
regimen.
Pathologic changes after chemopreventive agent
treatment would be reasonable end points to con-
sider, especially when persistent dysplastic alter-
ations occur in initial bronchial biopsy specimens,
bronchial washings, or sputum specimens. Fluores-
cence bronchoscopy
81
as an adjunct to white-light
bronchoscopy may be useful to monitor responses in
lung cancer chemoprevention trials. Candidate ge-
netic, biochemical, or cellular markers in the lung
include changes in genomic DNA methylation or
acetylation, the presence of LOH, as well as apopto-
sis, angiogenesis, inflammation, or changes in cellu-
lar proliferation in the affected tissues.
Conclusions
Cancer chemoprevention is an attractive approach
with which to reduce lung cancers by treating early
steps in lung carcinogenesis. There is a convergence
of basic scientific and clinical findings in lung cancer
chemoprevention. Smoking cessation and prevention
are essential to reduce the societal impact of tobacco
carcinogenesis. Pharmacologic interventions also can
be used to reverse or arrest the progression of lung
carcinogenesis. While promising pharmacologic evi-
dence exists for lung cancer chemoprevention, the
optimal agent has not yet been identified that is
active in primary or secondary lung cancer chemo-
prevention. For this reason, additional clinical trials
are needed that emphasize a mechanistic approach
in which mechanisms identified in vitro can be
validated in vivo. Given the long-term nature of the
interventions for cancer chemoprevention, pharma-
cologic agents should be administered with few, if
any, associated clinical toxicities. Combination che-
mopreventive regimens offer a strategy to reduce the
number of associated clinical toxicities since each
agent could be prescribed at dosages lower than
when administered as single agents. One way to
develop future lung cancer chemoprevention trials is
to explore chemopreventive activities in relevant
preclinical models. These models should help to
identify important pathways that are responsible for
chemopreventive effects and help to validate known
therapeutic targets or highlight novel therapeutic
targets for lung cancer prevention. A potential exam-
ple of this is the pharmacologic effect on G1 cyclins
found in human bronchial epithelial cells.
38,39
Biomarkers or surrogate end points should prove
useful in identifying chemopreventive targets as well
as in highlighting changes that place the affected
pulmonary cells or tissues at high risk for the devel-
opment of lung cancer. Changes in these markers
represent candidate surrogate end points for clinical
cancer chemoprevention trials. In the near term, as
the clinical lung cancer chemoprevention field ad-
vances, it will be important to understand mechanis-
tically how chemopreventive agents act and when
they should be prescribed. An improved understand-
ing of the pharmacology of cancer chemoprevention
agents should aid in the design and conduct of lung
cancer chemoprevention trials.
Summary of Recommendations
1. For all individuals, smoking prevention should
be strongly encouraged to decrease the risk of
lung cancer. Level of evidence, good; benefit,
2. For all individuals, school-based and commu-
nity-based interventions that are aimed at re-
ducing tobacco exposure should be recom-
mended, including a life skills training
approach that is aimed at reducing tobacco,
alcohol, and illicit drug use, campaigns with
brief recurring antismoking messages, high to-
bacco excise taxes, and restrictions on smoking
in the workplace. Level of evidence, good; bene-
fit, substantial; grade of recommendation, A
3. Smokers should be identified as smoking ces-
sation reduces the risk of lung cancer. Level of
recommendation, A
4. Current smokers should be advised to quit
smoking, and, when appropriate, clinicians
should prescribe and monitor pharmacother-
apy. Individuals who smoke and want to quit
also should have access to psychosocial treat-
ment and behavioral modification therapies as
indicated. There is sufficient to strong evidence
that indicates these practices will help to in-
crease long-term smoking abstinence rates.
5. Individuals who are at risk for lung cancer and
were treated with -carotene, retinol, isotreti-
noin, or N-acetyl-cysteine for lung cancer pre-
vention did not experience clinical benefits.
There is also evidence that the use of -
carotene or isotretinoin for lung cancer chemo-
prevention in high-risk individuals may increase
the risk for lung cancer, especially in individu-
als who continue to smoke. These agents
should not be used outside of a clinical trial for
primary, secondary, or tertiary lung cancer
prevention. Level of evidence, good; benefit,
harmful; grade of recommendation, D
6. For individuals who are at risk for lung cancer
and for patients with a history of lung cancer,
there are not yet sufficient data to recommend
the use of any agent either alone or in combi-
nation for primary, secondary, or tertiary lung
cancer chemoprevention outside of a clinical
trial. Level of evidence, insufficient; benefit,
lacking data; grade of recommendation, I
References
1 Weston A, Harris CC. Chemical carcinogenesis. In: Holland
JF, Bast RC, Kufe DW, et al, eds. Cancer medicine. Balti-
more, MD: Williams & Wilkins, 1997; 261276
2 Slaughter DP, Southwick HW, Snejkal WP. Field canceriza-
tion in oral stratified squamous epithelium: clinical implica-
tions for multicentric origin. Cancer 1953; 6:963968
3 Saccomanno G, Archer VE, Auerbach O, et al. Development
of carcinoma of the lung as reflected in exfoliated cells.
Cancer 1974; 33:256270
4 Mao L, Hruban RH, Boyle JO, et al. Detection of oncogene
mutations in sputum precedes diagnosis of lung cancer.
Cancer Res 1994; 54:16341637
5 Sozzi G, Miozzo M, Tagliabue E, et al. Cytogenetic abnor-
malities and overexpression of receptors for growth factors in
normal bronchial epithelium and tumor samples of lung
cancer patients. Cancer Res 1991; 51:400404
6 Sozzi G, Musso K, Ratcliffe C, et al. Detection of microsat-
ellite alterations in plasma DNA of non-small cell lung cancer
patients: a prospect for early diagnosis. Clin Cancer Res 1999;
5:26892692
7 Ihde DC, Minna JD. Non-small cell lung cancer. Part I:
Biology, diagnosis, and staging. Curr Probl Cancer 1991;
15:61104
8 Lonardo F, Rusch V, Langenfeld J, et al. Overexpression of
cyclins D1 and E is frequent in bronchial preneoplasia and
precedes squamous cell carcinoma development. Cancer Res
1999; 59:24702476
9 Rusch V, Klimstra D, Linkov I, et al. Aberrant expression of
p53 or the epidermal growth factor receptor is frequent in
early bronchial neoplasia and coexpression precedes squa-
mous cell carcinoma development. Cancer Res 1995; 55:
13651372
10 Albanell J, Lonardo F, Rusch V, et al. High telomerase
activity in primary lung cancers: association with increased
cell proliferation rates and advanced pathologic stage. J Natl
Cancer Inst 1997; 89:16091615
11 Wistuba II, Behrens C, Milchgrub S, et al. Sequential
molecular abnormalities are involved in the multistage devel-
opment of squamous cell lung carcinoma. Oncogene 1999;
18:643650
12 Park I-W, Wistuba II, Maitra A, et al. Multiple clonal
abnormalities in the bronchial epithelium of patients with
lung cancer. J Natl Cancer Inst 1999; 91:18631868
13 Pitterle DM, Jolicoeur EM, Bepler G. Hot spots for molec-
ular genetic alterations in lung cancer. In Vivo 1998; 12:643
658
14 Boyle JO, Lonardo F, Chang JH, et al. Multiple high-grade
bronchial dysplasia and squamous cell carcinoma: concordant
and discordant mutations. Clin Cancer Res 2001; 7:259266
15 Sporn MB, Dunlop NM, Newton DL, et al. Prevention of
chemical carcinogenesis by vitamin A and its synthetic analogs
(retinoids). Fed Proc 1976; 35:13321338
16 Fisher B, Costantino JP, Wickerham DL, et al. Tamoxifen for
prevention of breast cancer: report of the National Surgical
Adjuvant Breast and Bowel Project P-1 Study. J Natl Cancer
Inst 1998; 90:13711388
17 Hong WK, Sporn MB. Recent advances in chemoprevention
of cancer. Science 1997; 278:10731077
18 Dmitrovsky E, Sporn MB. Pharmacology of cancer chemo-
prevention. In: Bertino JR, ed. Encyclopedia of Cancer. San
Diego, CA: Academic Press, 2002; 449456
19 Bunn PA Jr, Soriano A, Johnson G, et al. New therapeutic
strategies for lung cancer: biology and molecular biology
come of age. Chest 2000; 117:163S168S
20 Langenfeld J, Lonardo F, Kiyokawa H, et al. Inhibited
transformation of immortalized human bronchial epithelial
cells by retinoic acid is linked to cyclin E down-regulation.
Oncogene 1996; 13:19831990
21 Hibi K, Liu Q, Beaudry GA, et al. Serial analysis of gene
expression in non-small cell lung cancer. Cancer Res 1998;
58:56905694
22 Hellmann GM, Fields WR, Doolittle DJ. Gene expression
profiling of cultured human bronchial epithelial and lung
carcinoma cells. Toxicol Sci 2001; 61:154163
23 Chen JJ, Peck K, Hong TM, et al. Global analysis of gene
expression in invasion by a lung cancer model. Cancer Res
2001; 61:52235230
24 Goodwin LO, Mason JM, Hajdu SI. Gene expression patterns
of paired bronchioloalveolar carcinoma and benign lung
tissue. Ann Clin Lab Sci 2001; 31:369375
25 Hong WK, Itri LM. Retinoids and human cancer. In: Sporn
MB, Roberts AB, Goodman DS, et al, eds. The retinoids:
biology, chemistry, and medicine. New York, NY: Raven
Press, 1994; 597630
26 Wolbach SB, Howe PR. Tissue changes following deprivation
of fat-soluble vitamin A. J Exp Med 1925; 42:753777
27 Hong WK, Endicott J, Itri LM, et al. 13-cis-retinoic acid in
the treatment of oral leukoplakia. N Engl J Med 1986;
315:15011505
28 Meyskens FL Jr, Surwit E, Moon TE, et al. Enhancement of
regression of cervical intraepithelial neoplasia II (moderate
dysplasia) with topically applied all-trans-retinoic acid: a
randomized trial. J Natl Cancer Inst 1994; 86:539543
29 Kraemer KH, DiGiovanna JJ, Moshell AN, et al. Prevention
of skin cancer in xeroderma pigmentosum with the use of oral
isotretinoin. N Engl J Med 1988; 318:16331637
30 Hong WK, Lippman SM, Itri LM, et al. Prevention of second
primary tumors with isotretinoin in squamous-cell carcinoma
of the head and neck. N Engl J Med 1990; 323:795801
31 Pastorino U, Infante M, Maioli M, et al. Adjuvant treatment
of stage I lung cancer with high-dose vitamin A. J Clin Oncol
1993; 11:12161222
32 Muto Y, Moriwaki H, Ninomiya M, et al. Prevention of
second primary tumors by an acyclic retinoid, polyprenoic
acid, in patients with hepatocellular carcinoma: Hepatoma
Prevention Study Group. N Engl J Med 1996; 334:15611567
33 Lippman SM, Lee JJ, Karp DD, et al. Randomized phase III
intergroup trial of isotretinoin to prevent second primary
tumors in stage I non-small-cell lung cancer. J Natl Cancer
Inst 2001; 93:605618
34 The Alpha-Tocopherol, Beta-Carotene Cancer Prevention
Study Group. The effect of vitamin E and beta carotene on
the incidence of lung cancer and other cancers in male
smokers. N Engl J Med 1994; 330:10291035
35 Hennekens CH, Buring JE, Manson JE, et al. Lack of effect
of long-term supplementation with beta carotene on the
incidence of malignant neoplasms and cardiovascular disease.
N Engl J Med 1996; 334:11451149
a combination of beta carotene and vitamin A on lung cancer
and cardiovascular disease. N Engl J Med 1996; 334:1150
1155
37 Khuri FR, Rigas JR, Figlin RA, et al. Multi-institutional phase
I/II trial of oral bexarotene in combination with cisplatin and
vinorelbine in previously untreated patients with advanced
non-small-cell lung cancer. J Clin Oncol 2001; 19:26262637
38 Langenfeld J, Kiyokawa H, Sekula D, et al. Posttranslational
regulation of cyclin D1 by retinoic acid: a chemoprevention
mechanism. Proc Natl Acad Sci U S A 1997; 94:1207012074
39 Boyle JO, Langenfeld J, Lonardo F, et al. Cyclin D1 prote-
olysis: a retinoid chemoprevention signal in normal, immor-
talized, and transformed human bronchial epithelial cells.
40 Chambon P. A decade of molecular biology of retinoic acid
receptors. FASEB J 1996; 10:940954
41 Mangelsdorf D, Umesono K, Evans RM. In: Sporn MB,
Roberts AB, Goodman DS, et al, eds. The retinoids: biology,
chemistry, and medicine. New York, NY: Raven Press, 1994;
319349
42 Glass CK, Rose DW, Rosenfeld MG. Nuclear receptor
coactivators. Curr Opin Cell Biol 1997; 9:222232
43 A unified nomenclature system for the nuclear receptor
superfamily [letter]. Cell 1999; 97:161163
44 Boylan JF, Lohnes D, Taneja R, et al. Loss of retinoic acid
receptor gamma function in F9 cells by gene disruption
results in aberrant Hoxa-1 expression and differentiation
upon retinoic acid treatment. Proc Natl Acad Sci U S A 1993;
90:96019605
45 Boylan JF, Lufkin T, Achkar CC, et al. Targeted disruption of
retinoic acid receptor alpha (RAR alpha) and RAR gamma
results in receptor-specific alterations in retinoic acid-
mediated differentiation and retinoic acid metabolism. Mol
Cell Biol 1995; 15:843851
46 Chiba H, Clifford J, Metzger D, et al. Distinct retinoid X
receptor-retinoic acid receptor heterodimers are differen-
tially involved in the control of expression of retinoid target
genes in F9 embryonal carcinoma cells. Mol Cell Biol 1997;
17:30133020
47 Lehmann JM, Dawson MI, Hobbs PD, et al. Identification of
retinoids with nuclear receptor subtype-selective activities.
Cancer Res 1991; 51:48044809
48 Lehmann JM, Jong L, Fanjul A, et al. Retinoids selective for
retinoid X receptor response pathways. Science 1992; 258:
19441946
49 Lotan R, Xu XC, Lippman SM, et al. Suppression of retinoic
acid receptor-beta in premalignant oral lesions and its up-
regulation by isotretinoin. N Engl J Med 1995; 332:1405
1410
50 Leid M, Kastner P, Lyons R, et al. Purification, cloning, and
RXR identity of the HeLa cell factor with which RAR or TR
heterodimerizes to bind target sequences efficiently. Cell
1992; 68:377395
51 Kliewer SA, Umesono K, Mangelsdorf DJ, et al. Retinoid X
receptor interacts with nuclear receptors in retinoic acid,
thyroid hormone and vitamin D3 signaling. Nature 1992;
355:446449
52 Zhang XK, Lehmann J, Hoffmann B, et al. Homodimer
formation of retinoid X receptor induced by 9-cis retinoic
acid. Nature 1992; 358:587591
53 Chen JD, Evans RM. A transcriptional co-repressor that
interacts with nuclear hormone receptors. Nature 1995;
377:454457
54 Nason-Burchenal K, Dmitrovsky E. The retinoids: cancer ther-
apy and prevention mechanisms. In: Nau H, Blaner W, eds.
Retinoids: the biochemical and molecular basis of vitamin A and
retinoid action; handbook of experimental pharmacology. Vol
139. Berlin, Germany: Springer, 1999; 301322
55 Muindi J, Frankel S, Miller W, et al. Continuous treatment
with all-trans retinoic acid causes a progressive reduction in
plasma drug concentrations: implications for relapse and
retinoid resistance in patients with acute promyelocytic
leukemia. Blood 1992; 79:299303
56 Cornic M, Delva L, Guidez F, et al. Induction of retinoic
acid-binding protein in normal and malignant human myeloid
cells by retinoic acid in acute promyelocytic leukemia pa-
tients. Cancer Res 1992; 52:33293334
57 Heyman RA, Mangelsdorf DJ, Dyck JA, et al. 9-cis retinoic
acid is a high affinity ligand for the retinoid X receptor. Cell
1992; 68:397406
58 Levin AA, Sturzenbecker LJ, Kazmer S, et al. 9-cis retinoic
acid stereoisomer binds and activates the nuclear receptor
RXR alpha. Nature 1992; 355:359361
59 Fanjul A, Dawson MI, Hobbs PD, et al. A new class of
retinoids with selective inhibition of AP-1 inhibits prolifera-
tion. Nature 1994; 372:107111
60 Oridate N, Suzuki S, Higuchi M, et al. Involvement of
reactive oxygen species in N-(4-hydroxyphenyl)retinamide-
induced apoptosis in cervical carcinoma cells. J Natl Cancer
Inst 1997; 89:11911198
61 Kitareewan S, Spinella MJ, Allopenna J, et al. 4HPR triggers
apoptosis but not differentiation in retinoid sensitive and
resistant human embryonal carcinoma cells through an RARy
independent pathway. Oncogene 1999; 18:57475755
62 Veronesi U, De Palo G, Marubini E, et al. Randomized trial of
feuretinide to prevent second breast malignancy in women with
early breast cancer. J Natl Cancer Inst 1999; 91:18471856
63 Lotan R. Aberrant expression of retinoid receptors and lung
carcinogenesis. J Natl Cancer Inst 1999; 91:989991
64 Moon RC, Mehta RG, Rao KVN. Retinoids and cancer in
experimental animals. In: Sporn MB, Roberts AB, Goodman
DS, et al, eds. The retinoids: biology, chemistry, and medi-
cine. New York, NY: Raven Press, 1994; 573595
65 Johnson L, Mercer K, Greenbaum D, et al. Somatic activation
of the K-ras oncogene causes early onset lung cancer in mice.
Nature 2001; 410:11111116
66 Spinella MJ, Freemantle SJ, Sekula D, et al. Retinoic acid
promotes ubiquitination and proteolysis of cyclin D1 during
induced tumor cell differentiation. J Biol Chem 1999; 274:
2201322018
67 Tamayo P, Slonim D, Mesirov J, et al. Interpreting patterns of
gene expression with self-organizing maps: methods and
application to hematopoietic differentiation. Proc Natl Acad
Sci U S A 1999; 96:29072912
68 Steinbach G, Lynch PM, Phillips RK, et al. The effect of
celecoxib, a cyclooxygenase-2 inhibitor, in familial adenoma-
tous polyposis. N Engl J Med 2000; 342:19461952
69 Khuri FR, Wu H, Lee JJ, et al. Cyclooxygenase-2 overexpres-
sion is a marker of poor prognosis in stage I non-small cell
lung cancer. Clin Cancer Res 2001; 7:861867
70 Mendelsohn J. The epidermal growth factor receptor as a
target for cancer therapy. Endocr Relat Cancer 2001; 8:39
71 Lonardo F, Dragnev KH, Freemantle SJ, et al. Evidence for
the epidermal growth factor receptor as a target for lung
cancer prevention. Clin Cancer Res 2002; 8:5460
72 Perez-Soler R, Chachoua A, Huberman M, et al. A phase II
trial of the epidermal growth factor receptor (EGFR) tyrosine
kinase inhibitor OSI-774, following platinum-based chemo-
therapy, in patients (pts) with advanced, EGFR-expressing,
non-small cell lung cancer (NSCLC) [abstract]. Proc Am Soc
Clin Oncol 2001; 20:1235a
73 Arteaga CL, Johnson DH. Tyrosine kinase inhibitors-ZD1839
(Iressa). Curr Opin Oncol 2001; 13:491498
74 Suh N, Wang Y, Honda T, et al. A novel synthetic oleanane
triterpenoid, 2-cyano-3,12-dioxoolean-1,9-dien-28-oic acid, with
potent differentiating, antiproliferative, and anti-inflammatory
activity. Cancer Res 1999; 59:336341
75 He LZ, Tolentino T, Grayson P, et al. Histone deacetylase
inhibitors induce remission in transgenic models of therapy-
resistant acute promyelocytic leukemia. J Clin Invest 2001;
108:13211330
76 Marks PA, Richon VM, Breslow R, et al. Histone deacetylase
inhibitors as new cancer drugs. Curr Opin Oncol 2001;
13:477483
77 Spinella MJ, Dmitrovsky E. Aerosolized delivery and lung
cancer prevention: pre-clinical models show promise. Clin
Cancer Res 2000; 6:29632964
78 Wattenberg LW, Wiedmann TS, Estensen RD, et al. Che-
moprevention of pulmonary carcinogenesis by brief exposures
to aerosolized budesonide or beclomethasone dipropionate
and by the combination of aerosolized budesonide and dietary
myo-inositol. Carcinogenesis 2000; 21:179182
79 Wattenberg LW, Wiedmann TS, Estensen RD, et al. Che-
moprevention of pulmonary carcinogenesis by aerosolized
budesonide in female A/J mice. Cancer Res 1997; 57:5489
5492
80 Dahl AR, Grossi IM, Houchens DP, et al. Inhaled isotretinoin
(13-cis. retinoic acid) is an effective lung cancer chemopre-
ventive agent in A/J mice at low doses: a pilot study. Clin
Cancer Res 2000; 6:30153024
81 Lam S, Kennedy T, Unger M, et al. Localization of bronchial
intraepithelial neoplastic lesions by fluorescence bronchos-
copy. Chest 1998; 113:696702
2003;123;60-71 Chest
Konstantin H. Dragnev, Diane Stover and Ethan Dmitrovsky
Lung Cancer Prevention: The Guidelines
& Services
Updated Information
References
#BIBL
free at:
Reprints
2003;123;72-82 Chest
Peter B. Bach, Michael J. Kelley, Ramsey C. Tate and Douglas C. McCrory
Screening for Lung Cancer: A Review of the Current Literature
ISSN: 0012-3692.
Screening for Lung Cancer*
A Review of the Current Literature
Peter B. Bach, MD; Michael J. Kelley, MD; Ramsey C. Tate, BS; and
Douglas C. McCrory, MD, MHS
Study objectives: To review the available data on the early detection of lung cancer, with a focus
on three technologies: chest x-ray (CXR), sputum cytology, and low-dose CT (LDCT) scanning.
Design, setting, participants: Review of published clinical studies of early detection technologies.
The best available evidence on each topic was selected for analysis. Randomized trials were used
to evaluate CXR and sputum cytology. Cohort studies, as well as studies providing evidence
regarding rates of overdiagnosis and efficacy of initial treatment, were considered in evaluation
of LDCT. Study design and results were summarized in evidence tables. Statistical analyses of
combined data were not performed.
Measurement and results: Five randomized trials of CXR with or without sputum cytology have
been conducted, each which reports disease-specific mortality as well as other end points. None
of these studies provide support for the use of either CXR or sputum cytology for the early
detection of lung cancer in asymptomatic individuals. Eight completed and ongoing trials of
LDCT were identified. All of these studies report the frequency and stage distribution of lung
cancers found during initial (prevalence) screening, and several studies also report rates of
detection at the time of annual follow-up. No outcome data on survival or treatment are available.
A number of studies support the hypothesis of overdiagnosisthat some lung cancers detected
by LDCT may behave in an indolent manner.
Conclusions: The use of either CXR or sputum cytology for the early detection of lung cancer is
not supported by the published evidence. The evidence for LDCT appears promising, in that the
technology typically identifies lung cancer at an early stage, although corollary studies suggest
that these findings in isolation may be misleading. Further high-quality research is needed to
better define the role of LDCT in the evaluation of asymptomatic high-risk individuals.
(CHEST 2003; 123:72S82S)
Key words: evidence-based medicine; lung neoplasms; mass chest x-ray; mass screening; review; sputum; tomography,
x-ray computed
Abbreviations: CXR chest x-ray; ELCAP Early Lung Cancer Action Project; LDCT low-dose CT;
MSKCCMemorial Sloan-Kettering Cancer Center; NCI National Cancer Institute; RCT randomized controlled trial
L
ung cancer accounts for approximately 6% of all
deaths in the United States each year, and is the
leading cause of cancer death for both men and
women.
1
At present, most patients who receive an
initial diagnosis of lung cancer have advanced stage
disease, making cure with currently available therapies
unlikely. In contrast, individuals with early stage disease
can achieve cure through surgical resection. Because of
this dichotomy in outcome associated with stage at
diagnosis, there has been persistent interest in design-
ing and testing methods for early detection of lung
cancer. To place these studies in context, we articulate
the critical elements of successful screening tests and
describe the research methods that can be used to
evaluate them. We then focus on lung cancer screening
studies with chest x-ray (CXR), sputum cytology, and
low-dose CT (LDCT) scanning. For the first two
modalities, we review published randomized controlled
trials (RCTs). For LDCT, we review published and
ongoing studies, discuss the evidence that has fueled
the recent debate over efficacy, and discuss future
plans of assessment.
Elements of a Beneficial Screening Test
The benefit of a screening test is typically assessed
in two domains. First, the screening test must pro-
*From the Health Outcomes Research Group, Department of
Epidemiology and Biostatistics (Dr. Bach and Ms. Tate), and the
Department of Medicine, Memorial Sloan-Kettering Cancer
Center, New York, NY; the Department of Medicine (Dr.
Kelley), and the Center for Clinical Health Policy Research (Dr.
McCrory), Duke University, Durham, NC; the Durham Veterans
Affairs Medical Center, Durham, NC.
Correspondence to: Peter B. Bach, MD, MAPP, Health Outcomes
Research Group, Memorial Sloan-Kettering Cancer Center, 1275
York Ave, Box 221, New York, NY 10021; e-mail: bachp@mskcc.org
vide benefit to individuals who have the illness,
typically through increasing life expectancy. In order
to prolong life expectancy, the test must be capable
of detecting the disease at a point in its natural
history at which the course can be altered through
treatment, typically earlier than is encountered in
sporadic practice. Patients survive longer after diag-
nosis of early stage lung cancer than they do after
diagnosis of more advanced stage lung cancer; there-
fore, it is widely believed that early detection fol-
lowed by definitive treatment should alter the natu-
ral history of the disease and thereby reduce lung
cancer mortality.
Second, the screening test should not be danger-
ous or painful, nor should it have numerous false-
positive results that cause anxiety or necessitate
invasive or dangerous follow-up tests. From a soci-
etal perspective, the screening test should also not be
harmful to the large fraction of the population who
do not have the disease, either by consuming vast
amounts of resources or by directly affecting the
capacity of the health-care system to provide for
others.
Research Methods for Evaluating
Screening Tests
Three general methods have been advocated for
the evaluation of screening tests: randomized trials
of screening, population-based studies of screen-
ing, and observational studies of screening in
select cohorts. Traditionally, the scientific commu-
nity ranks RCTs at the top of this hierarchy of
study designs. In a randomized trial, individuals
are subjected to different intensities of screening,
and the outcome that is observed is disease-
specific mortality. The efficacy of fecal-occult
blood testing for the detection of colon cancer was
established in this manner.
24
As another example,
the Mayo Lung Project randomized individuals to
receive regularly scheduled screening (interven-
tion arm) or routine care; the primary outcome of
interest was lung cancer mortality.
57
An alterna-
tive highly indicative design is the population-
based study. In studies of this design, the impact
of a broadly implemented screening program is
assessed through changes in disease-specific mor-
tality rates within the population. The efficacy of
cervical cancer screening was established through
multiple studies that documented large declines in
the rates of cervical cancer mortality in the 1960s
and 1970s in Sweden, Finland, and a number of
other countries.
8,9
In the third designobservational
studies of screening in select cohortsthe efficacy of
the screening test is inferred from the frequency with
which it can detect early stage cancers. The subsequent
impact on disease-specific mortality is then ascertained
from historical data on the outcomes of individuals with
early stage disease, or through the documentation of
survival for the individuals enrolled in the study.
Readers should be aware that all of these study
designs have potential weaknesses. Negative results
of randomized trials may be viewed with skepticism
either because of defects in design (such as inade-
quate power), or defects in study conduct (such as
having excessive crossover between arms). For ex-
ample, despite seven separate RCTs, the efficacy of
mammography for the prevention of death from
breast cancer remains uncertain. Some investigators
believe that significant design flaws in the majority of
mammography studies limit the ability of those
studies to measure the impact of mammography.
These investigators have pointed out that the best
designed studies show no discernible impact on
overall mortality,
10,11
an assertion that has led an
independent panel of US National Cancer Institute
(NCI) advisers to revise their information summaries
about screening with mammography.
12,13
Similarly,
population-based studies can be confounded by nu-
merous factors, including the implementation of a
screening program itself. For example, there is a
substantial amount of evidence suggesting that the act
of screening for prostate cancer with the prostate-
specific antigen test may artifactually raise incidence
and mortality rates from that disease.
14,15
A study from
Black et al
16
suggested that screening interventions
in RCTs may also influence the ascertainment of
cause of death. Observational studies are subject to
particular biases, including those resulting from
lead time and length time, and those resulting
from the recruitment and selection of volunteer
cohorts.
Sputum Cytology and CXR
Sputum cytology and CXR have each been
assessed in RCTs. The first study was begun in
1960, in which 55,034 London men were random-
ized to CXR every 6 months for 3 years, or to CXR
at the beginning and the end of the 3-year peri-
od.
17,18
Follow-up exceeded 99% in both study
arms. In this study, a total of 36 additional cases of
lung cancer were identified in the frequently
screened group (132 cases) than in the group
screened only at the study inception and study
completion (96 cases). Of the cancers detected
during the 3-year study period, 44% were resected
in the screened group, while only 29% were
resected in the control group. During the 3-year
study period, lung cancer mortality was similar: 62
individuals in the frequently screened group died
of lung cancer, compared to 59 individuals in the
less frequently screened group. This study, as well
as several observational studies, did not support
the hypothesis that CXR screening would be asso-
ciated with a mortality benefit, and widespread
screening was not pursued.
19,20
Improvements in CXR and sputum cytology tech-
nologies led the NCI to readdress this question in
the 1970s. The NCI funded the Cooperative Early
Lung Cancer Group,
21
which conducted three ran-
domized studies of sputum cytologic examination
and CXR. During the same time period, a random-
ized study of CXR and sputum cytology was con-
ducted in Czechoslovakia.
22
Two of the NCI-funded RCTs (the Johns Hopkins
Lung Project and the Memorial Sloan-Kettering
Cancer Center [MSKCC] Lung Cancer Screening
Program) randomized subjects to either CXR alone
or CXR plus sputum cytology. In neither study were
there differences in the number of cancers detected,
the fraction of cancers that were resectable, or the
lung cancer mortality rate (Table 1).
2325
These
studies were primarily designed to assess the incre-
mental impact of sputum cytology in a CXR-
screened cohort. Thus, the absence of a mortality
benefit in these studies may have been due to the
insensitivity of sputum cytology as it was imple-
mented.
The NCI-funded Mayo Lung Project focused on
the combined impact of CXR and sputum cytology
for screening.
57
This study recruited male smok-
ers 45 years old between 1971 and 1976 who
had at least one pack-per-day cigarette use, an
estimated survival of at least 5 years, no evidence
of lung cancer on an initial evaluation, and suffi-
cient lung function to tolerate lobectomy. Among
the 10,933 subjects initially recruited, 91 prevalent
cases of lung cancer were detected on screening
studies (0.83%), 51 cases by CXR, 17 cases by
sputum cytology alone, and 15 cases by a combi-
nation of both tests. Subsequently, individual sub-
jects with negative and satisfactory prevalence
screens were randomized to one of two groups.
The screening group had CXRs and sputum
cytology every 4 months for 6 years; the control
group received a recommendation at the start of
the study to receive a yearly CXR and sputum
cytologic examination (that being the standard
Mayo Clinic recommendation at that time). Sub-
jects were then followed up from 1 to 5.5 years
(median, 3 years). Compliance in the screened
group was 85% during the first year and fell to
75% by the end of the study. More than one half
of the control group also underwent CXR during
the study period.
Lung cancer was detected more often in the
screened group (206 cases) than in the control
group (160 cases; Table 1). Forty-eight percent of
the cancers detected in the screened group were
localized, and thus amenable to surgical resection
with curative intent, while only 32% of the cancers
detected in the control group were localized,
suggesting that screening may increase the cure
rate. However, at the first analysis, there were no
substantive differences in lung cancer-specific
mortality rates: 3.2 per 1,000 person-years in the
screened group and 3.0 per 1,000 person-years in
the control group. All-cause mortality was also
similar (Table 1). The investigators of the Mayo
Lung Project therefore concluded that the combi-
nation of CXR and sputum cytology performed
every 4 months does not decrease lung cancer
mortality compared to the recommendation of
annual screening.
6
An analysis of the same cohort
performed 15 years after study completion yielded
a similar finding: no substantive difference in lung
cancer mortality rate between the screened group
(4.4 per 1,000 person-years) and the control group
(3.9 per 1,000 person-years).
6,26
The contemporaneous study conducted in
Czechoslovakia randomly assigned individuals to
CXR and sputum cytology every 6 months or to
initial screening followed by screening again after 3
years had elapsed. Subjects in both arms were then
screened annually for an additional 3 years. The
results of this study at 3 years were similar to the
Mayo study; a greater number of cancers had been
detected in the intervention group than the control
group (39 cases vs 27 cases), including a greater
number of cases with early stage disease (20 cases vs
10 cases; Table 1). However, there was effectively no
difference between the number of individuals in the
screened group and the control group with late-stage
cancer (19 patients vs 17 patients), nor was there a
difference in the number of patients dying of lung
cancer.
22,27
Conclusions: CXR and Sputum Cytology
The results of the five RCTs suggest that neither
CXR nor sputum cytology satisfy the primary
criterion of a beneficial screening test. Neither
test appears to prolong the life expectancy of an
individual with the disease. Whether either test
fits the second criterionthat the test is not
particularly painful or harmfulwas not ad-
dressed in sufficient detail in any of these studies
to determine.
Readers should appreciate that despite the consis-
tency of the findings in these studies, there is still
skepticism about their interpretation. One observa-
T
a
b
l
e
1
R
C
T
s
o
f
C
X
R
W
i
t
h
o
r
W
i
t
h
o
u
t
S
p
u
t
u
m
C
y
t
o
l
o
g
y
*
S
t
u
d
y
S
i
t
e
S
t
u
d
y
A
r
m
I
n
t
e
r
v
e
n
t
i
o
n
S
a
m
p
l
e
S
i
z
e
B
a
s
e
l
i
n
e
S
c
r
e
e
n
R
e
p
e
a
t
S
c
r
e
e
n
i
n
g
L
u
n
g
C
a
n
c
e
r
M
o
r
t
a
l
i
t
y
C
a
n
c
e
r
s
,
N
o
.
P
e
r
1
,
0
0
0
P
e
r
s
o
n
s
C
a
n
c
e
r
s
,
N
o
.
P
e
r
1
,
0
0
0
P
e
r
s
o
n
-
Y
e
a
r
s
P
e
r
1
,
0
0
0
P
e
r
s
o
n
-
Y
e
a
r
s
L
o
n
d
o
n
1
9
6
0
1
9
6
4
1
7
,
1
8
A
l
l
5
5
,
0
3
4
5
1
0
.
9
1
7
7
N
R
2
.
2
I
n
t
e
r
v
e
n
t
i
o
n
C
X
R
e
v
e
r
y
6
m
o
f
o
r
3
y
r
2
9
,
7
2
3
3
1
1
.
0
1
0
1
N
R
2
.
1
C
o
n
t
r
o
l
C
X
R
a
t
e
n
d
o
f
3
y
r
2
5
,
3
1
1
2
0
0
.
8
7
6
N
R
2
.
4
M
a
y
o
1
9
7
1
1
9
8
3
5
7
A
l
l
1
0
,
9
3
3
9
1
8
.
3
3
6
6
N
R
N
R
I
n
t
e
r
v
e
n
t
i
o
n
C
X
R
a
n
d
s
p
u
t
u
m
c
y
t
o
l
o
g
y
e
v
e
r
y
4
m
o
f
o
r
6
y
r
4
,
6
1
8
N
A
N
A
2
0
6
5
.
5
3
.
2
C
o
n
t
r
o
l
A
d
v
i
s
e
d
t
o
h
a
v
e
C
X
R
a
n
d
s
p
u
t
u
m
c
y
t
o
l
o
g
y
a
n
n
u
a
l
l
y
4
,
5
9
3
N
A
N
A
1
6
0
4
.
3
3
.
0
C
z
e
c
h
o
s
l
o
v
a
k
i
a
1
9
7
6
1
9
8
0
2
2
,
2
7
A
l
l
6
,
3
6
4
1
8
2
.
8
6
6
N
R
N
R
I
n
t
e
r
v
e
n
t
i
o
n
C
X
R
a
n
d
s
p
u
t
u
m
c
y
t
o
l
o
g
y
e
v
e
r
y
6
m
o
f
o
r
3
y
r
3
,
1
7
2
N
A
N
A
3
9
N
R
3
.
6
C
o
n
t
r
o
l
C
X
R
a
n
d
s
p
u
t
u
m
c
y
t
o
l
o
g
y
a
t
e
n
d
o
f
3
y
r
3
,
1
7
4
N
A
N
A
2
7
N
R
2
.
6
M
S
K
C
C
1
9
7
4
1
9
8
2
2
1
,
2
5
A
l
l
1
0
,
0
4
0
5
3
5
.
3
2
3
5
N
R
N
R
I
n
t
e
r
v
e
n
t
i
o
n
C
X
R
a
n
n
u
a
l
l
y
a
n
d
s
p
u
t
u
m
c
y
t
o
l
o
g
y
e
v
e
r
y
4
m
o
4
,
9
6
8
3
0
6
.
0
4
1
1
4
3
.
7
2
.
7
C
o
n
t
r
o
l
C
X
R
a
n
n
u
a
l
l
y
5
,
0
7
2
2
3
4
.
5
1
2
1
3
.
8
2
.
7
J
o
h
n
s
H
o
p
k
i
n
s
1
9
7
3
1
9
8
2
2
1
,
2
3
A
l
l
1
0
,
3
8
6
7
9
7
.
6
3
9
6
N
R
I
n
t
e
r
v
e
n
t
i
o
n
C
X
R
a
n
n
u
a
l
l
y
a
n
d
s
p
u
t
u
m
c
y
t
o
l
o
g
y
e
v
e
r
y
4
m
o
5
,
2
2
6
3
9
7
.
5
1
9
4
N
R
3
.
4
C
o
n
t
r
o
l
C
X
R
a
n
n
u
a
l
l
y
5
,
1
6
1
4
0
7
.
8
2
0
2
N
R
3
.
8
*
N
A
n
o
t
a
v
a
i
l
a
b
l
e
;
N
R
n
o
t
r
e
p
o
r
t
e
d
.
R
a
n
d
o
m
i
z
a
t
i
o
n
o
c
c
u
r
r
e
d
s
u
b
s
e
q
u
e
n
t
t
o
b
a
s
e
l
i
n
e
s
c
r
e
e
n
i
n
g
;
t
h
e
r
e
f
o
r
e
,
t
h
e
s
a
m
p
l
e
s
i
z
e
s
o
f
t
h
e
s
t
u
d
y
a
r
m
s
d
o
n
o
t
s
u
m
t
o
t
h
e
n
u
m
b
e
r
o
f
t
o
t
a
l
e
n
r
o
l
l
e
e
s
.
R
a
n
d
o
m
i
z
a
t
i
o
n
o
c
c
u
r
r
e
d
p
r
i
o
r
t
o
b
a
s
e
l
i
n
e
s
c
r
e
e
n
i
n
g
;
t
h
e
r
e
f
o
r
e
,
t
h
e
t
o
t
a
l
n
u
m
b
e
r
o
f
d
e
a
t
h
s
m
a
y
i
n
c
l
u
d
e
i
n
d
i
v
i
d
u
a
l
s
f
o
u
n
d
t
o
h
a
v
e
l
u
n
g
c
a
n
c
e
r
d
u
r
i
n
g
t
h
e
b
a
s
e
l
i
n
e
s
c
r
e
e
n
.
I
n
c
l
u
d
e
s
3
7
9
c
a
n
c
e
r
d
e
t
e
c
t
e
d
d
u
r
i
n
g
s
c
r
e
e
n
i
n
g
p
e
r
i
o
d
a
n
d
1
7
c
a
n
c
e
r
s
d
e
t
e
c
t
e
d
a
f
t
e
r
t
h
e
e
n
d
o
f
s
c
r
e
e
n
i
n
g
.
tion, initially provided by the International Union
Against Cancer,
28
is that many of these studies may
be invalid because they did not include a no-
screening study arm and thus no determination of
true efficacy could be made. Other authors have
argued that the sample size of these studies was
inadequate. Both the Mayo Lung Project and the
Czechoslovakian studies were powered to detect a
50% reduction in lung cancer mortality in the
screened group compared with the control group.
27
In contrast, the power to detect a smaller but
clinically meaningful effect, such as a 10% reduction
in mortality, was much lower (only 0.21 and 0.16,
respectively).
7,27,2931
Adding to the controversy, case-control studies
have suggested that there might still be a benefit
from CXR screening.
32
The ongoing NCI-sponsored
Prostate, Lung, Colorectal, and Ovarian Trial is a
randomized trial designed to further determine the
role of CXR screening in a low-risk population of
both genders, but the results from this study have
not yet been made available.
33,34
LDCT Scanning
LDCT scanning is a technique that allows a
low-resolution image of the entire thorax to be
obtained in a single breath-holding with low radia-
tion exposure. There is enormous enthusiasm for
LDCT as a screening test. To date, LDCT has only
been evaluated in observational studies of volunteer
cohorts (Table 2).
Studies of LDCT Screening
A prevalence screen using a single-spiral CT scan
in 5,483 persons aged 40 to 74 years was performed
in 1996 in Matsumoto, Japan.
35
Most of the subjects
had undergone yearly CXRs and sputum cytologic
screening, and 3,967 subjects underwent CXR con-
currently with spiral CT scanning. Smokers also
underwent sputum cytology. Among subjects who
had abnormalities prompting further evaluation
were 59 subjects (1%) with noncancerous but suspi-
cious lung lesions, 84 subjects (2%) with lesions
suspicious for lung cancer, and 80 subjects (2%) with
indeterminate small lung nodules. Nineteen lung
cancers were diagnosed (8.5% of subjects with ab-
normal findings prompting further evaluation). The
initial radiographic appearance of these lesions was
suspicious for lung cancer in 14 cases, benign but
suspicious in 3 cases, and indeterminate in 2 cases.
Eighteen of the 19 cases were surgically confirmed,
and 1 case was clinically diagnosed. Pathologic stag-
ing in 16 of 19 tumors (84%) was stage I, while the
remainder (3 of 19 tumors) were stage IV. Four
tumors were 1 cm, and 14 tumors were between
1 cm and 2 cm. Only 1 of the 17 tumors that were
2 cm was seen on CXR. One lung cancer that was
not seen on CT scan was diagnosed by sputum
cytology. Overall, the rate of lung cancer diagnosis
was 0.48%, which was significantly higher than the
rate of 0.03 to 0.05% in the same area prior to spiral
CT screening.
A historical comparison of two screening strategies
performed in Japan within the Anti-Lung Cancer
Association was updated at the annual meeting of
American Society of Clinical Oncology in May
1999.
36,37
The report compared the outcome of
screening with CXR and sputum cytology done from
September 1975 to August 1993 (a total of approxi-
mately 26,000 screenings) to the same screening
strategy plus spiral CT scan done from September
1993 to December 1998. Forty-three patients with
primary lung cancer were found prior to CT scan-
ning, compared to 36 patients with primary lung
cancer in the CT-scan period. In addition, the
percentage of stage IA tumors increased from
42 to 81%, and the 5-year survival improved from 48
to 82%. The results of this nonrandomized, historical
comparison suggest that screening with CT scans can
increase the ability to diagnose lung cancers at
earlier stages but does not constitute strong evidence
of a mortality benefit due to the potential for lead
time and other biases.
The Early Lung Cancer Action Project (ELCAP)
performed yearly low-dose spiral CT scans and CXR
in a single-arm study of 1,000 smokers in New York
City.
38
One to 6 noncalcified lung nodules were
found at baseline (prevalence) in 233 participants;
27 nodules (2.7%) were malignant. Among these 27
malignant nodules, 20 nodules (74%) were not found
on standard CXR; in contrast, no malignant nodules
were detected by CXR that were not also seen on CT
scanning. The majority of these malignancies (23 of
27; 85%) were stage I, and all but one nodule (97%)
were resectable. In the second year of screening
(incidence), seven cancers were detected, and six
of these were stage I. For relevance, in the United
States, only approximately 20% of sporadically diag-
nosed lung cancer is stage I.
39
The findings of these
and related studies are summarized in Table 3. All of
the studies had a high rate of false-positive results
and a preponderance of stage I lesions among found
cancers.
Conclusions: LDCT
It is difficult to determine whether LDCT meets
either of the two criteria of a beneficial screening
T
a
b
l
e
2
L
D
C
T
S
c
r
e
e
n
i
n
g
S
t
u
d
i
e
s
*
S
t
u
d
y
S
i
t
e
D
a
t
e
o
f
P
u
b
l
i
c
a
t
i
o
n
B
a
s
e
l
i
n
e
S
c
r
e
e
n
A
n
n
u
a
l
R
e
p
e
a
t
S
c
r
e
e
n
i
n
g
P
a
t
i
e
n
t
s
S
c
r
e
e
n
e
d
,
N
o
.
A
b
n
o
r
m
a
l
R
e
s
u
l
t
s
,
N
o
.
(
%
o
f
T
o
t
a
l
)
M
a
l
i
g
n
a
n
c
i
e
s
D
e
t
e
c
t
e
d
,
N
o
.
(
%
T
o
t
a
l
)
D
e
t
e
c
t
e
d
M
a
l
i
g
n
a
n
c
i
e
s
S
t
a
g
e
I
,
%
P
a
t
i
e
n
t
s
S
c
r
e
e
n
e
d
,
N
o
.
N
e
w
l
y
I
d
e
n
t
i
f
i
e
d
A
b
n
o
r
m
a
l
R
e
s
u
l
t
s
,
N
o
.
(
%
o
f
T
o
t
a
l
)
M
a
l
i
g
n
a
n
c
i
e
s
D
e
t
e
c
t
e
d
,
N
o
.
(
%
T
o
t
a
l
)
D
e
t
e
c
t
e
d
M
a
l
i
g
n
a
n
c
i
e
s
S
t
a
g
e
I
,
%
H
a
d
a
s
s
a
h
U
n
i
v
e
r
s
i
t
y
,
I
s
r
a
e
l
4
0
2
0
0
1
6
1
4
N
R
4
(
0
.
6
5
)
1
0
0
N
R
N
R
N
R
N
R
C
o
r
n
e
l
l
U
n
i
v
e
r
s
i
t
y
,
U
n
i
t
e
d
S
t
a
t
e
s
3
8
,
4
0
,
4
1
1
9
9
9
,
2
0
0
1
1
,
0
0
0
2
3
3
(
2
3
)
2
7
(
2
.
7
0
)
8
1
1
,
1
8
4
6
3
(
5
)
7
(
0
.
5
9
)
8
5
M
a
y
o
C
l
i
n
i
c
,
U
n
i
t
e
d
S
t
a
t
e
s
4
2
2
0
0
2
1
,
5
2
0
7
8
2
(
5
1
)
2
2
(
1
.
4
)
5
9
1
,
4
6
4
1
9
1
(
1
3
)
3
(
0
.
2
0
)
0
M
o
f
f
i
t
t
C
a
n
c
e
r
C
e
n
t
e
r
,
U
n
i
t
e
d
S
t
a
t
e
s
2
0
0
1
4
9
3
N
R
1
1
(
2
.
2
3
)
1
8
9
1
N
R
0
(
0
)
0
M
u
e
n
s
t
e
r
U
n
i
v
e
r
s
i
t
y
,
G
e
r
m
a
n
y
2
0
0
1
9
1
9
N
R
1
7
(
1
.
8
5
)
7
6
N
R
N
R
2
(
N
R
)
1
0
0
N
a
t
i
o
n
a
l
C
a
n
c
e
r
C
e
n
t
e
r
H
o
s
p
i
t
a
l
,
J
a
p
a
n
4
3
2
0
0
2
1
,
6
1
1
1
8
6
(
1
1
.
5
)
1
4
(
0
.
8
7
)
7
7
7
,
8
9
1
7
2
1
(
9
.
1
)
2
2
(
0
.
2
8
)
8
2
S
h
i
n
s
h
u
U
n
i
v
e
r
s
i
t
y
,
J
a
p
a
n
4
4
2
0
0
1
5
,
4
8
3
6
7
6
(
1
2
)
2
2
(
0
.
4
0
)
1
0
0
8
,
3
0
3
5
1
8
(
6
)
3
4
(
0
.
4
1
)
8
6
H
i
t
a
c
h
i
H
e
a
l
t
h
C
a
r
e
C
e
n
t
e
r
,
J
a
p
a
n
2
0
0
1
8
,
5
4
6
N
R
3
5
(
0
.
4
1
)
9
7
7
,
4
3
4
N
R
7
(
0
.
0
9
)
1
0
0
*
N
R
n
o
t
r
e
p
o
r
t
e
d
.
U
n
p
u
b
l
i
s
h
e
d
p
r
o
c
e
e
d
i
n
g
s
o
f
t
h
e
4
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
S
c
r
e
e
n
i
n
g
f
o
r
L
u
n
g
C
a
n
c
e
r
;
F
e
b
r
u
a
r
y
2
3
2
5
,
2
0
0
1
;
N
e
w
Y
o
r
k
,
N
Y
.
test. Some have argued that the data from the
observational studies are so compelling that we can
conclude that screening with LDCT will be associ-
ated with a reduction in lung cancer mortality (the
first criterion).
40,41,45
Others have argued that the find-
ings in the observational studies may reflect the impact
of biases, rather than a true effect.
30,4650
Whether
LDCT meets the second criterionthat it is not
harmful or painfulis also the subject of a contentious
debate.
Those arguing that LDCT will reduce lung cancer
mortality point to very consistent findings in obser-
vational studies: LDCT detects far more lung can-
cers than CXR, and the vast majority of those
detected are early stage. However, those arguing that
the finding of early stage disease does not necessarily
equate to an improvement in life expectancy raise
two concerns. LDCT may be detecting nodules of
histologic cancer that are often indolent in their
behavior (ie, overdiagnosis), so treating these indi-
viduals cannot increase their life expectancy. Also,
there are questions about whether surgical treatment
of screen-detected cancers alters the natural history
of the disease. Critics addressing the second crite-
rion of a beneficial screening test have also pointed
out that the aggregate harm to screened individuals
(and to society) may exceed the benefit conferred to
those who have lung cancer detected. In this case,
these harms would be due to the high costs of
screening and follow-up, and the morbidity and
anxiety associated with false-positive results. We
review the evidence substantiating each of these
claims below.
Evidence Supporting the Claim That LDCT
May Be Associated With Overdiagnosis
Much of the work on determining whether LDCT
may overdiagnose lung cancer has focused on estab-
lishing whether indolent lung cancer exists. Some
evidence from the randomized trials of CXR and
sputum cytology has been advanced in support of the
hypothesis that indolent lung cancer can be detected
by a screening intervention. To understand this
argument, consider that in a randomized trial of
screening, if all screen-detected cancers grow,
spread, and cause disease, one would expect that an
equal number of cancers would be detected in the
screened and unscreened groups after a sufficiently
long period of follow-up. In the screened group,
cancers would be found at earlier, presymptomatic
stages while cancers in the nonscreened group would
more typically be found at a later, symptomatic stage
(stage shift). In contrast, if screening led to the
identification of cancers that did not grow, spread, or
cause death, these cancers would be detected in the
screened group, but would never manifest with
symptoms and therefore would be missed in the
nonscreened group. The detection of small lesions
Table 3Characteristics of Incident Cancers Detected in RCTs of CXR and Sputum Cytology
Study Site Intervention
Incident Cancers
Early Stage Late Stage Screen Detected Interval Detected
Mass Radiography Service,
London
17,18
CXR every 6 mo for 3 yr 44* 57* 65 36
CXR at end of 3 yr 22* 54* 17 59
Mayo Clinic,
United States
57
CXR and sputum cytology
every 4 mo for 6 yr
99 107 90 116
Advised to have CXR and
sputum cytology annually
51 109 0 160
Research Institute of
Tuberculosis and Respiratory
Diseases, Czechoslovakia
22,27
CXR and sputum cytology
every 6 mo for 3 yr
20 19 31 8
CXR and sputum cytology at
end of 3 yr
10 17 8 19
Memorial Sloan-Kettering
Cancer Center,
United States
21,25
CXR annually and sputum
cytology every 4 mo
50 64 70 44
CXR annually 58 63 65 56
Johns Hopkins,
United States
21,23
CXR annually and sputum
cytology every 4 mo
139 240 186 193
CXR annually
*Excluding cancers detected by the initial screening.
Includes 73 cases detected by symptoms and 43 cases detected by screening outside of study protocol.
Includes 112 cases detected by symptoms and 48 cases detected by screening outside of study protocol.
Includes 15 cases detected by symptoms, 1 case detected by screening outside of study protocol, and 3 cases detected by autopsy.
Not reported by study arm.
Calculated value based on reported proportions among screen-detected and interval cases.
that do not grow, spread, or cause death has been
referred to as overdiagnosis. It is important to note
that a higher percentage of early stage tumors is also
found in the overdiagnosis scenario, so that the
detection of a higher percentage of early stage
tumors in the screened group cannot by itself distin-
guish stage shift from overdiagnosis.
In both the Mayo and Czechoslovakian random-
ized trials of CXR, substantially more cancers were
detected in the screened than the unscreened group.
On extended follow-up, the excess cancers detected
by screening appear to have been overdiagnosed.
Marcus et al
26
reported on follow-up of all cases
diagnosed during the enrollment and study period in
the Mayo study (206 in the screened group and 160
in the control group). Nearly all of the 20% excess
cancers detected in the screened group were early
stage cancers. Failure to detect an equivalent num-
ber of early stage cancers in the control group was
without apparent ill effect, because the control group
experienced no excess number of lung cancer deaths,
even when follow-up extended beyond 17 years for
all participants.
26
Kubik et al
27
presented long-term
follow-up data for the screened and control groups
from the Czechoslovakian study. After the third year
of this study, both the screened and control groups
were entered into a program of annual screening.
Despite the fact that substantially more lung cancers
were detected in the screened group than in the
control group during the first years of the study (year
1 to year 3), the rate of lung cancer detection in the
two groups in the latter years (year 4 to year 6), while
they were being screened in a similar fashion, was
the same. In other words, greater detection of lung
cancer in the screened group early in the study did
not lead to a compensatory fall in detection rates
later on, suggesting that screening was detecting
excess lesions that would not have progressed to
more advanced disease.
27
A number of investigators have also looked to data
from autopsy series for evidence of indolent lung
cancers that could be overdiagnosed by a sensitive
screening test. These investigators argue that if
indolent lung cancer exists, then numerous people
who have died of other causes should have small
nodules of lung cancer found during their autopsies.
A study based at the Yale-New Haven hospital
documented that of all patients who at autopsy had
primary lung cancers, 28% were not diagnosed dur-
ing their lifetime. The overall prevalence of undiag-
nosed lung cancer among all autopsies was 0.09%.
52
A study from Duke University recently suggested
that some lung cancers may not be detected at
postmortem examination. Only 19 of 28 nodules
(68%) seen on CT scanning within 2 months prior to
death were found at autopsy.
52
These findings are
provocative but somewhat hard to interpret. Diag-
nosed lung cancer is relatively rare: the incidence in
50- to 55-year-olds is only 82 per 100,000 (0.08%),
and only rises to about 340 per 100,000 (0.34%) in
75- to 79-year-olds (Surveillance, Epidemiology, and
End Results Program, 1973 to 1998 Public Use
Data).
1
Also, nodules seen by CT scan that were
undetected at autopsy may not have been primary
lung cancers. As a result, it is difficult to know what
rate of discovery would constitute strong evidence
for or against the hypothesis that indolent lung
cancer is a substantial concern in screening.
53
If screen-detected cancers are found at rates in
excess of those expected based on population inci-
dence statistics, or are discovered in patterns incon-
sistent with well established risk factors, this would
suggest that some of the screen detected cancers are
atypical (or overdiagnosed). Evidence that LDCT
is detecting atypical lung cancer can be seen in two
studies. In a large Japanese study of mobile CT
scanning, the rate of detection of lung cancer was
equivalent in smokers and nonsmokers, strongly
contravening the epidemiologic data on sporadically
diagnosed cancer.
44
In the New York-based ELCAP
study, the number of cancers detected in the annual
follow-up LDCT is far less than that detected during
the initial scan, even though the size of the detected
lesions is similar. This difference in rates of detection
is inconsistent with what one would expect if all of
the lesions grew at similar rates (ie, all were aggres-
sive malignancies).
40,41
Instead, that the size of the
prevalent lesions detected at initial scan are consis-
tent with those found at subsequent scans but the
rates of detection are substantially higher suggests
that a meaningful proportion of cancers detected
during the initial scan would have behaved in an
indolent manner.
Evidence Supporting a Limited Impact
of Surgical Resection on Outcome of
Screen-Detected Disease
The intervention universally advocated for early
stage lung cancer is surgical resection.
54
Studies in
which survival is assessed within a standardized
population of individuals with early stage disease
strongly support the efficacy of this intervention.
55,56
However, there are no RCTs in which surgical
resection has been compared to no treatment, and
such a trial will probably never be done. Some
indirect evidence raises doubts about the benefit of
surgery in screen-detected disease. In the Mayo
RCT of CXR, for example, more patients in the
screened group had early stage lung cancer and
underwent surgery, but there was no statistically
significant difference in lung cancer mortality (in
fact, lung cancer mortality rates were marginally
higher in the screened group). In other words,
increased rates of surgery did not confer a benefit to
the group as a whole.
57
Similar results were seen in
the RCT of CXR conducted in Czechoslovakia.
22
Evidence that LDCT May Result in
Aggregate Harm to Screened Individuals
Distinct from the issues of lung cancer mortal-
ity, there are a number of concerns about the
potential harms that could be caused by LDCT if
implemented as a screening tool. First among
these is the identification of large numbers of
individuals with abnormalities on LDCT that do
not require treatment, but may require serial
monitoring and biopsies. These individuals likely
experience anxiety when an abnormality is found
during their screening test for lung cancer. The
likelihood of such events appears high. In the
ELCAP study, 23% of participants had at least one
noncalcified nodule, and only 2.9% had diagnosed
lung cancer, although it should be noted that all of
the individuals who were identified as requiring
biopsies because their lesions were highly suspi-
cious did in fact have histologically confirmed lung
cancer.
41
In a parallel study conducted at the
Mayo Clinic in Rochester, MN, 51% had at least
one noncalcified nodule on CT scan, and only
1.4% had diagnosed lung cancer.
58
In other words,
the apparent false-positive to true-positive ratio is
probably in the range of 10 to 30. A corollary
source of harm may result from some histologically
proven lung cancers being in fact overdiagnosed.
Unnecessary treatments may include thoracotomy,
lung resection, chemotherapy, and radiation.
59
The cost of mass screening with LDCT is
unknown. A large potential population of high-risk
subjects, combined with a high rate of findings
that necessitate follow-up CT scanning, subspe-
cialist consultation, and other testing, all endorse
the expectation of a high total cost. An additional
source of excess cost will come from the fact that
costs of care for individuals with early stage lung
cancer exceed costs of care for individuals with
more advanced stage disease.
60
Whether the
LDCT will be cost-effective, meaning that the
incremental improvement in survival per dollar
spent is reasonable, is a related question that
cannot be answered until effectiveness is estab-
lished. The few published cost-effectiveness anal-
yses on this topic rely on assumptions about
effectiveness that are unproven.
6163
Summary
Sputum cytology screening has been evaluated in
the context of RCTs and does not appear to meet the
criteria of a beneficial screening test. It appears to
detect only a minority of all lung cancers, and does
not appear to reduce lung cancer mortality, although
it has been primarily assessed in combination with
other screening evaluations. More sensitive analyses
of sputum for abnormalities suggestive of malignancy
are being investigated currently. CXR screening
every 4 months or 6 months in combination with
sputum cytology has been assessed in two RCTs, and
CXR alone every 6 months as a single modality was
assessed in a third RCT. In none of these studies was
the control arm a no-screening intervention, and the
power of these studies to detect small reductions in
mortality was limited. Yet, in none of these studies
was CXR associated with a reduction in lung cancer
mortality. CXR is being further evaluated in the
nearly completed Prostate, Lung, Colorectal, and
Ovarian Trial.
LDCT has only been evaluated in the context of
observational studies with volunteer cohorts. The
vast majority of cancers detected by LDCT scanning
in these studies are stage I at the time of discovery.
This suggests that the preponderance of them might
be cured through surgical intervention. However,
evidence from numerous sources casts doubt on the
extent to which these screen-detected lesions repre-
sent aggressive malignancies, and therefore the ex-
tent to which surgical treatment will alter outcome.
Readers should note that the finding of larger num-
bers of early stage cancers also characterizes all of
the RCTs of sputum cytology and CXR, and in none
of these studies was there an alteration in lung
cancer mortality. The impact of LDCT on lung
cancer mortality to date is consequently unknown.
References
1 Ries LAG, Eisner MP, Kosary CL, et al. SEER Cancer
Statistics Review, 19731997. Bethesda, MD: National Can-
cer Institute, 2000
2 Mandel J, Bond J, Church T. Reducing mortality from
colorectal cancer by screening for fecal occult blood. N Engl
J Med 1993; 328:13651371
3 Hardcastle J, Chamberlain J, Robinson M, et al. Randomized
controlled trial of faecal-occult blood screening for colorectal
cancer. Lancet 1996; 348:14721477
4 Kronborg O, Fenger C, Olsen J, et al. Randomised study of
screening for colorectal cancer with faecal-occult blood test.
Lancet 1996; 348:14671471
5 Fontana R, Sanderson DR, Woolner LB, et al. Lung cancer
screening: the Mayo program. J Occup Med 1986; 28:746
750
6 Fontana RS, Sanderson DR, Taylor WF, et al. Early lung
cancer detection: results of the initial (prevalence) radiologic
and cytologic screening in the Mayo Clinic study. Am Rev
Respir Dis 1984; 130:561565
7 Fontana RS, Sanderson DR, Woolner LB, et al. Screening for
lung cancer: a critique of the Mayo Lung Project. Cancer
1991; 67(4 Suppl):11551164
8 Gatta G, Capocaccia R, Hakulinen T, et al. Variations in
survival for invasive cervical cancer among European women,
19781989. Cancer Causes Control 1999; 10:575581
9 Kjellgren O. Mass screening in Sweden for cancer of the
uterine cervix: effect on incidence and mortality. Gynecol
Obstet Invest 1986; 22:5763
10 Gotzsche P, Olsen O. Is screening for breast cancer with
mammography justifiable? Lancet 2000; 355:129134
11 Olsen O, Gotzsche P. Screening for breast cancer with
mammography (Cochrane Review). The Cochrane Library.
Oxford, UK: Update Software, 2002.
12 Charatan F. US panel finds insufficient evidence to support
mammography. BMJ 2002; 324:255
13 Kolata G. Expert panel cites doubts on mammograms worth.
New York Times. January 24, 2002;A16
14 Feuer EJ, Merrill RM, Hankey BF. Interpreting trends in
prostate cancer: part II. Cause of death misclassification and
the recent rise and fall in prostate cancer mortality. Cancer
Surveillance Series. J Natl Cancer Inst 1999; 91:10251032
15 Hankey BF, Feuer EJ, Clegg LX, et al. Interpreting trends in
prostate cancer: part I. Evidence of the effects of screening in
recent prostate cancer incidence, mortality, and survival rates.
Cancer Surveillance Series. J Natl Cancer Inst 1999; 91:1017
1024
16 Black WC, Haggstrom DA, Welch HG. All-cause mortality in
randomized trials of cancer screening. J Natl Cancer Inst
2002; 94:167173
17 Brett GZ. Earlier diagnosis and survival in lung cancer. BMJ
1969; 4:260262
18 Brett GZ. The value of lung cancer detection by six-monthly
chest radiographs. Thorax 1968; 23:414420
19 Boucot KR, Weiss W. Is curable lung cancer detected by
semiannual screening? JAMA 1973; 224:13611365
20 Lilienfeld A, Archer PG, Burnett CH, et al. An evaluation of
radiologic and cytologic screening for the early detection of
lung cancer: a cooperative pilot study of the American Cancer
Society and the Veterans Administration. Cancer Res 1966;
26:20832121
21 Berlin N. Overview of the NCI Cooperative Early Lung
Cancer Detection Program. Cancer 2000; 89(11 Suppl):
23492351
22 Kubik A, Polak J. Lung cancer detection: results of a random-
ized prospective study in Czechoslovakia. Cancer 1986; 57:
24272437
23 Frost J, Ball WC, Levin M, et al. Early lung cancer detection:
results of the initial (prevalence) radiologic and cytologic
screening in the Johns Hopkins Study. Am Rev Respir Dis
1984; 130:549554
24 Tockman MS. Lung cancer screening: the Johns Hopkins
study. Chest 1986; 89(suppl):324324
25 Melamed MR. Lung cancer screening results in the National
Cancer Institute New York study. Cancer 2000; 89(11 Suppl):
23562362
26 Marcus PM, Bergstralh EJ, Fagerstrom RM, et al. Lung
cancer mortality in the Mayo Lung Project: impact of ex-
tended follow-up. J Natl Cancer Inst 2000; 92:13081316
27 Kubik AK, Parkin DM, Zatloukal P. Czech Study on Lung
Cancer Screening: post-trial follow-up of lung cancer deaths
up to year 15 since enrollment. Cancer 2000; 89(11 Suppl):
23632368
28 Prorok PC, Chamberlain J, Day NE, et al. UICC workshop on
the evaluation of screening programmes for cancer. Int J
Cancer 1984; 34:14
29 Eddy DM. Screening for lung cancer. Ann Intern Med 1989;
111:232237
30 Patz EF Jr, Goodman PC, Bepler G. Screening for lung
31 Bailar JC. Screening for lung cancer: where are we now? Am
Rev Respir Dis 1984; 130:541542
32 Okamoto N, Suzuki T, Hasegawa H, et al. Evaluation of a
clinic-based screening program for lung cancer with a case-
control design in Kanagawa, Japan. Lung Cancer 1999;
25:7785
33 Gohagan J, Prorok PC, Hayes RB, et al. The Prostate, Lung,
Colorectal, and Ovarian Cancer Screening Trial of the Na-
tional Cancer Institute: history, organization, and status.
Control Clin Trial 2000; 21(6 Suppl):251S272S
34 Prorok PC, Andriole GL, Bresalier RS, et al. Design of the
Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer
Screening Trial. Control Clin Trial 2000; 21(6 Suppl):273S
309S
35 Sone S, Takashima S, Li F, et al. Mass screening for lung
cancer with mobile spiral computed tomography scanner.
Lancet 1998; 351:12421245
36 Kaneko M, Eguchi K, Ohmatsu H, et al. Peripheral lung
cancer: screening and detection with low-dose spiral CT vs
radiography. Radiology 1996; 201:798802
37 Kakinuma R, Ohmatsu H, Kaneko M, et al. Detection failures
in spiral CT screening for lung cancer: analysis of CT findings.
Radiology 1999; 212:6166
38 Henschke CI, McCauley DI, Yankelevitz DF, et al. Early
Lung Cancer Action Project: overall design and findings from
baseline screening. Lancet 1999; 354:99105
39 Feinstein MB, Bach PB. Epidemiology of lung cancer. Chest
Surg Clin N Am 2000; 10:653661
40 Henschke CI, McCauley DI, Yankelevitz DF, et al. Early
Lung Cancer Action Project: a summary of the findings on
baseline screening. Oncologist 2001; 6:147152
41 Henschke CI, Naidich DP, Yankelevitz DF, et al. Early Lung
Cancer Action Project: initial findings on repeat screenings.
Cancer 2001; 92:153159
42 Swenson SJ, Jett JR, Sloan JA, et al. Screening for lung cancer
with low-dose spiral computed tomography. Am J Respir Crit
Care Med 2002; 165:433434
43 Sobue T, Morigama N, Kaneko M, et al. Screening for lung
cancer with low-dose helical computed tomography: anti-
lung cancer association project. J Clin Oncol 2002; 20:911
920
44 Sone S, Li F, Yang ZG, et al. Results of three-year mass
screening programme for lung cancer using mobile low-dose
spiral computed tomography scanner. Br J Cancer 2001;
84:2532
45 Miettinen OS, Yankelevitz DF, Henschke CI. Screening for
lung cancer [letter]. N Engl J Med 2001; 344:935
46 Jett JR. Spiral computed tomography screening for lung
cancer is ready for prime time. Am J Respir Crit Care Med
2001; 163:812815
47 Patz EFJ, Goodman PC. Low-dose spiral computed tomog-
raphy screening for lung cancer: not ready for prime time.
48 Lung screening study ties up advisors, board sends it back to
NCI for revision. Cancer Lett 2001; 27[26]:14
49 Grannis FWJ. Lung cancer overdiagnosis bias: the gyanousa
am loose! Chest 2001; 119:322323
50 Heffner JE, Silvestri G. CT screening for lung cancer. Am J
Respir Crit Care Med 2002; 165:433434
51 McFarlane MJ, Feinstein AR, Wells CK. Clinical features of
lung cancers discovered as a postmortem surprise. Chest
1986; 90:520523
52 Dammas S, Patz EF Jr, Goodman PC. Identification of small
lung nodules at autopsy: implications for lung cancer screen-
ing and overdiagnosis bias. Lung Cancer 2001; 33:1116
53 Cancer rates and risks. Bethesda, MD: National Cancer
Institute, Cancer Statistics Branch, 1996
54 Ettinger DS, Cox JD, Ginsberg RJ, et al. NCCN (National
Comprehensive Cancer Network) non-small cell lung cancer
practice guidelines. Oncology 1996; 10:S81S111
55 Bach PB, Cramer LD, Warren JL, et al. Racial differences in
the treatment of early-stage lung cancer. N Engl J Med 1999;
341:11981205
56 Flehinger BJ, Kimmel M, Melamed MR. The effect of
surgical treatment on survival from early lung cancer. Chest
1992; 101:10131018
57 Lederle FA, Niewoehner DE. Lung cancer surgery: a critical
review of the evidence. Arch Intern Med 1994; 154:23972400
58 Swenson SJ, Jett JR, Sloan JA, et al. Screening for lung cancer
with low-dose spiral computed tomography. Am J Respir Crit
Care Med 2002; 165:508513
59 Black WC. Overdiagnosis: an underrecognized cause of con-
fusion and harm in cancer screening. J Natl Cancer Inst 2000;
92:12801282
60 Fireman BH, Quesenberry CP, Somkin CP, et al. Cost of care
for cancer in a health maintenance organization. Health Care
Financ Rev 1997; 18:5176
61 Marshall D, Simpson KN, Earle CC, et al. Potential cost-
effectiveness of one-time screening for lung cancer (LC) in a
high risk cohort. Lung Cancer 2001; 32:227236
62 Marshall D, Simpson KN, Earle CC, et al. Economic decision
analysis model of screening for lung cancer. Eur J Cancer
2001; 37:17591767
63 Miettinen OS. Screening for lung cancer: can it be cost-
effective? Can Med Assoc J 2000; 162:14311436
2003;123;72-82 Chest
Peter B. Bach, Michael J. Kelley, Ramsey C. Tate and Douglas C. McCrory
Screening for Lung Cancer: A Review of the Current Literature
& Services
Updated Information
References
#BIBL
free at:
Reprints
2003;123;83-88 Chest
Peter B. Bach, Dennis E. Niewoehner and William C. Black
Screening for Lung Cancer: The Guidelines
ISSN: 0012-3692.
Screening for Lung Cancer*
The Guidelines
Peter B. Bach, MD, MAPP; Dennis E. Niewoehner, MD, FCCP; and
William C. Black, MD
Although virtually all individuals with advanced lung cancer succumb to the disease, a substantial
portion of individuals diagnosed at an earlier stage can be cured. This dichotomy has provoked
interest in lung cancer screening. To date, randomized controlled trials of chest x-ray and sputum
cytology have failed to demonstrate that screening with either modality decreases lung cancer
mortality; neither of these technologies can be recommended. Early studies of lung cancer
screening with low-dose CT (LDCT) appear promising; however, only data from observational
studies are available. We recommend that individuals should only be screened with LDCT in the
context of well-designed clinical trials. (CHEST 2003; 123:83S88S)
Key words: evidence-based medicine; lung neoplasms; mass chest x-ray; mass screening; practice guideline; sputum;
tomography, x-ray computed
Abbreviations: CXR chest x-ray; LDCT low-dose CT; NCI National Cancer Institute; PLCO Prostate, Lung,
Colorectal and Ovarian; RCT randomized controlled trial
L
ung cancer is the number-one cancer killer in
the United States for both men and women,
expected to claim the life of roughly 150,000 people
in 2002. At present, the only therapy that achieves a
high rate of cure is surgical resection of early stage
disease. Hence, screening programs that can in-
crease the rate of detection of early stage lung cancer
make intuitive sense, and have been the subject of
considerable enthusiasm for more than 50 years.
The recommendations we make regarding differ-
ent screening approaches are based on a consider-
ation of the evidence that is reviewed in this article
and the accompanying evidence review. Our recom-
mendations are broadly consistent with those pro-
duced by other organizations when evaluating the
available early detection methods. These other
guidelines are reviewed in Table 1.
Readers should appreciate that for some of these
topics, the evidence is sparse and changing. As new
data become available, these recommendations may
change. The election to screen an individual at risk
for lung cancer should be based on shared, informed
decision making between provider and patient.
These guidelines should complement that process.
CXR
Background
The rationale for CXR screening is based on the
observation that most patients who are diagnosed
with lung cancer have advanced stage disease that
causes them to have symptoms. In contrast, CXR has
sufficient resolution to detect small asymptomatic
nodules that are often stage I disease. As stage I lung
cancer can be treated through surgery, the efficacy
of CXR would be mediated through the detection
of lung cancer at an earlier stage, followed by a cura-
tive intervention such as removal of the cancer-
containing lobe of the lung.
Prior Studies
Three randomized controlled trials (RCTs), one
conducted in London in the 1960s, one conducted at
the Mayo Clinic in Rochester, MN, in the 1970s, and
one conducted in Czechoslovakia in the 1970s, each
evaluated the impact on lung cancer mortality of
regular CXR compared to less frequent CXR.
810
The two latter studies also collected sputum cytology
in conjunction with the CXR but the great majority
*From the Health Outcomes Research Group (Dr. Bach), De-
partment of Epidemiology and Biostatistics, and the Department
of Medicine, Memorial Sloan-Kettering Cancer Center, New
York, NY; the Pulmonary Section (Dr. Niewoehner), Veterans
Affairs Medical Center, Minneapolis, MN; and the Department
of Radiology (Dr. Black), Dartmouth-Hitchcock Medical Center,
Lebanon, and the Department of Community and Family Med-
icine, Center for the Evaluative Clinical Sciences, Dartmouth
Medical School, Hanover, NH.
Correspondence to: Peter B. Bach, MD, MAPP, Health Outcomes
Research Group, Memorial Sloan-Kettering Cancer Center,
1275 York Ave, Box 221, New York, NY 10021; e-mail:
bachp@mskcc.org
of incident lung cancers were independently de-
tected by CXR. In all of these studies, more lung
cancers were detected in the screened group than in
the control group, but there was no discernible
difference in cumulative lung cancer mortality. A
provocative finding in all three studies was that the
Table 1Other Guidelines Related to Screening and Early Detection of Lung Cancer
Recommending Body Topic Grade of Recommendation Recommendations
Society of Surgical Oncology
1
CXR A (strong evidence) Mass screening of high-risk individuals using annual
sputum cytology and CXR failed to improve
ultimate survival from lung cancer, although early
cases can be detected by such screening.
Harvard Community Health
Plan Adult Screening and
Prevention Task Force
2
CXR/sputum cytology A (strong evidence) Routine screening for lung cancer with CXRs or
sputum cytology in asymptomatic adults should
not be performed, regardless of the patients
smoking habits.
3
Lung cancer
screening
C (consensus) The American Cancer Society recommends that, to
the extent possible, individuals at risk for lung
cancer due to current or prior smoking history,
history of significant exposure to second-hand
smoke, or occupational history, be aware of their
continuing lung cancer risk. Those who seek
testing for early lung cancer detection should be
informed about what is currently known about
the benefits, limitations, and risks associated with
conventional and emerging early detection
technologies, as well as the associated diagnostic
procedures and treatment.
3
CT C (consensus) Given the high rate of positive results that occur
with CT screening for lung cancer and the
complexity of the algorithm for working up small
nodules, there is reason to be concerned about
broad dissemination of lung screening outside of
experienced, multispecialty settings and prior to
validation of this new technology. For this
reason, it is critically important during this period
of evolving investigations into the efficacy of
spiral CT and other modalities that appropriate
and influential professional organizations provide
a foundation for best practices based upon the
current state-of-the-art, and also promote
informed decision making for patients about
possible benefits, risks, and limitations of testing
for early lung cancer detection. Individuals
interested in early detection also should be
encouraged to participate in trials.
NCI
4
CXR/sputum cytology A (strong evidence) RCTs have not demonstrated a reduction in lung
cancer mortality resulting from screening with
CXR and/or sputum cytology. The current
evidence does not support lung cancer screening.
United States Preventive
Services Task Force
5
CXR/sputum cytology B (weak evidence) Routine screening for lung cancer with CXR or
sputum cytology in asymptomatic persons is not
recommended.
Society of Thoracic Radiology
6
LDCT C (Consensus) We do not recommend mass screening for lung
cancer at this time, but strongly encourage
appropriate subjects to participate in trials so that
the true effectiveness of lung cancer screening
with LDCT can be determined at the earliest
possible time.
National Cancer Guidance
Group
7
Mass screening A (strong evidence) There is no evidence that population screening for
lung cancer is effective. Even among high-risk
groups, RCTs failed to demonstrate that
screening, by a combination of CXR and sputum
cytology, reduces death rates.
excess cases of lung cancer seen in the screened
group appeared to reflect an increased number of
individuals detected with early stage disease, yet
there were no discernible differences in the number
with advanced stage disease.
Ongoing Studies
The Prostate, Lung, Colorectal and Ovarian
(PLCO) Cancer Screening Trial began recruitment
for its main phase in 1994. The PLCO trial is an RCT
in which 74,000 individuals aged 55 to 74 years will
be screened for prostate, lung, colorectal, and ovar-
ian cancers and followed up for at least 13 years from
randomization. An additional 74,000 individuals will
serve as control subjects receiving routine medical
care. For the lung cancer-screening portion of the
study, smokers will undergo a baseline posteroante-
rior CXR at entry and annual CXR for 3 years,
whereas nonsmokers will undergo only two annual
repeat screenings. The PLCO trial has an 89% power
to detect a 10% reduction in lung cancer mortality.
11
Recommendation
1. For individuals without symptoms or a history
of cancer, we recommend against the use of
serial CXRs to screen for the presence of lung
cancer. Level of evidence, good; benefit, none
or negative; grade of recommendation, D
Remarks
Of the three RCTs of CXR screening, none dem-
onstrated a mortality benefit. Despite the limitations
of these studies, including relatively substantial
crossover and contamination, it is incorrect to con-
clude that these methodologic problems negate the
findings of these studies. In all three RCTs of CXR,
there were clear and consistent differences in the
rates and the stage distribution of lung cancer detec-
tion between the frequently screened and less fre-
quently screened groups, substantiating the claim
that the results of these studies do in fact reflect the
impact of an active intervention. Therefore, although
these studies provide incomplete knowledge about
CXR, they are informative. None of these studies
directly address the utility of a one-time baseline
CXR in high-risk patients, which has perceived but
undocumented value. Further information on the
utility of serial CXR screening is anticipated after the
completion of the PLCO trial.
Studies of Sputum Cytology
Background
The rationale for sputum cytology is based on the
observation that many individuals have cancerous
cells in their sputum at the time of lung cancer
diagnosis. Sputum cytology is more sensitive for
detecting squamous cell carcinomas, which tend to
occur proximally in the bronchial tree, than for
adenocarcinomas, which tend to arise more periph-
erally. Hence, the recent shift in histologic cell type
from predominantly squamous cell to predominantly
adenocarcinoma, noted in epidemiologic studies,
may lessen the value of sputum cytology as a screen-
ing tool.
Prior Studies
Two randomized trials designed to examine the
impact of sputum cytology supplementing CXR were
conducted in the 1970s and 1980s.
12,13
In these
studies, conducted at Johns Hopkins and Memorial
Sloan-Kettering Cancer Center, individuals were
randomized to regular CXR with or without sputum
cytology at 4-month intervals for at least 5 years. In
neither of these studies was there a difference
between the two study arms in the number of lung
cancer cases detected, the percentage that was
resectable, or the lung cancer mortality rates. Since
the intervention (sputum cytology) failed to detect
an appreciable number of new cancers not seen on
CXR, the lack of any mortality benefit might be due
to the insensitivity of the detection method that was
used at the time.
Ongoing Studies
Improvements in detection of shed cancer cells in
sputum samples has been proposed using multiple
techniques, including immunohistochemical detec-
tion of aberrant proteins,
14
computer-assisted image
analysis,
15
detection of genetic alterations,
16,17
and
detection of epigenetic alterations.
18
These en-
hanced techniques are currently under study.
Recommendation
2. For individuals without either symptoms or a
history of cancer, we recommend against the
use of single or serial sputum cytologic evalua-
tion to screen for the presence of lung cancer.
Level of evidence, fair; benefit, none or nega-
tive; grade of recommendation, D
Remarks
At present, there is good evidence from RCTs that
screening with sputum cytology does not appreciably
affect lung cancer detection rates or lung cancer
mortality, although in no case has sputum cytology
been tested in individuals not being actively
screened with CXR. Newer technologies for the
analysis of sputum may markedly improve the sensi-
tivity of this test, which could lead to an alternative
recommendation.
LDCT Scanning
Background
LDCT scanning is a technique that allows a
low-resolution image of the entire thorax to be
obtained in a single breath-hold with low radiation
exposure. The test is very sensitive, and is capable of
routinely detecting nodules as small as 2 to 3 mm in
their greatest diameter. In addition, data obtained
through LDCT and standard CT scans can be used
to reconstruct three-dimensional images that can be
assessed sequentially for evidence of growth. The
rationale for LDCT as an improved early detection
technology is therefore based on this twofold en-
hancement: the ability to detect very small nodules,
and the ability to assess shape and growth patterns
prior to invasive diagnostic tests. The expectation is
that these enhancements will increase the proportion
of individuals who receive diagnoses at an earlier
stage, while also minimizing the number of individ-
uals who undergo unnecessary procedures after an
abnormality is first observed.
Prior Studies
LDCT is a relatively new technology that has only
recently been studied in observational studies.
Therefore, the evidence regarding LDCT is subject
to the biases particular to this type of study. Cur-
rently available studies of LDCT demonstrate four
phenomena. First, for individuals screened with
LDCT and CXR, LDCT detects a far greater num-
ber of lung cancers than does CXR. Second, the vast
majority of lung cancers detected by LDCT are stage
I. Third, LDCT detects many more noncancerous
than cancerous nodules. Fourth, the use of serial CT
scans and three-dimensional reconstruction appears
to lessen the number of invasive procedures per-
formed on individuals who have an abnormality, but
do not have lung cancer. No study has reported on
survival outcomes for individuals who have screen-
detected cancer. These observational data, coupled
with concerns about overdiagnosis and inefficacy of
treatment for screen detected disease, limit our
ability to determine the potential efficacy of the
intervention.
Ongoing Studies
At present, large observational and randomized
studies of LDCT are either planned or underway. A
large, New York-based LDCT study has begun, and
is funded to enroll 10,000 current or former smokers
for annual screening. The study should yield useful
information on the frequency of abnormal test re-
sults, the diagnostic workup of patients with abnor-
malities, and the frequency of unnecessary proce-
dures and interval-diagnosed cancers. In addition,
after six PLCO sites successfully randomized
3,000 individuals to LDCT or CXR screening in 2
months (the Lung Screening Study), the National
Cancer Institute (NCI) approved a $200 million
dollar study in which 50,000 individuals with a
smoking history will be randomized to annual
screening with LDCT or CXR at approximately 10
PLCO sites and approximately 20 American College
of Radiology Imaging Network sites.
19
This study is
expensive and could take many years to complete,
but it should yield useful answers. It is designed to
have a 90% power to detect a mortality reduction of
20% and should be completed in 2009.
Recommendation
of cancer, we recommend against the use of a
single LDCT or serial LDCTs to screen for the
presence of lung cancer. At-risk individuals
who express an interest in undergoing LDCT
screening should be made aware of several
ongoing high-quality clinical studies of this
technology. Level of evidence, poor; benefit,
none or negative; grade of recommendation, I
Remarks
Although studies of LDCT based on observational
designs appear promising, in that LDCT detects a
preponderance of early stage lesions, a similar pat-
tern accompanied the early studies of CXR and
sputum cytology. The fact that prior randomized
studies of CXR and sputum cytology, related autopsy
series, and preliminary findings in LDCT studies all
raise concerns that some cancers detected by LDCT
are overdiagnosed elevates the importance of proper
evaluation of the technology. In addition, concerns
about false-positive results and unnecessary treat-
ment raise the possibility that even if LDCT leads to
an improvement in lung cancer mortality through
early detection, the test may in aggregate lead to
greater harm than benefit.
As such, LDCT should be considered an experi-
mental procedure that requires evaluation in the
context of well-designed studies. Recently, the NCI
approved the funding of a randomized study of
LDCT that will accrue patients at approximately 10
of the NCI PLCO sites and approximately 20 sites
designated by the American College of Radiology
Imaging Network. In addition, numerous observa-
tional studies are underway. The NCI study is
designed to estimate the efficacy of the test in
lowering lung cancer mortality. Ongoing observa-
tional studies may provide supplemental insights
about LDCT in terms of costs, frequency of
follow-up tests, and rates of complications, but it is
unlikely that they will in isolation answer the
paramount question of efficacy. Whether random-
ized or uncontrolled:
(1) Studies should be designed to evaluate high-
risk subjects only, such as individuals 60 years old
with at least a 30 pack-year smoking history. If
individuals have quit smoking, it should have oc-
curred no more than 10 years prior to enrollment.
Studies that enroll low-risk subjects are unlikely to
provide useful information about efficacy, and may
cause harm during follow-up of findings.
(2) Screening of subjects should be conducted at
regular intervals (such as annually). Without longitu-
dinal screening and extended follow-up, judgments
about overdiagnosis and health-care events cannot
be made. As a corollary, specific efforts should be
made in the event of loss to follow-up.
(3) It is vital that all study subjects maintain a
record of health-care events occurring after screen-
ing visits. Of particular interest are interval-
diagnosed cases of lung cancer and health-care
encounters spurred by findings at LDCT. Without
this information, neither the extent to which LDCT
misses highly aggressive lung cancers, nor the cost
and burden of positive findings can be assessed.
Conclusion
The most effective treatment for lung cancer
remains surgical resection of early stage disease;
however, sporadic lung cancer is rarely diagnosed in
its earliest stages. The promise of screening tech-
niques for increasing rates of early stage lung cancer
detection, and thus the expectation of more treatable
cases, has driven considerable research and ongoing
development of screening technologies. RCTs of
CXR and sputum cytology have failed to demon-
strate a mortality benefit for either technique, and
we do not recommend screening with serial CXR or
sputum cytology for asymptomatic individuals or
individuals without a history of cancer. LDCT scan-
ning is a promising technology due to its sensitivity
and ability to assess growth of nodules, and ongoing
studies may provide additional information about the
costs and benefits of screening with this technology.
However, due to the absence of evidence regarding
mortality and concerns about overdiagnosis, we rec-
ommend against screening with LDCT for individu-
als without symptoms or a history of cancer. As
further research enhances our understanding of the
risks and benefits, these recommendations may
change.
Recommendations
of cancer, we recommend against the use of
serial CXRs to screen for the presence of lung
cancer. Level of evidence, good; benefit, none
or negative; grade of recommendation, D
2. For individuals without either symptoms or a
history of cancer, we recommend against the
use of single or serial sputum cytologic evalua-
tion to screen for the presence of lung cancer.
Level of evidence, fair; benefit, none or nega-
of cancer, we recommend against the use of a
single LDCT or serial LDCTs to screen for the
presence of lung cancer. At-risk individuals
who express an interest in undergoing LDCT
screening should be made aware of several
ongoing high quality clinical studies of this
technology. Level of evidence, poor; benefit,
none or negative; grade of recommendation, I
References
practice guidelines: Society of Surgical Oncology practice
guidelines. Oncology 1997; 11:889892,895
2 Lawrence T. Screening for lung cancer. HCHP Adult Screen-
ing and Prevention Task Force. Harvard Community Health
Plan Clinical Guidelines and Algorithms. Boston, MA: Har-
vard Community Health Plans, 1989
3 Smith R, von Eschenbach A, Wender R, et al. American
Cancer Society guidelines for the early detection of cancer:
update of early detection guidelines for prostate, colorectal,
and endometrial cancers. Also: update 2001; testing for early
lung cancer detection. CA Cancer J Clin 2001; 51:3875
4 National Cancer Institute. Screening for lung cancer (PDQ).
National Cancer Institute CancerNet Web site. 2001. http://
www.cancer.gov/cancerinfo/pdq/screening/lung/healthprofes-
sional. Accessed December 17, 2002
5 Screening for lung cancer. In: United States Preventive
Services Task Force, ed. Guide to clinical preventive services.
Baltimore, MD: Williams & Wilkins, 1996; 135139
6 Aberle DR, Gamsu G, Henschke CI, et al. A consensus
statement of the Society of Thoracic Radiology: screening for
lung cancer with helical computed tomography. J Thorac
Imaging 2001; 16:6568
7 National Cancer Guidance Group. Improving outcomes in
lung cancer: the manual. NHS Executive, Department of
Health, 1998 London
8 Brett GZ. The value of lung cancer detection by six-monthly
chest radiographs. Thorax 1968; 23:414420
9 Fontana R, Sanderson DR, Woolner LB, et al. Lung cancer
screening: the Mayo program. J Occup Med 1986; 28:746
750
10 Kubik A, Polak J. Lung cancer detection: results of a random-
ized prospective study in Czechoslovakia. Cancer 1986; 57:
24272437
11 Prorok PC, Andriole GL, Bresalier RS, et al. Design of the
Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer
Screening Trial. Control Clin Trials 2000; 21(6 Suppl):273S
309S
12 Tockman MS. Lung cancer screening: the Johns Hopkins
study. Chest 1986; 89(suppl):324
13 Melamed MR. Lung cancer screening results in the National
Cancer Institute New York study. Cancer 2000; 89(11 Suppl):
23562362
14 Tockman MS, Mulshine JL. Sputum screening by quantita-
tive microscopy: a new dawn for detection of lung cancer?
Mayo Clin Proc 1997; 72:788790
15 Payne PW, Sebo TJ, Doudkine A, et al. Sputum screening by
quantitative microscopy: a reexamination of a portion of the
National Cancer Institute Cooperative Early Lung Cancer
Study. Mayo Clin Proc 1997; 72:697704
16 Mao L, Hruban RH, Boyle JO, et al. Detection of oncogene
mutations in sputum precedes diagnosis of lung cancer.
Cancer Res 1994; 54:16341637
17 Miozzo M, Sozzi G, Musso K, et al. Microsatellite alterations
in bronchial and sputum specimens of lung cancer patients.
Cancer Res 1996; 56:22852288
18 Palmisano WA, Divine KK, Saccomanno G, et al. Predicting
lung cancer by detecting aberrant promoter methylation in
sputum. Cancer Res 2000; 60:59545958
19 Lung screening study ties up advisors, board sends it back to
NCI for revision. Cancer Lett 2001; 27[26]:14
2003;123;83-88 Chest
Peter B. Bach, Dennis E. Niewoehner and William C. Black
Screening for Lung Cancer: The Guidelines
& Services
Updated Information
References
#BIBL
at:
Citations
#otherarticles
Reprints
2003;123;89-96 Chest
Bethany B. Tan, Kevin R. Flaherty, Ella A. Kazerooni and Mark D. Iannettoni
The Solitary Pulmonary Nodule
ISSN: 0012-3692.
The Solitary Pulmonary Nodule*
Bethany B. Tan, MD; Kevin R. Flaherty, MD; Ella A. Kazerooni, MD; and
Mark D. Iannettoni, MD, FCCP
More than 150,00 patients a year present to their physicians with the diagnostic dilemma of a
solitary pulmonary nodule (SPN) found either on chest radiography or chest CT. A thoughtful and
timely workup of this finding is essential if lung cancer is to be recognized early and the chance
for cure optimized. Based on the literature to date, recommendations are made for appropriate
imaging modalities and diagnostic testing, as well as indications for obtaining preoperative tissue
diagnosis for the patient with an SPN. (CHEST 2003; 123:89S96S)
Key words: diagnostic workup; lung cancer; malignancy; solitary pulmonary nodule
Abbreviations: CXR chest radiograph; FDG 18-fluorodeoxyglucose; NSCLC non-small cell lung cancer;
PET positron emission tomography; SPN solitary pulmonary nodule; TTNA transthoracic needle aspiration
A
solitary pulmonary nodule (SPN) is radiologi-
cally defined as an intraparenchymal lung lesion
that is 3 cm in diameter and is not associated with
atelectasis or adenopathy.
1
Lung lesions 3 cm in
size are defined as lung masses. One of 500 chest
radiographs (CXRs) demonstrates a lung nodule.
Ninety percent of these are incidental radiologic
findings, found unexpectedly in radiographs ob-
tained for unrelated diagnostic workups. More than
150,000 patients per year in the United States
present their physicians with the diagnostic dilemma
of an SPN. This number has increased even further
due to incidental findings of lung nodules on chest CT.
2
The tragedy of lung cancer is directly associated
with its delayed presentation. Signs and symptoms
are rarely present until the malignancy has become
advanced and possibly unresectable. Patients with
the best prognosis are those found to have stage IA
(T1N0M0) disease. These patients have a 61 to 75%
5-year survival following surgical resection.
3,4
Unfor-
tunately, approximately one half of all lung cancers
have extrapulmonary spread at the time of diagnosis.
As a result, the average patient with a diagnosis of
lung cancer has a 5-year survival of only 10 to 15%.
5
Therefore, a timely and accurate diagnosis of the
etiology of an SPN is essential to providing the
patient with malignancy a potential for cancer cure.
The occult nature of a lung nodule with its few
symptoms and inability to be detected on physical
examination does not lend itself to the sense of
immediacy prompted by the discovery of other
potential malignancies, such as a breast mass.
Differential Diagnosis
The differential diagnosis of an SPN includes
neoplastic, infectious, inflammatory, vascular, trau-
matic, and congenital lesions.
2
Other benign etiolo-
gies for SPNs are rheumatoid nodules, intrapulmo-
nary lymph nodes, plasma cell granulomas, and
sarcoidosis. Although most SPNs are benign,
2,5
pri-
mary malignancy may be found in approximately
35% of SPNs, and solitary metastases can account for
another 23%.
68
Clinical characteristics such as older
age, a history of cigarette smoking, and a previous
history of cancer all increase the probability that an
SPN is malignant.
9
Radiologic characteristics (dis-
cussed below) can also influence the probability of
malignancy. Bayes theorem, logistic regression mod-
els, and neural network analysis have all been devel-
oped in an attempt to use both patient history and
nodule appearance to accurately predict the likeli-
hood of malignancy.
912
While these represent laud-
able efforts, they have proven to be cumbersome and
of little practical use to the clinician evaluating a
patient with an SPN. In general, all SPNs should be
considered malignant until proven otherwise.
13
Radiologic Diagnostics: CXRs
Since the SPN is by definition a radiographic
finding, radiologic imaging is intrinsic to the diag-
nostic workup. Essentially all SPN are found on
*From the University of Michigan Section of Thoracic Surgery,
Ann Arbor, MI.
Correspondence to: Mark D. Iannettoni, MD, FCCP, University
of Michigan Section of Thoracic Surgery, 2120 Taubman Center,
Box 0334, 1500 E Medical Center Dr, Ann Arbor, Michigan
48109; e-mail: mdi@umich.edu
CXRs or incidentally on CT. The CXR is an excellent
initial imaging study for patients with symptoms, or
as routine follow-up for patients with known pulmo-
nary lesions. The indications and efficacy of CXRs
for routine cancer surveillance are beyond the scope
of this chapter and are discussed in the chapter on
Follow-up and Surveillance in these guidelines.
CXRs are inexpensive, readily available, quickly ob-
tained, and can be read by both the clinician and the
radiologist.
However, the diagnosis of lung cancer from CXRs
alone can be quite difficult. The failure to recognize
lung cancer on the CXR is one of the most frequent
causes of missed diagnosis in radiology.
14
The rate of
failure to diagnose lung cancer from CXRs varies
from 25 to 90% in a number of different studies with
differing study designs.
1517
In the radiologic litera-
ture, an error rate of 20 to 50% for radiologic
detection of lung cancer is generally accepted.
18
If an
SPN is missed on the CXR, the delay in diagnosis can
be substantial. Quekel et al
19
looked at CXRs retro-
spectively in 259 patients with proven non-small cell
lung cancer (NSCLC) and found a 19% incidence of
missed diagnoses. Those patients with missed lesions
had significantly smaller nodules (median diameter
16 mm), more superimposing structures, and more
indistinct border edges on CXRs than those SPNs
that were correctly identified. The delay in diagnosis
from the time of initial radiologic appearance was
also significant at 472 days vs 29 days. This resulted
in 43% of lesions being upstaged from T1 to T2
lesions during the delay period.
19
Traditionally the presence of benign calcification
or the absence of growth over a 2-year time period
has been believed to be reliable indicators of benign
disease. These criteria have been known since the
1950s.
20,21
In the 50 years since, we have learned that
there are really no other characteristics that can
consistently differentiate a benign nodule from a
malignant nodule based on appearance on the CXR.
Therefore, for patients with an SPN that is visible on
the CXR, all previous CXRs should be reviewed. For
all patients with previous CXRs, an SPN that is
unchanged for 2 years does not require further
diagnostic evaluation.
Benign calcification refers to central, diffuse, lam-
inar, or popcorn patterns.
2
Other types of calcifica-
tions such as eccentric or stippled calcifications are
radiologically indeterminant and are seen in both
benign and malignant lesions. CXRs may also falsely
suggest that calcium is present, leading the clinician
and patient to have false confidence that the nodule
is benign. In a recent series by Berger et al,
22
7% of
nodules that were believed to be definitely calci-
fied by the CXR lacked calcium on the CT scan.
The growth rate of a lesion may also be an
unreliable predictor of a benign nodule. Benign
lesions typically have a doubling time of either 1
month or 16 months.
23
Malignant nodules have a
doubling time from anywhere from 40 to 360 days.
24
The CXR is also less sensitive than CT for detecting
changes in size of an SPN, as a doubling in spherical
tumor volume may result in a change in diameter of
only a few millimeters.
The morphology of a nodule that is spherical with
rounded edges is associated with benign disease.
However, 20 to 34% of SPNs with this appearance
are malignant, most notably those nodules that rep-
resent metastatic disease.
25,26
Recommendations
1. For patients with an SPN that is visible on the
CXR, all previous CXRs should be reviewed.
Level of evidence, poor; benefit, substantial;
grade of recommendation, C
2. For all patients with previous CXRs, an SPN
that is unchanged for 2 years does not re-
quire further diagnostic evaluation. Level of
evidence, fair; benefit, substantial; grade of
recommendation, B
3. For patients with an SPN visible on the CXR in
which benign central calcification is present, no
further diagnostic evaluation is necessary.
Radiographic Diagnostics: Chest CT
Spiral CT with IV contrast enhancement is the
imaging modality of choice for the SPN and should
be obtained on all newly diagnosed SPNs. CT pro-
vides ideal imaging for characterizing the nodule and
its location. The CT scan can also be used to identify
synchronous lung lesions or metastatic liver or adre-
nal lesions or mediastinal lymph nodes. Chest CT is
also helpful for assessment of chest wall, mediastinal,
or diaphragmatic invasion or for evaluation of supe-
rior sulcus (Pancoast) tumors. Of course, by strict
definition the SPN does not invade the chest wall,
mediastinum, or superior sulcus.
CT has a 50% sensitivity and 89% specificity for
detecting mediastinal invasion and 14% sensitivity
and 99% specificity for identifying chest wall inva-
sion,
27
and it is well demonstrated to be more
sensitive than the CXR in characterization of the
nodule and the mediastinum. Studies demonstrate
that MRI has a similar sensitivity and specificity for
evaluation of the mediastinum and chest wall.
28,29
CT evaluation of mediastinal adenopathy has a wide
range of reported sensitivities and specificities in the
literature. Shea and Lillington
29
of the Lung Cancer
Study Group reviewed this literature on CT and
MRI mediastinal evaluation and found a CT sensi-
tivity of 70 to 90% and a specificity of 60 to 90%. If
mediastinal adenopathy was not appreciated on CT,
there was only a 15% chance of finding positive N2
disease at the time of surgery. Mediastinal detection
of N2 disease was similar for CT and MRI. If
mediastinal adenopathy is visualized radiographi-
cally, the most important role of CT is to provide a
road map for further procedures that will give a
tissue diagnosis.
28
A number of benign etiologies for SPNs have a
characteristic appearance on CT. The nodule con-
taining a fat density can be classified as a hamartoma
with confidence. Arteriovenous fistulas demonstrate
presence of a feeding artery and a draining vein as
well as contrast enhancement on CT. Rounded
atelectasis is associated with a dense comet tail sign
on CT. A fungus ball can be identified as an SPN
within a cavity. Pulmonary infarcts may be charac-
terized on CT as a wedge shape abutting the pleura
with air bronchograms.
The patient with a new finding of an SPN and a
recent history of pneumonia or pulmonary symptoms
may warrant following the lesion for 4 to 6 weeks to
rule out an infectious etiology. However, persistence
of the nodule in such a patient should not further
delay the diagnostic workup. In a 3-year retrospec-
tive study, 1% of all SPNs were found to have an
infectious etiology.
30
This incidence may be some-
what higher in regions endemic for fungal infections
or tuberculosis.
Malignant pulmonary nodules may be ill defined
with irregular margins and spiculated borders. In
fact, 84 to 90% of spiculated nodules are malig-
nant.
25,26
The size of a lung nodule is also a good
indicator of the likelihood of malignancy. The vast
majority of nodules 2 cm in size are malignant,
compared to a 50% rate of malignancy in all nodules
2 cm in size.
31
The incidence of malignancy in a
lung lesion 3 cm is so great that all these lesions
should be surgically resected unless medically con-
traindicated. Air bronchograms and pseudocavitation
are characteristics seen on CT imaging that are more
common in malignant (30%) than benign (5%) le-
sions.
25
Cavitation of a nodule is also indicative of
malignancy, but inflammatory and infectious disease
may also present with this morphology. In these
situations, wall thickness can further aid in determin-
ing the probability that an SPN is benign or malig-
nant. Woodring and Fried
32
found that 95% of all
nodules with a wall 5 mm were benign in origin,
84% of all cavitated lesions with a wall 15 mm in
thickness were malignant, and 73% of nodules with a
wall thickness of 5 to 15 mm were benign.
For those SPNs with indeterminant morphology,
IV contrast enhancement with helical CT imaging
may be a helpful adjunct. Swenson et al
33
found
nodules enhancing to 20 Hounsfield units to be a
predictive feature of malignancy while contrast en-
hancement 15 Hounsfield units was characteristic
of benignancy with a sensitivity of 98%, specificity of
73%, and 85% accuracy.
Recommendation
4. For patients with an SPN, spiral CT of the
chest with contrast is indicated to better char-
acterize the nodule, parenchyma, and medias-
tinum. CT can be useful in identifying nodules
more likely to be benign and obviate the need
for further diagnostic evaluation. Additionally,
chest CT plays an important role in staging (as
delineated in the chapter on Noninvasive Stag-
ing elsewhere in these guidelines). Level of
evidence, good; benefit, moderate; grade of
recommendation, B
Radiologic Diagnostics: MRI
MRI has a very limited role in the evaluation of the
SPN. It may be beneficial in the patient who cannot
tolerate IV contrast. MRI may also allow better
anatomic evaluation of the lung apices, thoracic inlet,
chest wall, or diaphragm due to its ability to provide
sagittal, coronal, and oblique images. In general, the
cost of MRI is not worth the lower risk of contrast-
induced toxicity for most patients, as the imaging
accuracy of CT is as good for most locations of SPNs.
With the exception of special instances, MRI is not
indicated for the routine workup of the SPN. Addi-
tionally, the cost of MRI is not worth the lower risk
of contrast-induced toxicity for most patients be-
cause the imaging accuracy is at least as good as CT
for most locations of lesions.
Recommendation
5. For patients with an SPN, MRI is not indicated
except in these special instances. Level of
evidence, good; benefit, none; grade of recom-
mendation, D
Radiologic Diagnostics: Positron Emission
Tomography
Positron emission tomography (PET) with 18-
fluorodeoxyglucose (FDG) has proven to be an
excellent mode of tumor imaging. FDG is taken up
by cells in glycolysis but is bound within these cells
and cannot enter the normal glycolytic pathway.
Increased activity is demonstrated in cells with high
metabolic rates, as is seen in tumors and areas of
inflammation. Gould et al
34
performed a meta-
analysis of the literature on pulmonary nodules and
masses and PET scanning and found 40 good studies
with an overall sensitivity of 96.8% and specificity of
77.8% for detecting malignancy. PET scans also have
a 96% sensitivity and 88% specificity with 94%
accuracy in the diagnosis of benign nodules. High
diagnostic accuracy for detecting tumor also makes
PET more accurate than CT for detecting mediasti-
nal lymph node metastases and distant metastases.
Indeed, T1 lung cancer may have up to a 21%
incidence of regional lymph node metastases.
35,36
The spatial resolution of PET is currently 7 to 8 mm,
and so the imaging of SPNs 1 cm is unreliable with
the current generation of PET scanners and should not
be performed. PET with FDG may also give false-
negative results for nodules that are carcinoid tumors
or bronchoalveolar carcinomas, as these tumors may
not have high FDGuptake.
37
False-positive results may
be seen in lung lesions with an infectious or inflamma-
tory etiology, such as tuberculosis, histoplasmosis, or
rheumatoid nodules. PET had been available only in
large academic centers but is now becoming much
more accessible. PET scans are more expensive than
other imaging modalities, with Medicare reimburse-
ment of $1,912 compared to $276 reimbursement for
chest CT or $560 reimbursement for transthoracic
needle aspiration (TTNA).
38
PET scan is not only an excellent imaging study for
tumor, but it can potentially change patient manage-
ment by detecting unsuspected nodal and metastatic
disease. Therefore, the question regarding when to
include PET scans as part of the workup of the SPN
is not one of diagnostic accuracy but of when clinical
decision making will be changed by its findings and
warrants the cost of the study. For low-risk patients
with a pretest likelihood of malignancy of only 20%,
the posttest likelihood of malignancy with a negative
finding PET is 1%.
34
The high negative predictive
value of PET in this patient population would sup-
port observation for the SPN with a negative PET
finding. However, high-risk patients with a pretest
likelihood of malignancy of 80% still have a 14%
posttest likelihood of malignancy with a negative
PET finding.
34
The patient with high risk of malig-
nancy should have a tissue diagnosis of the SPN, and
the only question should be the most efficient means
of obtaining this tissue diagnosis. There is no indica-
tion for PET in the workup of an SPN with a
negative mediastinal evaluation on CT if operative
intervention is definitely planned or if it will other-
wise not change patient management. Likewise,
there is no indication for PET in a patient with a
known malignancy who has a questionable pulmo-
nary metastasis vs lung cancer primary tumor.
Recommendations
6. For patients with an SPN 1 cm in size, PET
scanning is not currently recommended. Level
of evidence, good; benefit, none/negative;
grade of recommendation, D
7. For patients with an SPN who are surgical
candidates and have a negative mediastinal
evaluation on CT, PET scanning with FDG as
an investigational tool, where available, may be
warranted. Level of evidence, fair; benefit,
moderate; grade of recommendation, B
8. For patients with an SPN who are marginal
surgical candidates, if PET with FDG results
are negative, a repeat CT scan is required at
least once in 3 months. Level of evidence,
good; benefit, substantial; grade of recommen-
dation, A
9: For patients with an SPN who are marginal
surgical candidates, if there are unchanged
results from prior CXRs and negative PET scan
findings, serial follow-up is recommended, con-
sisting of an initial CXR, and CT scanning at 3,
6, 12, and 24 months. Level of evidence, fair;
benefit, substantial; grade of recommendation, B
Tissue Diagnosis: TTNA
Obtaining a tissue diagnosis via TTNA is some-
what less invasive than bronchoscopy and Wang
needle biopsy and does not require IV sedation.
Certainly, it is much less invasive than surgery, but a
nonmalignant diagnosis may not be believed and
TTNA and bronchoscopy can at best only be diag-
nostic, not therapeutic. The sensitivity for malig-
nancy is 64 to 100%.
39,40
Unfortunately, the sensitiv-
ity of TTNA for a specific benign diagnosis is 12 to
68% but only 12% in a number of studies.
41
Adding
automated cutting (core needle biopsy) to TTNA
may increase the yield of a specific diagnosis of
benign disease from 12 to 75%.
42
Yield is also
increased by having an on-site pathologist to assess
the quality of biopsy samples at the time of the
procedure.
43
TTNA is contraindicated in the patient
with a single lung. Relative contraindications to this
procedure are the patient with pulmonary hyperten-
sion, coagulopathy or a bleeding diathesis, severe
COPD, or vascular malformations. The most fre-
quent complication of TTNA is pneumothorax in 25
to 30% of patients, with 5 to 10% of these patients
requiring a chest tube. Pneumothorax is decreased
by avoiding crossing pulmonary fissures and multiple
punctures of the lung parenchyma. There can be up
to a 10% incidence of hemoptysis and hemorrhage,
which is increased by the use of cutting needles. Air
embolus and tumor seeding are rare, 0.1% and
0.05% respectively.
44
Recommendations
10. For patients with an SPN who are operable
candidates, TTNA is not indicated. Level of
evidence, good; benefit, none; grade of rec-
ommendation: D
11. For operable patients with an SPN who de-
cline surgical intervention, TTNA or trans-
bronchial needle biopsy is the preferred pro-
cedure for establishing a diagnosis. Level of
recommendation, B
12. For patients with an SPN who are not opera-
ble candidates, or are at high risk, TTNA may
be helpful to establish tissue diagnosis. Level
of evidence, good; benefit, moderate; grade of
recommendation, B
Tissue Diagnosis: Bronchoscopy
Bronchoscopy may be an good approach for ob-
taining a tissue diagnosis in the large, central lung
mass or in those with endobronchial encroachment,
as diagnostic yield is 70% and 90%, respectively.
4549
The best application of bronchoscopy and TBNA is
for staging of NSCLC by aspirating enlarged medi-
astinal lymph nodes. The discovery of metastatic
disease will change patient management and obvi-
ates any further surgical staging. However, for the
patient with a peripheral lung nodule, there is little
role for bronchoscopy. Although the reported diag-
nostic yield of 40 to 80% for peripheral lesions is
surprisingly high, these reports are primarily from
centers that employ routine use of fluoroscopy and
multiple sampling methods.
46
Bronchoscopic diag-
nostic yield is proportional to the size of the lung
lesion. In the evaluation of the SPN, bronchoscopy
has been shown to provide no measurable preoper-
ative benefit to the patient, as it does not obviate the
need for surgery.
49
Recommendation
13. In patients with an SPN, bronchoscopy is
usually not indicated. Level of evidence, good;
benefit, none; grade of recommendation, D
Surgery
The patient with an SPN that is new and does not
have benign appearing calcifications should be con-
sidered to have a malignancy until proven otherwise.
Surgical resection is the ideal approach, as it is both
diagnostic and therapeutic. If it is believed that
based on patient history that the SPN may not be
NSCLC but rather metastatic disease, then thoracos-
copy and wedge resection is an accepted initial
surgical approach. The specimen should be sent for
frozen section, so that conversion to a thoracotomy
and lobectomy can be performed in the same setting
should the nodule prove to be NSCLC. Localization
techniques such as methylene blue dye injection and
wire localization may assist thoracoscopic resection
of small nodules or those not in the lung periph-
ery.
50,51
Suzuki et al
52
suggest that nodules 1 cm or
5 mm from the nearest pleural surface should
have preoperative localization to optimize thoraco-
scopic resection. For the surgical candidate with an
SPN proven to be NSCLC, lobectomy and system-
atic mediastinal lymph node dissection is the stan-
dard of care for complete oncologic resection and
staging.
53
Five-year survival following complete re-
section of stage 1A or 1B NSCLC is 65 to 80% and
50 to 60%, respectively. For the patient who is a
marginal surgical candidate and whose pulmonary or
cardiac status would benefit from a limited resection,
wedge resection or segmentectomy is acceptable
for treatment of NSCLC. Warren and Faber
54
demonstrated similar long-term survival for patients
who had segmentectomy vs lobectomy for stage I
NSCLC, but overall local recurrence was 23% vs 5%.
In the only prospective trial, performed by the Lung
Cancer Study Group, Ginsberg and Rubinstein
55
reported a local recurrence rate for segmentectomy
or wedge resection of T1N0 NSCLC three times
greater than that for lobectomy, but long-term sur-
vival was not as impressively decreased. Because of
the greatly increased rate of recurrence, patients
who have a limited resection require close postoper-
ative surveillance.
Recommendations
14. In operable patients with an SPN, if the lesion
is amenable to a wedge resection, then a
wedge resection is the procedure of choice
followed by a lobectomy if the pathologic
finding is positive for cancer. Level of evi-
dence, fair; benefit, substantial; grade of rec-
ommendation, B
15. In operable patients with an SPN, if the lesion
is not amenable to a wedge resection, a diag-
nostic lobectomy is acceptable. Level of evi-
dence, good; benefit, substantial; grade of
recommendation, A
16. All pulmonary resections, anatomic or non-
anatomic, must include a systematic lymph
node dissection. Level of evidence, good; ben-
efit, substantial; grade of recommendation, A
surgical candidates, a wedge resection or seg-
mentectomy is acceptable. Level of evidence,
fair; benefit, substantial; grade of recommen-
dation, B
Follow-up
The patient with an SPN who does not have a
tissue diagnosis and who is deemed acceptable for
observation should be followed up closely for a
minimum of 2 years. This should include an initial
CXR, and CT scanning at 3, 6, 12, and 24 months for
best monitoring for nodule growth. There is very
little objective evidence for frequency of surveillance
monitoring.
Recommendation
18. For patients with an SPN without a definitive
tissue diagnosis, a minimum follow-up of 2
years is recommended. This should include an
initial CXR, and CT scanning at 3, 6, 12, and
24 months. Level of evidence, poor; benefit,
moderate; grade of recommendation, C
1. For patients with an SPN that is visible on
CXR, all previous CXRs should be reviewed.
2. For all patients with previous CXRs, an SPN
that is unchanged for 2 years does not
require further diagnostic evaluation. Level of
recommendation, B
3. For patients with an SPN visible on CXR in
which benign central calcification is present,
no further diagnostic evaluation is necessary.
4. For patients with an SPN, a spiral CT of the
chest with contrast is indicated to better char-
acterize the nodule, parenchyma, and medias-
tinum. CT can be useful in identifying nodules
more likely to be benign and obviate the need
for further diagnostic evaluation. Additionally,
chest CT plays an important role in staging (as
delineated in the chapter on noninvasive stag-
ing elsewhere in these guidelines). Level of
recommendation, B
5. For patients with an SPN, MRI is not indi-
cated except in these special instances. Level
of evidence, good; benefit, none; grade of
recommendation, D
6. For patient with an SPN 1 cm in size, PET
scanning is not currently recommended. Level
of evidence, good; benefit, none/negative;
7. For patients with an SPN who are surgical
candidates and have a negative mediastinal
evaluation on CT, PET scanning with FDG as
an investigational tool, where available, may
be warranted. Level of evidence, fair; benefit,
surgical candidates, if PET scanning with
FDG results are negative, a repeat CT scan is
required at least once in 3 months. Level of
evidence, poor; benefit, substantial; grade of
recommendation, C
surgical candidates, if there are unchanged
results from prior CXR and negative PET scan
findings, serial follow-up is recommended,
consisting of an initial CXR, and CT scanning
at 3, 6, 12, and 24 months. Level of evidence,
fair; benefit, substantial; grade of recommenda-
tion, B
10. For the patients with an SPN who are opera-
ble candidates, TTNA is not indicated. Level
of evidence, good; benefit, none; grade of
recommendation, D
11. For operable patients with an SPN who de-
cline surgical intervention, TTNA or trans-
bronchial needle biopsy is the preferred pro-
cedure for establishing a diagnosis. Level of
recommendation, B
12. For patients with an SPN who are not opera-
ble candidates, or are at high risk, TTNA may
be helpful to establish tissue diagnosis. Level
recommendation, B
13. For patients with an SPN, bronchoscopy is
usually not indicated. Level of evidence, good;
benefit, none; grade of recommendation, D
14. For operable patients with an SPN, if the
lesion is amenable to a wedge resection, then
a wedge resection is the procedure of choice
followed by a lobectomy if the pathologic
finding is positive for cancer. Level of evi-
ommendation, B
15. For operable patients with an SPN, if the
lesion is not amenable to a wedge resection, a
diagnostic lobectomy is acceptable. Level of
recommendation, A
16. All pulmonary resections, anatomic or non-
anatomic, must include a systematic lymph
node dissection. Level of evidence, good; ben-
surgical candidates, a wedge resection or seg-
mentectomy is acceptable. Level of evidence,
fair; benefit, substantial; grade of recommen-
dation, B
18. For patients with an SPN without a definitive
tissue diagnosis, a minimum follow-up of 2
years is recommended. This should include an
initial CXR, and CT scanning at 3, 6, 12, and
24 months. Level of evidence, poor; benefit,
References
1 Tuddenham WI. Glossary of terms for thoracic radiology:
recommendations of the Nomenclature Committee of the
Fleischner Society. Am J Roentgenol 1984; 43:509517
2 Leef JL 3rd, Klein IS. The solitary pulmonary nodule. Radiol
Clin North Am 2002; 40:123143
3 Abeloff MD, Annitage IO, Lichter AS, et al. Clinical oncol-
ogy. 2nd ed. New York, NY; W.B. Saunders, 2000; 13981464
4 Mountain CF. Revisions in the international system for
staging lung cancer. Chest 1997; 111:17101717
5 Yankelvitz DF, Hensche CI. Lung cancer: small solitary
pulmonary nodules. Radiol Clin North Am 2000; 38:19
7 Jemal A, Murray T, Thun LA. Cancer statistics, 2002. CA
Cancer J Clin 2002; 52:2347
8 Swanson SI, Iaklitsch MT, Mentzer SI, et al. Management of
the solitary pulmonary nodule: role of thoracoscopy in diag-
nosis and therapy. Chest 1999; 116:523S524S
9 Swenson SI, Silverstein MD, Ilstrup DM, et al. The proba-
bility of malignancy in solitary pulmonary nodules: application
to small radiologically indeterminate nodules. Arch Intern
Med 1997; 57:849855
10 Gurney IW. Determining the likelihood of malignancy in
solitary pulmonary nodules with Bayesian analysis: Part I.
Theory. Radiology 1993; 186:405413
11 Henske CT, Yankelevitz DF, Mateesscu I, et al. Neural
networks for the analysis of small pulmonary nodules. Clin
Imaging 1997; 21:390399
12 Nakamura K, Yoshida H, Engelmann R, et al. Computerized
analysis of the likelihood of malignancy in solitary pulmonary
with use of artificial neural networks. Radiology 2000; 214:
823830
13 Lillington GA. Management of solitary pulmonary nodules:
how to decide when resection is required. Postgrad Med
1997; 101:145150
14 Potchen El, Bisesi MA. When is it malpractice to miss lung
cancer on chest radiographs? Radiology 1990; 175:2932
15 Stitik FP, Tockrnan MS. Radiologic screening in the early
detection of lung cancer. Radiol Clin North Am 1978;
16:347366
16 Heenan RT, Flehinger BJ, Melamed MR, et al. Non-small
cell lung cancer: results of the New York screening program.
Radiology 1984; 151:289293
17 Muhm JR, Miller WE, Fontana RS, et al. Lung cancer
detected during a screening program using 4-month chest
radiographs. Radiology 1983; 143:609615
18 Guiss LW, Kuenstler. A retrospective review of survey pho-
tofluorograms of persons with lung cancer. Cancer 1960;
13:9195
19 Quekel LO, Kessels, AG, Goei R, et al. Miss rate of lung
cancer on the chest radiograph in clinical practice. Chest
1999; 115:720732
20 Good CA. Management of patient with solitary mass in lung.
Chic Med Soc Bull 1953; 55:893896
21 Good CA, Wilson TW. The solitary circumscribed pulmonary
nodule. JAMA 1958; 166:210215
22 Berger WG, Erly WK, Krupinski EA, et al. The solitary
pulmonary nodule on chest radiography: can we really tell if
the nodule is calcified? AJR Am J Roentgenol 1999; 176:201
204
23 Fraser RG, Pare JA. Diagnosis of diseases of the chest. 3rd
ed. Vol 2. Philadelphia, PA; W.B. Saunders, 1989; 13271699
24 Lillington GA, Caskey CI. Evaluation and management of
solitary multiple pulmonary nodules. Clin Chest Med 1993;
14:111119
25 Zwirewich CV, Vedal S, Miller RR, et al. Solitary pulmonary
nodule: high resolution CT and radiologic-pathologic corre-
lation. Radiology 1991; 179:469481
26 Zerhouni EA, Stilik FP, Siegelman SS, et al. CT of the
pulmonary nodule: a cooperative study. Radiology 1986;
160:319327
27 White P, Adams H, Crane M, et al. Preoperative staging of
carcinoma of the bronchus: can computed tomographic scan-
ning reliably identify stage II tumours. Thorax 1994; 49:951
957
28 Shaffer K. Radiologic evaluation in lung cancer: diagnosis and
staging. Chest 1997; 112:235S238S
29 Shea J, Lillington GA. Preoperative staging of lung cancer.
West J Med 1994; 161:508509
30 Rolston KV, Rodriguez S, Dholakia N, et al. Pulmonary
infections mimicking cancer: a retrospective three year re-
view. Support Care Cancer 1997; 5:590593
31 Shure D, Fedullo PF. Transbronchial needle aspiration of
peripheral masses. Am Rev Respir Dis 1983; 728:10901092
32 Woodring JH, Fried AM. Significance of wall thickness in
solitary cavities of the lung: a follow-up study. AJR Am J
Roentgenol 1983: 140:473474
33 Swenson SI, Brown LR, Colby TV, et al. Lung nodule
enhancement at CT: prospective findings. Radiology 1996;
201:447455
34 Gould MK, Maclean CC, Kuschner WG, et al. Accuracy of
positron emission tomography for diagnosis of pulmonary
nodules and mass lesions: a meta-analysis. JAMA 2001;
285:914924
35 Seely IM, Mayo IR, Miller RR, et al. T1 lung cancer:
prevalence of mediastinal nodal metastases and diagnostic
accuracy of CT. Radiology 1993; 186:129132
36 Decker GL, Whitlock WL, Schaefer PS, et al. The impact of
thoracic computed tomography in clinically staged T1N0M0
chest lesions. Arch Intern Med 1990; 150:557559
37 Goldsmith SI, Kostakoglu L. Nuclear medicine imaging of
lung cancer. Radiol Clin North Am 2000; 38:511524
38 1999 national physician fee schedule relative value file.
Healthcare Financing Administration. Available at http://
cms.hhs.gov. Accessed December 20, 2002
39 Weisbrod GL. Transthoracic needle biopsy. World J Surg
1993; 17:705711
40 Sanders C. Transthoracic needle aspiration. Clin Chest Med
1992; 13:1116
41 Lacasse Y, Wong E, Guyatt GH, et al. Transthoracic needle
aspiration biopsy for the diagnosis of localized pulmonary
lesions: a meta-analysis. Thorax 1999; 54:884893
42 Doiselle PM, Shepard, IO, Mark EL. Routine addition of an
automated biopsy device to fine-needle aspiration of the lung:
a prospective assessment. AJR Am J Roentgenol 1997; 169:
661666
43 Stewart CI, Stewart IS. Immediate assessment of fine needle
aspiration cytology of the lung. J Clin Pathol 19; 49:839843
44 Murphy JM, Gleeson FV, Flower CD. Percutaneous needle
biopsy of the lung and its impact on patient management.
World J Surg 2001; 25:373380
45 American Thoracic Society/European Respiratory Society.
Pretreatment evaluation of non-small-cell lung cancer. Am J
46 Arroliga AC, Matthay RA. The role of bronchoscopy in lung
cancer. Clin Chest Med 1993; 14:8798
47 Stringfield JT, Markowitz DJ, Bentz RR, et al. The effect of
tumor size and location on diagnosis by fiberoptic bronchos-
copy. Chest 1977; 72:474476
48 Baaklini WA, Reinoso MA, Gorin AB, et al. Diagnostic yield
of fiberoptic bronchoscopy in evaluating solitary pulmonary
nodules. Chest 2000; 117:10491054
49 Torrington KG, Kern JD. The utility of fiberoptic bronchos-
copy in the evaluation of solitary pulmonary nodules. Chest
1993; 104:10211024
50 Thaete FL, Peterson MS, Plunkett MB, et al. Computed
tomography: guided wire localization of pulmonary lesions
before thoracoscopic lung resection: results in 101 cases.
J Thorac Imaging 1999; 14:9098
51 Mullan BF, Stanford W, Barnhart W, et al. Lung nodules:
improved wire for CT-guided localization. Radiology 1999;
211:561565
52 Suzuki K, Nagai K, Yoshida J, et al. Video-assisted thoraco-
scopic surgery for small indeterminate pulmonary nodules.
Chest 1999; 115:563568
53 Martini N, Ginsberg RJ. Treatment of stage I and stage II
disease. In: Aisner R, Arriga R, Green N, et al, eds. The
comprehensive textbook of thoracic oncology. Baltimore,
MD: Williams and Wilkens, 1996; 338350
54 Warren WH, Faber LP. Segmentectomy vs lobectomy in
patients with stage I pulmonary carcinoma. J Thorac Cardio-
vasc Surg 1994; 107:10871093
55 Ginsberg RJ, Rubinstein LV. Randomized trial of lobectomy
vs limited resection for T1N0 non-small cell lung cancer:
Lung Cancer Study Group. Ann Thorac Surg 1995; 57:615
622
2003;123;89-96 Chest
Bethany B. Tan, Kevin R. Flaherty, Ella A. Kazerooni and Mark D. Iannettoni
The Solitary Pulmonary Nodule
& Services
Updated Information
References
#BIBL
free at:
Citations
#otherarticles
Reprints
2003;123;97-104 Chest
Michael A. Beckles, Stephen G. Spiro, Gene L. Colice and Robin M. Rudd
Laboratory Tests, and Paraneoplastic Syndromes
Initial Evaluation of the Patient With Lung Cancer: Symptoms, Signs,
ISSN: 0012-3692.
Initial Evaluation of the Patient With
Lung Cancer*
Symptoms, Signs, Laboratory Tests, and
Paraneoplastic Syndromes
Michael A. Beckles, MB, BS; Stephen G. Spiro, MD; Gene L. Colice, MD, FCCP;
and Robin M. Rudd, MD
This chapter describes the components of the initial evaluation for a patient either suspected or
known to have lung cancer. The components of the initial evaluation are based on the recognized
manifestations of localized lung cancer, ie, symptoms referable to the primary tumor, intratho-
racic spread of lung cancer, and patterns of metastatic dissemination. Features of the history and
physical signs may be useful indicators of the extent of disease. A standardized evaluation, relying
on symptoms, signs, and routinely available laboratory tests, can serve as a useful screen for
metastatic disease. Also described are the common features of the various paraneoplastic
syndromes associated with lung cancer. (CHEST 2003; 123:97S104S)
Key words: evaluation; laboratory test; paraneoplastic; signs; symptoms
Abbreviations: ACTH adrenocorticotrophin hormone; ADH antidiuretic hormone; SIADH syndrome of inap-
propriate antidiuretic hormone; SVC superior vena cava; SVCO superior vena cava obstruction
L
ung cancer, like most other solid tumors, is
unfortunately usually recognized late in its nat-
ural history. The 5-year mortality from the time of
presentation remains at approximately 85 to 90%.
Of 100 newly presenting patients with lung cancer,
80 patients will be inoperable at presentation and
approximately 20 patients will proceed to attempted
resection, of whom 5 to 10 patients will be alive
5 years later.
1
More than 90% of patients with lung cancer will
be symptomatic at presentation. The diagnosis is
usually suspected following a chest radiograph. In a
series of 678 consecutive patients with newly diag-
nosed lung cancer, only 44 patients (6%) were
asymptomatic.
2
A minority (183 patients, 27%) pre-
sented with symptoms related to the primary tumor.
Most patients had either nonspecific systemic symp-
toms, including anorexia, weight loss, and fatigue
(232 patients, 27%), or specific symptoms indicating
metastatic disease (219 patients, 32%) at initial pre-
sentation. Prognosis was clearly related to the type of
presenting symptoms. There was a better 5-year
survival in asymptomatic patients (18%), than for
those with symptoms related to the primary tumor
(12%). Those with nonspecific symptoms had a 6%
5-year survival, and those with symptoms indicating
metastatic disease fared the worst, with none alive at
5 years.
Chest Radiography
The chest radiograph plays a pivotal role in the
recognition of lung cancer. Certainly, in the asymp-
tomatic patient, an abnormality on the chest radio-
graph would be the first clue to the presence of lung
cancer. In patients with symptoms related to the
primary tumor, the chest radiograph may often
strongly support a suspicion of carcinoma of the
lung. For patients presenting with either nonspecific
systemic complaints or symptoms suggestive of met-
astatic disease, the chest radiograph will be helpful in
focusing attention quickly on the lung as the most
likely primary site.
Comprehensive reviews of the abnormalities seen
on the chest radiograph in patients with carcinoma of
the lung may be obtained elsewhere.
3,4
However, it
is useful to appreciate that clues from the chest
radiograph may not only suggest the diagnosis of
*From the Department of Respiratory Medicine (Dr. Beckles),
Royal Free Hospital, London, UK; Department of Respiratory
Medicine (Dr. Spiro), Middlesex Hospital, London, UK; Pulmo-
nary, Critical Care and Respiratory Services (Dr. Colice), Wash-
ington Hospital Center, Washington, DC; and Department of
Medical Oncology (Dr. Rudd), St. Bartholomews Hospital,
Smithfield, London, UK.
Correspondence to: Stephen G. Spiro, MD, Department of Re-
spiratory Medicine, The Middlesex Hospital, Mortimer St, Lon-
don W1T 3AA, United Kingdom; e-mail: stephen.spiro@uclh.org
lung cancer, but may also point toward the histologic
subtype. Adenocarcinoma is the most common type
of lung cancer, accounting for 30 to 35% of all cases.
5
Adenocarcinomas are more frequently seen periph-
erally. Squamous cell carcinoma, which accounts for
up to 30% of all lung cancers, typically arises in the
central bronchi and commonly extends into the
hilum and mediastinum. They may be slower grow-
ing and metastasize late.
6
Squamous cell carcinoma
may also occur in the lung parenchyma where they
may cavitate.
6,7
Large cell carcinoma comprises 10 to
20% of all lung cancers and is also seen more
commonly peripherally.
3
Small cell lung carcinoma
comprises 15 to 25% of all lung cancers and, like
squamous cell carcinoma, also usually develops in
the proximal airways and involves the hilum and
mediastinum.
3
Symptoms, Signs, and Laboratory Tests in
Lung Cancer
Symptoms, signs and abnormalities in laboratory
tests relating to the lung cancer can be classified
as follows: (1) those related to the primary lesion,
(2) those related to intrathoracic spread, (3) those
related to distant metastasis, and (4) those related to
paraneoplastic syndromes.
Symptoms and Signs Related to the
Primary Tumor
The frequencies of local and systemic symptoms
in patients with lung cancer are summarized in
Table 1.
813
Cough
Cough is the most common presenting symptom
in lung cancer. Many lung cancers occur in central
airways and may lead to postobstructive pneumonia,
or cause lymph node enlargement that may lead to
cough. Failure of acute exacerbations of COPD to
clear should raise the suspicion of a neoplasm.
Dyspnea
Dyspnea develops early in up to 60% of patients. It
is usually associated with increasing cough and spu-
tum. If the tumor is occluding a main airway, it can
cause breathlessness, which may be associated with a
unilateral wheeze.
Hemoptysis
Hemoptysis is a common presenting symptom. It
is rarely severe and usually only consists of blood
streaking of the sputum. The most common descrip-
tion is that of coughing up blood for several days in
succession. Although chest radiograph findings are
usually abnormal in patients with hemoptysis from
lung cancer, in a small percentage of patients the
finding will either be normal or not show a localizing
abnormality.
14
In patients 40 years old with
COPD and a smoking history presenting with he-
moptysis, even though chest radiograph findings may
be unremarkable there should still be a high index of
suspicion for lung cancer. Besides careful observa-
tion, the clinician may consider further diagnostic
tests, including sputum cytology, bronchoscopy, or
chest CT.
Chest Discomfort
Chest discomfort is common and occurs in up to
50% of patients at diagnosis. This is often of an
ill-defined nature, intermittent and aching in quality.
Definite pleuritic pain may occur as a result of direct
spread of the tumor to the pleural surface.
Signs and Symptoms of Intrathoracic
Spread
Intrathoracic spread of lung cancer, either by
direct extension or lymphatic spread, produces a
variety of symptoms and signs. These may be caused
by involvement of the following structures:
(1) nerves, eg, recurrent laryngeal nerve, phrenic
nerve, brachial plexus, and sympathetic nerve trunks;
(2) chest wall and pleura; (3) vascular involvement,
eg, superior vena cava (SVC), pericardium, and
heart; and (4) viscera, including esophagus.
Recurrent Laryngeal Nerve Palsy
Recurrent laryngeal nerve palsy has been reported
in 2 to 18% of cases, and is more common in
Table 1Range of Frequencies of Initial Symptoms
and Signs of Lung Cancer*
Symptoms and Signs Range of Frequency, %
Cough 875
Weight loss 068
Dyspnea 360
Chest pain 2049
Hemoptysis 635
Bone pain 625
Clubbing 020
Fever 020
Weakness 010
SVCO 04
Dysphagia 02
Wheezing and stridor 02
*Modified from Andersen and Prakash,
8
Grippi,
9
Hyde and Hyde,
10
Cromartie et al,
11
Karsell and McDougall,
12
and American Thoracic
Society/European Respiratory Society.
13
left-sided tumors because of the circuitous route of
the left recurrent laryngeal nerve around the aortic
arch, and causes hoarseness. It is associated with
poor expectoration with coughing, and an increased
risk of aspiration.
Phrenic Nerve Paralysis
Phrenic nerve dysfunction may be noted on the
chest radiograph by the presence of an elevated
hemidiaphragm, or can present with breathlessness
in patients already compromised by lung disease.
Pancoast Tumor
Pancoast tumor is also called superior sulcus tu-
mor, and arises posteriorly in the apex of an upper
lobe near the brachial plexus, commonly infiltrating
the eighth cervical and first and second thoracic
nerve roots. This causes pain, cutaneous temperature
change, and muscle wasting along the relevant nerve
root. There is often a delay of many months before
the true diagnosis is revealed.
Horner Syndrome
This is due to involvement of the sympathetic
chain and stellate ganglion causing unilateral enoph-
thalmos, ptosis, small pupil, and ipsilateral lack of
facial sweating.
Chest Wall
Pain from chest wall involvement is a common
presenting symptom. More than 50% of patients
with lung cancer complain of chest pain during the
course of their disease. The pain is usually dull, tends
to be persistent, poorly localized, and unrelated to
breathing or coughing. Retrosternal pain may be due
to massive hilar and mediastinal nodal involvement.
When chest pain is particularly severe and localized,
it is usually related to either direct invasion of the
pleura or chest wall by the primary tumor, or due to
a rib metastasis. Tenderness may be elicited at the
site of rib involvement and, rarely, a soft-tissue mass
can be palpated.
Pleura
Pleural involvement occurs in 8 to 15% of patients
with lung cancer. Pleuritic chest pain can occur with
the early phase of neoplastic pleural invasion and can
disappear with the onset of a pleural effusion. Pleu-
ral effusion, which may result in dyspnea, is generally
caused by direct pleural extension, but may also be
secondary to mediastinal node involvement and lym-
phatic obstruction. Signs of a pleural effusion include
dullness to percussion and decreased breath sounds.
SVC Obstruction
Lung cancer accounts for 46 to 75% of all cases of
SVC obstruction (SVCO); the most common histo-
logic subtype associated with SVCO is small cell
carcinoma
10,15,16
and is due to direct invasion by the
primary tumor, or from enlarged right paratracheal
metastatic lymph nodes. The patient will complain of
facial swelling, including the neck and eyelids, with
dilated veins visible over the upper torso, shoulders,
and arms. There may also be headache, dizziness
(particularly on bending forwards), drowsiness, blur-
ring of vision, cough, and dysphagia.
15,17
Heart and Pericardium
Metastases to the heart and pericardium usually
occur by direct lymphatic spread. At autopsy, cardiac
involvement occurs in approximately 15% of cases,
and a small number will have tamponade.
18
In
primary lung cancer, the pericardium is the most
common site of cardiac involvement.
19
Esophagus
Enlargement of hilar and mediastinal nodes is
usually due to metastatic spread and seldom causes
symptoms unless it is massive, when it can compress
the esophagus and limit swallowing.
Symptoms, Signs, and Laboratory Tests
Indicating Extrathoracic Metastases
Approximately one third of patients present with
symptoms as a result of distant metastases. The most
common sites of distant metastasis from lung cancer
are the bones; liver, adrenal glands, and intra-
abdominal lymph nodes; brain and spinal cord; and
lymph nodes and skin.
Bones
Lung cancer can metastasize to virtually any bone,
although the axial skeleton and proximal long bones
are most commonly involved. The primary symptom
resulting from bone involvement is pain, which may
have a pleuritic component when the ribs are in-
volved. Bone pain is present in up to 25% of all
patients at presentation.
Liver, Adrenal Glands, and Intra-abdominal
Lymph Nodes
Liver metastases occur commonly with lung can-
cer. However, liver function test results are seldom
abnormal until the metastases are numerous and
large. Hepatic metastases most commonly produce
symptoms of weakness and weight loss. When
present, hepatic metastases carry a very poor prog-
nosis. Adrenal lesions and para-aortic lymph node
metastases may occur and are most commonly seen
with small cell lung cancers; in the latter cell type,
they are often discovered during staging. Clinical
evidence of adrenal insufficiency is rarely seen.
Brain And Spinal Cord
Intracranial metastases occur in 10% of patients at
presentation. Spinal cord metastases are less com-
mon and tend to occur in patients with cerebral
metastases. Brain metastasis may produce headache,
nausea and vomiting, focal neurologic symptoms or
signs, seizures, confusion, and personality changes.
The lung is the primary site of approximately 70% of
cancers that initially present with symptomatic brain
metastases.
20
Lymph Nodes and Skin
The most common site of palpable lymphadenop-
athy is the supraclavicular fossa, which can be in-
volved in 15 to 20% of cases during the course of the
disease. Identifying an enlarged lymph node or
subcutaneous nodule due to metastatic lung cancer
is extremely helpful in facilitating both diagnosis and
staging. Fine needle aspiration can be performed
quickly at the bedside or as an outpatient with little
morbidity and with a high sensitivity.
21
Standardized Evaluation For Systemic Metastases
Carbone et al
2
and Feinstein and Wells
2226
have
explored the relationship between symptoms at pre-
sentation and prognosis in a large cohort of consec-
utive patients with lung cancer. Patients with the
best prognosis were either asymptomatic or had
symptoms referable only to the primary tumor. In
patients with either systemic symptoms of anorexia,
weight loss, and fatigue or symptoms attributable to
metastatic disease, prognosis was especially poor.
The relationship between systemic symptoms and
prognosis was conserved with standard staging of
lung cancer. Within any individual tumor stage,
there was a gradient of worsening prognosis in
patients who presented with anorexia, weight loss,
and fatigue. The biological association between sys-
temic symptoms and worse prognosis was not en-
tirely clear, although, intuitively, patients with sys-
temic symptoms would be clinically suspected of
having extensive disease.
Hooper and colleagues
27,28
incorporated the con-
cept of systemic symptoms as a reflection of more
extensive disease into an approach for identifying
lung cancer patients more likely to have evidence of
metastases on radiographic studies. They used a
cluster of clinical factors, consisting of symptoms,
signs, and standard laboratory tests, which were
thought to reflect metastatic disease. Included
within these clinical factors were the nonspecific
variables weight loss and anemia. They found that
abnormalities in these clinical factors were associ-
ated with radiographic evidence of metastatic dis-
ease. The more abnormalities noted in the clinical
assessment, the more likely that metastases would be
detected. They also found that patients with no
abnormalities in these clinical factors were extremely
unlikely to have scan evidence of metastatic disease.
Silvestri et al
29,30
adapted Hoopers criteria (Table 2)
and retrospectively asked whether they would be a
useful screen for detecting adrenal metastases. As
with the work of Hooper and colleagues,
27,28
if no
clinical abnormalities were noted adrenal metastases
were not found by CT. The more clinical abnormal-
ities found, the more likely adrenal metastases would
be found. Both the work of Silvestri et al
29,30
and a
study by Quinn and coworkers
31
pointed out that
abnormalities in the clinical assessment would often
not be helpful in identifying the site of metastases.
However, the recognition of abnormalities in the
clinical screen strongly suggested the presence of
metastases.
Silvestri et al
31
expanded on this work by asking
whether the clinical evaluation would be useful in
identifying which patients with lung cancer would
have extrathoracic metastases detected by CT of the
brain or abdomen or radionuclide bone scans. They
performed a meta-analysis of all studies in patients
with lung cancer, which provided data on both
radiographic studies and the clinical factors adapted
from Hoopers criteria. Consistent with earlier work,
this meta-analysis showed that patients with clinical
abnormalities were often found to have metastatic
disease. However, if no abnormalities were noted in
the clinical assessment, patients were very unlikely to
have evidence of metastatic disease on CT of the
brain or abdomen or radionuclide bone scans. These
authors concluded that performing an assessment of
various clinical factors through a thorough history
and physical examination and standard laboratory
tests would be a useful screen for identifying patients
with a higher likelihood of metastatic disease.
Recommendation
1. All patients with known or suspected lung
cancer should undergo a thorough history,
physical examination, and standard laboratory
tests as a screen for metastatic disease. Level of
recommendation, B
Paraneoplastic Syndromes
Paraneoplastic syndromes are a group of clinical
disorders associated with malignant diseases that are
not directly related to the physical effects of primary
or metastatic tumor. The exact mechanism by which
paraneoplastic syndromes occur is not fully under-
stood. Paraneoplastic syndromes occur in at least
10% of patients with bronchogenic carcinoma.
8,9
The extent of paraneoplastic symptoms is unre-
lated to the size of the primary tumor, and in some
cases can precede the diagnosis of malignant disease.
At other times they may occur late in the illness, or
herald the first sign of recurrence. The paraneoplas-
tic syndromes may be due to the production of
biologically active substances either by the tumor, or
in response to the tumor (eg, polypeptide hormones,
hormone-like peptides, antibodies or immune com-
plexes, prostaglandins, or cytokines). Some of the
more common paraneoplastic syndromes are out-
lined below. There are comprehensive reviews else-
where.
32,33
Table 3 shows the myriad manifestations of para-
neoplastic syndromes associated with lung cancer.
The neurologic syndromes associated with lung can-
cer may occasionally occur through autoimmune
mechanisms.
34,35
Small cell carcinoma is the most
common type of lung cancer associated with para-
neoplastic autoimmune neurologic syndromes. Di-
rect metastatic effects or metabolic or infectious
processes must be excluded as contributors to the
neurologic findings. The severity of the neurologic
symptoms is unrelated to the tumor bulk; in fact, a
primary malignant lesion may be undetected before
death despite disabling symptoms.
36
Common Endocrine Paraneoplastic Syndromes
Associated With Lung Cancer
Hypercalcemia: Hypercalcemia is frequently sec-
ondary to bony metastases. It can, however, be due
to production of a parathyroid hormone-related pep-
tide.
37
It is most common with squamous cell carci-
noma; approximately 15% of patients acquire hyper-
calcemia before death. Symptoms include nausea,
vomiting, abdominal pain, constipation, polyuria,
thirst, dehydration, confusion, and irritability.
Syndrome of Inappropriate Antidiuretic Hormone
Production: Antidiuretic hormone (ADH) is pro-
duced in the hypothalamus and secreted from the
posterior lobe of the pituitary gland. This hormone is
involved in the maintenance of the extracellular fluid
environment by reducing free water clearance. Al-
though excess ADH production can be documented
in up to 70% of patients with lung cancer, syndrome
of inappropriate ADH (SIADH) is not as common.
However, all patients with hyponatremia related to
SIADH in lung cancer do have elevated serum levels
of ADH.
38
Atrial natriuretic peptide is another hor-
mone produced ectopically by lung cancer cells,
which affects renal salt and water handling. In
individual patients, increased levels of atrial natri-
uretic peptide may contribute to hyponatremia by
causing a natriuresis.
38
SIADH is mainly associated
with small cell lung cancer, although other malignant
tumors of the lung may rarely be associated with this
syndrome.
3941
The production of excess ADH is
not always related to symptoms attributed to
SIADH.
39,42,43
Only 1 to 5% of patients have symp-
toms attributable to the SIADH. Manifestations of
SIADH include confusion, unexplained seizures, de-
creased level of consciousness, and coma. Biochem-
ically, the syndrome is defined as low serum sodium,
a dilute plasma osmolality with a higher, or inap-
propriate urine osmolality, in the presence of con-
tinued urinary sodium excretion. The syndrome re-
solves promptly ( 3 weeks) with the initiation of
combination cytotoxic chemotherapy in 80% of pa-
tients with small cell lung cancer, but commonly
recurs with tumor progression.
44
Cushing Syndrome: Adrenocorticotrophic hor-
mone (ACTH) is the most commonly produced
ectopic hormone in lung cancer. It is derived from
the parent hormone, pro-opiomelanocortin polypep-
tide. As with ADH above, it is not unusual to find
increased serum levels of ACTH in patients with
lung cancer; it may be detectable in up to 50% of
Table 2Features of a Standardized Evaluation for Systemic Metastases*
Symptoms Signs Laboratory Tests
Constitutional, weight loss than 10 lb Lymphadenopathy ( 1 cm) Hematocrit 40% in male patients
Musculoskeletal, focal skeletal pain Hoarseness, SVC syndrome Hematocrit 35% in female patients
Neurological: headaches, syncope, seizures,
extremity weakness, or recent change in
mental status
Bone tenderness
Hepatomegaly ( 13-cm span)
Focal neurologic signs, papilledema
Soft-tissue mass
Elevated alkaline phosphatase, -glutamyltransferase,
or serum glutamicoxaloacetic transaminase
*Modified from Silvestri et al.
29,30
patients with lung cancer.
45
However, despite the
presence of increased levels of ACTH, Cushing
syndrome is rarely seen. Signs and symptoms of
Cushing syndrome develop in only 1 to 5% of
patients with small cell lung cancer.
46,47
This limited
clinical expression is due partly to the rapid progres-
sion of small cell lung cancer and also the release of
a biologically inactive form of the hormone. The
most common features associated with this condition
are clinical symptoms of weakness, muscle wasting,
drowsiness, confusion and psychosis, dependent
edema, hypokalemic alkalosis, and hyperglycemia.
Digital Clubbing and Hypertrophic
Osteoarthropathy
These disorders may be associated with any cell
type of lung cancer, although they are most fre-
quently associated with squamous and adenocarci-
noma and least frequently associated with small cell
lung carcinoma. The exact mechanism for clubbing
and hypertrophic osteoarthropathy is unknown, al-
though suggestions include neurogenic, hormonal,
and vascular mechanisms.
48
Clubbing is much more common than hypertro-
phic osteoarthropathy. In one study of 111 consecu-
tive patients with pathologically proven lung cancer,
clubbing was present in 29%. The phenomenon was
more common among women than men (40% vs
19%) and more common in non-small cell lung
cancer than small cell lung cancer (35% vs 4%).
49
Hypertrophic osteoarthropathy is seen in 5% of
patients with non small cell lung cancer
50
and is
characterized by a painful symmetrical arthropathy
(usually of the ankles, wrists, and knees) and perios-
teal new bone formation on the distal long bones of
the limbs. Small cell lung cancer is a rarer cause; in
one series, it accounted for only 1% of the cases with
hypertrophic osteoarthropathy.
51
Anecdotal observa-
tions indicate that clubbing and hypertrophic osteo-
arthropathy may resolve with successful treatment of
the primary tumor, particularly surgical resection of
a non-small cell lung cancer.
Neurologic Syndromes
A variety of poorly understood neurologic syn-
dromes may occur in lung cancer, and affect 4 to 5%
of patients.
52
The diagnosis of a neurologic paraneo-
plastic syndrome is made once other causes, such as
electrolyte imbalance, metastatic disease, cerebral
and spinal vascular disease, infections, and treatment
toxicity, are excluded. Small cell carcinoma is the
most common type of lung cancer associated with
paraneoplastic autoimmune neurologic syndromes.
The syndromes involved include Lambert-Eaton
myasthenic syndrome, peripheral neuropathy, corti-
cal cerebellar degeneration, and several other CNS
syndromes. Some of these syndromes may be iden-
tified by the presence of antibodies either in the
serum or the cerebrospinal fluid.
35
A more extensive
review of these and other paraneoplastic syndromes
may be found elsewhere.
33
Effective chemotherapy in patients with small cell
lung cancer and a neurologic paraneoplastic syn-
drome may result in sustained improvements in the
neurologic symptoms.
53
In a small series of patients
with small cell lung cancer, overall prognosis was
more favorable in those with the Lambert-Eaton
myasthenic syndrome than those without this para-
neoplastic syndrome.
54
Recommendation
2. Patients with lung cancer and a paraneoplastic
syndrome should not be precluded from poten-
tially curative therapy on the basis of these
symptoms alone. Level of evidence, fair; bene-
fit, substantial; grade of recommendation, C
Summary
More than 90% of patients with lung cancer will
be symptomatic at presentation. A minority present
with symptoms related to the primary tumor, and
most patients present with either nonspecific sys-
Table 3Paraneoplastic Syndromes Associated With Lung Cancer*
Endocrine Neurologic Skeletal Renal Metabolic
SIADH
Nonmetastatic hypercalcaemia
Cushing syndrome
Gynecomastia
Hypercalcitonemia
Elevated levels of LSH, FSH
Hypoglycaemia
Hyperthyroidism
Carcinoid syndrome
Subacute sensory neuropathy
Mononeuritis multiplex
Intestinal pseudo-obstruction
Lambert-Eaton syndrome
Encephalomyelitis
Necrotising myelopathy
Cancer-associated retinopathy
Hypertrophic osteoarthropathy
Clubbing
Glomerulo-
nephritis
Nephrotic
syndrome
Lactic acidosis
Hypouricemia
*Modified from Scagliotti,
1
Carbone et al,
2
and Theros.
7
temic symptoms, including anorexia, weight loss and
fatigue, or specific symptoms indicating metastatic
disease. Prognosis is related to the type of presenting
symptoms. Asymptomatic patients and patients with
symptoms related to the primary tumor have better
5-year survivals than those with systemic symptoms
or symptoms indicating metastatic disease. Assessing
symptoms, signs and standard laboratory tests in a
standardized manner in patients presenting with
known or suspected lung cancer may serve as a
useful screen for identifying those patients with a
higher likelihood of metastatic disease.
Paraneoplastic syndromes, which occur in up to
10% of patients with lung cancer, are a group of
clinical disorders associated with malignant diseases
that are not directly related to the physical effects of
primary or metastatic tumor. These syndromes may
be due to the production of biologically active
substances or other, presently unclear, mechanisms.
Paraneoplastic symptoms are unrelated to the size of
the primary tumor, in some cases can precede the
diagnosis of malignant disease, and at other times
may occur late in the illness, or herald the first sign
of recurrence.
cancer should undergo a thorough history,
physical examination, and standard laboratory
tests as a screen for metastatic disease. Level of
recommendation, B
2. Patients with lung cancer and a paraneoplastic
syndrome should not be precluded from poten-
tially curative therapy on the basis of these
symptoms alone. Level of evidence, fair; bene-
References
1 Scagliotti G. Symptoms, signs and staging of lung cancer. Eur
Respir Mon 2001; 17:86119
2 Carbone PP, Frost JK, Feinstein AR, et al. Lung cancer:
perspectives and prospects. Ann Intern Med 1970; 73:1003
1024
3 Patz EF Jr. Imaging bronchogenic carcinoma. Chest 2000;
117:90S95S
4 Haque AK. Pathology of carcinoma of lung. J Thorac Imaging
1991; 7:920
5 Vincent RG, Pickren JW, Lane WW, et al. The changing
histopathology of lung cancer. Cancer 1977; 39:16471655
6 Sider L. Radiographic manifestations of primary broncho-
genic carcinoma. Radiol Clin North Am 1990; 28:583597
7 Theros EG. 1976 Caldwell Lecture: varying manifestation of
peripheral pulmonary neoplasms: a radiologic-pathologic cor-
relative study. AJR Am J Roentgenol 1977; 128:893914
8 Andersen HA, Prakash UBS. Diagnosis of symptomatic lung
cancer. Semin Respir Med 1982; 3:165175
9 Grippi MA. Clinical aspects of lung cancer. Semin Roentge-
nol 1990; 25:1224
10 Hyde L, Hyde CI. Clinical manifestations of lung cancer.
Chest 1974; 65:299306
11 Cromartie RS 3rd, Parker EF, May JE, et al. Carcinoma of
the lung. Ann Thorac Surg 1980; 30:3035
12 Karsell PR, McDougall JC. Diagnostic tests for lung cancer.
Mayo Clin Proc 1993; 68:288296
13 Pretreatment evaluation of non-small-cell lung cancer: The
American Thoracic Society and The European Respiratory
Society. Am J Respir Crit Care Med 1997; 156:320332
14 Colice GL. Detecting lung cancer as a cause of hemoptysis in
patients with a normal chest radiograph. Chest 1997; 111:
877884
15 Yellin A, Rosen A, Reichert N, et al. Superior vena cava
syndrome. Am Rev Respir Dis 1990; 141:11141118
16 Van Houtte P, De Jager R, Lustman-Marechal J, et al.
Prognostic value of the superior vena cava syndrome as the
presenting sign of small cell anaplastic carcinoma of the lung.
Eur J Cancer 1980; 16:14471450
17 Ahmann FR. A reassessment of the clinical implications of
the superior vena caval syndrome. J Clin Oncol 1984; 2:961
969
18 Mukai K, Shinkai T, Tominaga K, et al. The incidence of
secondary tumors of the heart and pericardium. Jpn J Clin
Oncol 1988; 18:195201
19 Onuiigbo WIB. The spread of lung cancer to the heart,
pericardium and great vessels. Jpn Heart J 1974; 15:234238
20 Merchut MP. Brain metastases from undiagnosed systemic
neoplasms. Arch Intern Med 1989; 149:10761080
21 Rohwedder JJ, Handley JA, Kerr D. Rapid diagnosis of lung
cancer from palpable metastases by needle thrust. Chest
1990; 98:13931396
22 Feinstein AR. Symptomatic patterns, biologic behavior, and
prognosis in cancer of the lung. Ann Intern Med 1964;
61:2743
Table 3Paraneoplastic Syndromes Associated With Lung Cancer (continued)*
Systemic Collagen/Vascular Cutaneous Hematologic Coagulopathies
Anorexia,
cachexia
Fever
Dermatomyositis
Polymyositis
Vasculitis
Systemic lupus erythematosus
Acquires hypertrichosis languinosa
Erythema gyratum repens
Erythema multiforme
Tylosis
Erythrodermia
Exfoliative dermatitis
Acanthosis nigricans
Sweet syndrome
Pruritus, urticaria
Anemia
Leucocytosis, eosinophilia
Leukemoid reactions
Thrombocytosis
Thrombocytopenic purpura
Thrombophlebitis
Disseminated intravascular
coagulation
23 Feinstein AR. Symptoms as an index of biological behaviour
and prognosis in human cancer. Nature 1966; 209:241245
24 Feinstein AR. A new staging system for cancer and reap-
praisal of early treatment and cure by radical surgery.
N Engl J Med 1968; 279:747754
25 Feinstein AR, Wells CK. Lung cancer staging. Clin Chest
Med 1982; 63:291305
26 Feinstein AR, Wells CK. A clinical-severity staging system for
patients with lung cancer. Medicine 1990; 69:133
27 Hooper RG, Beechler CR, Johnson MC. Radioisotope scan-
ning in the initial staging of bronchogenic carcinoma. Am Rev
Respir Dis 1978; 118:279286
28 Hooper RG, Tenholder MF, Underwood GH, et al. Com-
puted tomographic scanning of the brain in the initial staging
of bronchogenic carcinoma. Chest 1984; 85:774776
29 Silvestri GA, Lenz JE, Harper SN, et al. The relationship of
clinical findings to CT scan evidence of adrenal gland metas-
tases in the staging of bronchogenic carcinoma. Chest 1992;
102:17481751
30 Silvestri GA, Littenberg B, Colice GL. The clinical evaluation
for detecting metastatic lung cancer. Am J Respir Crit Care
Med 1995; 152:225230
31 Quinn DL, Ostrow LB, Porter DK, et al. Staging of non-small
cell bronchogenic carcinoma. Chest 1986; 89:270275
32 Spiro S. Bronchial tumours. In: Brewis RAL, Corrin B,
Geddes DM, et al, eds. Respiratory medicine. London, UK:
W.B. Saunders, 1995; 924961
33 Patel AM, Davila DG, Peters SG. Paraneoplastic syndromes
associated with lung cancer. Mayo Clin Proc 1993; 68:278
287
34 Grunwald GB. Autoimmune paraneoplastic syndromes. In:
Fishman AP, ed. Update: pulmonary diseases and disorders.
New York, NY: McGraw-Hill, 1992; 137146
35 Anderson NE, Rosenblum MK, Graus F, et al. Autoantibod-
ies in paraneoplastic syndromes associated with small-cell
lung cancer. Neurology 1988; 38:13911398
36 Lennon VA. Anti-Purkinje cell cytoplasmic and neuronal
nuclear antibodies aid diagnosis of paraneoplastic autoim-
mune neurological disorders. J Neurol Neurosurg Psychiatry
1989; 52:14381439
37 Martin TJ, Moseley JM, Gillespie MT. Parathyroid hormone-
related protein. Crit Rev Biochem Mol Biol 1991; 26:377395
38 Johnson BE, Chute JP, Rushin J, et al. A prospective study of
patients with lung cancer and hyponatremia of malignancy.
39 Maurer LH, ODonnell JF, Kennedy S, et al. Human neuro-
physins in carcinoma of the lung. Cancer Treat Rep 1983;
67:971976
40 Padfield PL, Morton JJ, Brown JJ, et al. Plasma arginine
vasopressin in the syndrome of antidiuretic hormone excess
associated with bronchogenic carcinoma. Am J Med 1976;
61:825831
41 Vorherr H. Para-endocrine tumor activity with emphasis on
ectopic ADH secretion. Oncology 1974; 29:382416
42 Moses AM, Scheinman SJ. Ectopic secretion of neurohy-
pophyseal peptides in patients with malignancy. Endocrinol
Metab Clin North Am 1991; 20:489506
43 Bliss DP Jr, Battey JF, Linnoila RI, et al. Expression of the
atrial natriuretic factor gene in small cell lung cancer tumors
and tumor cell lines. J Natl Cancer Inst 1990; 82:305310
44 List AF, Hainsworth JD, Davis BW, et al. The syndrome of
inappropriate secretion of antidiuretic hormone (SIADH) in
small-cell lung cancer. J Clin Oncol 1986; 4:11911198
45 Mendelsohn G, Baylin SB. Ectopic hormone production:
biological and clinical implications. Prog Clin Biol Res 1984;
142:291316
46 Odell WD, Wolfsen AR, Bachelot I, et al. Ectopic production
of lipotropin by cancer. Am J Med 1979; 66:631638
47 Hansen M, Bork E. Peptide hormones in patients with lung
cancer. Recent Results Cancer Res 1985; 99:180186
48 Shneerson JM. Digital clubbing and hypertrophic osteoar-
thropathy. Br J Dis Chest 1981; 75:113131
49 Sridhar KS, Lobo CF, Altman RD. Digital clubbing and lung
cancer. Chest 1998; 114:15351537
50 Prakash U. Hypertrophic pulmonary osteoarthropathy and
clubbing. In: Sackner MA, ed. Weekly updates: pulmonary
medicine; lesson 30. Princeton, NJ: Biomedia, 1978; 27
51 Stenseth JH, Clagett OT, Woolner LB. Hypertrophic pulmo-
nary osteoarthropathy. Dis Chest 1967; 52:6268
52 Swash M, Schwartz MS. Paraneoplastic syndromes. In: John-
son RT, ed. Current therapy in neurologic disease. Philadel-
phia, PA: BC Decker, 1990; 236243
53 Croteau D, Owainati A, Dalmau J, et al. Response to cancer
therapy in a patient with a paraneoplastic choreiform disor-
der. Neurology 2001; 57:719722
54 Maddison P, Newsom-Davis J, Mills KR, et al. Favourable
prognosis in Lambert-Eaton myasthenic syndrome and small-
cell carcinoma. Lancet 1999; 353:117118
2003;123;97-104 Chest
Laboratory Tests, and Paraneoplastic Syndromes
Initial Evaluation of the Patient With Lung Cancer: Symptoms, Signs,
& Services
Updated Information
References
#BIBL
free at:
Reprints
2003;123;105-114 Chest
for Resectional Surgery
The Physiologic Evaluation of Patients With Lung Cancer Being Considered
ISSN: 0012-3692.
The Physiologic Evaluation of Patients
With Lung Cancer Being Considered for
Resectional Surgery*
Michael A. Beckles, MB, BS; Stephen G. Spiro, MD; Gene L. Colice, MD, FCCP;
Robin M. Rudd, MD
The preoperative physiologic assessment of a patient being considered for surgical resection of
lung cancer must consider the immediate perioperative risks from comorbid cardiopulmonary
disease, the long-term risks of pulmonary disability, and the threat to survival due to inadequately
treated lung cancer. As with any planned major operation, especially in a population predisposed
to atherosclerotic cardiovascular disease by cigarette smoking, a cardiovascular evaluation is an
important component in assessing perioperative risks. Measuring the FEV
1
and the diffusing
capacity of the lung for carbon monoxide (DLCO) measurements should be viewed as comple-
mentary physiologic tests for assessing risk related to pulmonary function. If there is evidence of
interstitial lung disease on radiographic studies or undue dyspnea on exertion, even though the
FEV
1
may be adequate, a DLCO should be obtained. In patients with abnormalities in FEV
1
or DLCO
identified preoperatively, it is essential to estimate the likely postresection pulmonary reserve. The
amount of lung function lost in lung cancer resection can be estimated by using either a perfusion
scan or the number of segments removed. A predicted postoperative FEV
1
or DLCO <40% indicates
an increased risk for perioperative complications, including death, from lung cancer resection.
Exercise testing should be performed in these patients to further define the perioperative risks prior
to surgery. Formal cardiopulmonary exercise testing is a sophisticated physiologic testing technique
that includes recording the exercise ECG, heart rate response to exercise, minute ventilation, and
oxygen uptake per minute, and allows calculation of maximal oxygen consumption (V
O
2
max). Risk for
perioperative complications can generally be stratified by V
O
2
max. Patients with preoperative
V
O
2
max >20 mL/kg/min are not at increased risk of complications or death; V
O
2
max <15 mL/kg/min
indicates an increased risk of perioperative complications; and patients with V
O
2
max <10 mL/kg/min
have a very high risk for postoperative complications. Alternative types of exercise testing include
stair climbing, the shuttle walk, and the 6-min walk. Although often not performed in a standardized
manner, stair climbing can predict V
O
2
max. In general terms, patients who can climb five flights of
stairs have V
O
2
max >20 mL/kg/min. Conversely, patients who cannot climb one flight of stairs have
V
O
2
max <10 mL/kg/min. Data on the shuttle walk and 6-min walk are limited, but patients who
cannot complete 25 shuttles on two occasions will have V
O
2
max <10 mL/kg/min. Desaturation during
an exercise test has been associated with an increased risk for perioperative complications. Lung
volume reduction surgery (LVRS) for patients with severe emphysema is a controversial procedure.
Some reports document substantial improvements in lung function, exercise capability, and quality of
life in highly selected patients with emphysema following LVRS. Case series of patients referred for
LVRS indicate that perhaps 3 to 6% of these patients may have coexisting lung cancer. Anecdotal
experience fromthese case series suggest that patients with extremely poor lung function can tolerate
combined LVRS and resection of the lung cancer with an acceptable mortality rate and good
postoperative outcomes. Combining LVRS and lung cancer resection should probably be limited to
those patients with heterogeneous emphysema, particularly emphysema limited to the lobe contain-
ing the tumor. (CHEST 2003; 123:105S114S)
Key words: cardiopulmonary exercise testing; diffusing capacity; predicted postoperative lung function; preoperative
assessment; spirometry
Abbreviations: BTS British Thoracic Society; CPET cardiopulmonary exercise test; Dlco diffusing capacity of the
lung for carbon monoxide; LVRS lung volume reduction surgery; ppo predicted postoperative; %ppo percentage of
predicted postoperative; Sao
2
arterial oxygen saturation; V
o
2
max maximal oxygen consumption
W
hile surgery remains the best option for cure
for lung cancer, many potentially resectable
tumors occur in individuals with abnormal pulmo-
nary function usually due to cigarette smoking.
These patients may be at increased risk for immedi-
ate perioperative complications and long-term dis-
ability following resection of functioning lung tissue.
Cigarette smoking will also predispose these patients
to other comorbid conditions, specifically atheroscle-
rotic cardiovascular disease, which will further in-
crease the perioperative risks. Consequently, the
preoperative physiologic assessment of a patient
being considered for surgical resection of lung can-
cer must consider the immediate perioperative risks
from comorbid cardiopulmonary disease, the long-
term risks of pulmonary disability, and the threat to
survival due to inadequately treated lung cancer.
Little information is available on long-term sur-
vival of patients deemed inoperable because of
physiologic limitations, especially compared to a
group with similar physiologic limitations who un-
derwent surgical resection. One study reported that
the long-term survival curve for five high-risk pa-
tients undergoing operation was no different than for
39 similar patients deemed inoperable, despite a
higher initial mortality in the group undergoing
resection.
1
The balance between perioperative risks
and inadequate cancer treatment may be shifting,
because surgical techniques and anesthetic and post-
operative care have improved. Morbidity and mor-
tality rates following lung resection are lower now
than in the past.
2
Postoperative cardiopulmonary
complications historically noted to be of greatest
concern after lung resection, eg, acute hypercapnea,
mechanical ventilation lasting 48 h, arrhythmias,
pneumonia, pulmonary emboli, myocardial infarc-
tion, and lobar atelectasis requiring bronchoscopy,
3
now may be more effectively managed.
Following lung resection, it is generally accepted
that postoperative lung function will decrease. Serial
studies have shown that lung function and exercise
capability decrease within the first several months
following lung cancer resection, but may recover to a
small extent by 6 months.
4,5
It had been assumed
that there would be lower limits on the acceptability
of postoperative lung function, below which quality
of life would be unacceptable due to pulmonary
disability.
6
However, data relating changes in actual
quality of life to removal of functioning lung tissue in
patients with compromised lung function are lim-
ited. This issue has become particularly difficult to
interpret with the recent resurgence of interest in
lung volume reduction surgery (LVRS), and the
possibility of simultaneously resecting a lung cancer
and improving lung function with LVRS.
Ideally, the task of the preoperative physiologic
assessment is to identify patients at high risk for
perioperative complications and long-term disability
from lung cancer resection surgery using the least
invasive tests possible. The purpose of this preoper-
ative physiologic assessment is twofold: to enable
adequate counseling of the patient on treatment
options and risks so that they can make a truly
informed decision, and to identify possible steps to
reduce the risks of perioperative complications and
long-term pulmonary disability.
General Issues for Lung Cancer Surgery
All patients with lung cancer should be seen by a
physician interested in the management of this dis-
ease. Patients seen by specialists will have higher
rates of diagnosis, referral to surgeons and oncolo-
gists, and treatment with better outcomes.
7,8
A mul-
tidisciplinary team approach is essential in the assess-
ment of these patients. The proposed procedure
should be discussed with the patient and relatives.
Age should not be a reason to deny patients with
lung cancer access to lung cancer services.
9
As the
population ages, the number of patients 70 years
old will rise; it is estimated that 40% of patients
with lung cancer in 2005 will be 75 years old.
8
For
elderly patients ( 70 years old), the mortality from
reported series for lobectomy is between 4% and
7%, and for pneumonectomy averages 14%.
2,10,11
These rates are higher than for patients 70 years
old and the differential between the mortality of
pneumonectomy and lobectomy is larger in elderly
patients when compared to younger patients, but
these differences may be more a function of comor-
bidity than age alone. Information is limited on the
mortality rates for lung cancer resection in the very
elderly ( 80 years old) but suggest that the very
elderly can tolerate lobectomy.
10
As with any planned major operation, especially in
a population predisposed to atherosclerotic cardio-
vascular disease by cigarette smoking, a preoperative
cardiovascular risk assessment should be performed.
The approach to this risk assessment (Table 1) has
been described in the American College of Cardiol-
ogy and American Heart Association guidelines for
perioperative cardiovascular evaluation for noncar-
*From the Department of Respiratory Medicine (Dr. Beckles),
Royal Free Hospital, London, UK; Department of Respiratory
Medicine (Dr. Spiro), Middlesex Hospital, London, UK; Pulmo-
nary, Critical Care and Respiratory Services (Dr. Colice), Wash-
ington Hospital Center, Washington, DC; and Department of
Medical Oncology (Dr. Rudd), St. Bartholomews Hospital,
Smithfield, London, UK.
Correspondence to: Stephen G, Spiro, MD, Department of Re-
spiratory Medicine, The Middlesex Hospital, Mortimer St, Lon-
don W1T 3AA, United Kingdom; e-mail: stephen.spiro@uclh.org
diac surgery.
12
The presence of coronary artery
disease increases the risk of nonfatal myocardial
infarction or death within 30 days of noncardiac
surgery.
Reommendations
1. Patients with lung cancer should be seen by
physicians interested in the management of this
disease. Level of evidence, fair; benefit, sub-
stantial; grade of recommendation, B
2. Patients with lung cancer should be assessed by
a multidisciplinary team for their suitability for
surgery; there should be liaison between the
chest physician, thoracic surgical team, and
oncologist in all cases prior to surgery. Level of
recommendation, C
3. Patients with lung cancer should not be denied
lung resection surgery on the grounds of age
alone. Level of evidence, fair; benefit, substan-
tial; grade of recommendation, B
4. Patients with lung cancer undergoing surgery
should have a preoperative cardiologic evalua-
tion carried out according to established guide-
lines. Level of evidence, fair; benefit, substan-
Spirometry and Diffusing Capacity
The FEV
1
obtained by spirometry is the most
commonly used test to assess suitability of patients
with lung cancer for surgery. Spirometry should be
performed when the patient is in clinically stable
condition and receiving maximal bronchodilator
therapy. The FEV
1
can be expressed in either abso-
lute values or as a percentage of predicted.
There have been several studies looking at the
minimum absolute values of FEV
1
that, as a single
measurement, will predict whether a patient will
survive a pneumonectomy and still have a good level
of habitual activity. Many studies are retrospective
and have small numbers of patients. A review of the
literature suggests an FEV
1
2 L as a safe lower
limit for pneumonectomy and 1.5 L for a lobec-
tomy.
13,14,15
In the British Thoracic Society (BTS)
guidelines, data from 2,000 patients in three large
series in the 1970s have shown that a mortality rate
of 5% can be achieved if the preoperative FEV
1
is
1.5 L for a lobectomy and 2 L for a pneumo-
nectomy.
10
A major pragmatic difficulty in assem-
bling our recommendations is that the literature is
heavily based on making predictions for resection
using absolute values of FEV
1
. This approach might
bias against older patients, people of small stature,
and female patients who might tolerate lower levels
of lung function. Although it is not possible to
recalculate percentage of predicted values from the
BTS data, an FEV
1
80% predicted also indicates
that the patient should be considered suitable for
pneumonectomy without further evaluation.
16
Interest in the diffusing capacity of the lung for
carbon monoxide (Dlco) as a useful marker of
operative risk was stimulated by a study of Ferguson
et al,
17
who related preoperative Dlco to postresec-
tion morbidity and mortality in 237 patients. Patients
were selected for surgery on the basis of clinical
evaluation and spirometry, but not the Dlco, which
was also measured. They found the preoperative
Dlco expressed as a percentage of predicted to have
a higher correlation with postoperative deaths than
the FEV
1
expressed as percentage of predicted, or
any other factor tested. They noted a Dlco of
60% predicted was associated with increased
mortality. Also, the risk of pulmonary complications
increased twofold to threefold with a Dlco 80%
normal.
Spirometry and Dlco measurements should be
viewed as complementary physiologic tests. If there
is evidence of interstitial lung disease on radio-
graphic studies or undue dyspnea on exertion, even
though the FEV
1
may be adequate, a Dlco should
be obtained. In a prospective study of 137 patients
with operable tumor, those with an FEV
1
80%
predicted, a Dlco 80% predicted, and no signif-
icant cardiac history were all suitable for pneumo-
nectomy.
16
There were no deaths in this group. In
Table 1Clinical Predictors of Increased Preoperative
Cardiovascular Risk*
Major
Unstable coronary syndromes
Recent myocardial infarction with evidence of important
ischemic risk by clinical symptoms or noninvasive study
Unstable or severe angina
Decompensated congestive heart failure
Significant arrhythmia
Severe valvular disease
Intermediate
Mild angina pectoris
Prior myocardial infarction by history or pathologic Q waves
Compensated or prior congestive heart failure
Diabetes mellitus
Minor
Advanced age
Abnormal ECG (left ventricular hypertrophy, left bundle-branch
block, ST-T abnormalities
Rhythm other than sinus rhythm
Low functional capacity (eg, inability to climb stairs)
History of stroke
Uncontrolled systemic hypertension
*Adapted from Eagle et al.
12
this study, patients with either an FEV
1
or a Dlco
80% predicted had additional physiologic testing
performed.
Recommendations
5. In patients being considered for lung cancer
resection, spirometry should be performed. If
the FEV
1
is 80% predicted normal or 2 L,
the patient is suitable for resection including
pneumonectomy without further evaluation. If
the FEV
1
is 1.5 L, the patient is suitable for
a lobectomy without further evaluation. Level
of evidence, fair; benefit, substantial; grade of
recommendation, B
resection, if there is evidence of interstitial lung
disease on radiographic studies or undue dys-
pnea on exertion, even though the FEV
1
might
be adequate, Dlco should be measured. Level
recommendation, B
resection, if either the FEV
1
or Dlco are
80% predicted, postoperative lung function
should be predicted through additional testing.
Level of evidence, fair; benefit, substantial;
grade of recommendation, B
Predicted Postoperative Values of
Lung Function
The extent of further evaluation in patients with
diminished pulmonary reserve depends on the ex-
tent of planned pulmonary resection: pneumonec-
tomy, lobectomy, wedge resection, or segmentec-
tomy. In patients with compromised lung function
preoperatively, it is therefore essential to estimate
the likely pulmonary reserve postresection. Ap-
proaches to obtaining the predicted postoperative
(ppo) lung function have relied on several different
methods to estimate the amount of functioning lung
tissue that would be lost along with the surgical
resection. The methods used, including ventilation
scans,
14,1821
perfusion scans,
6,14,1824
quantitative
CT,
25,26
and simply counting the number of seg-
ments to be removed,
23,27
seem to provide similar
quantitative estimates of ppo lung function. Recom-
mended approaches use a radionuclide perfusion
scan with Tc-labeled macroaggregates of albumin to
estimate the ppo FEV
1
and Dlco after pneumonec-
tomy and the number of segments remaining for
postlobectomy values.
10
The percentage of ppo
(%ppo) values for FEV
1
and Dlco are routinely
used instead of absolute values.
%ppo FEV
1
After Pneumonectomy
%ppo FEV
1
is calculated using the following
formula, which can also be used to calculate ppo and
%ppo Dlco:
ppoFEV1 preoperative FEV
1
(1 fraction of
total perfusion for the resected lung)
where ppo FEV
1
is expressed as percentage of
predicted to calculate the %ppo FEV
1
. The preop-
erative FEV
1
is taken as the best measure postbron-
chodilator. A quantitative radionuclide perfusion
scan is performed to measure the relative function of
each lung. Although several studies have demon-
strated good correlation between the actual postop-
erative FEV
1
and the ppo FEV
1
,
14,19,28
the %ppo
values estimated by the perfusion method may be up
to 10% less than actual measured values 3 months
postresection. This therefore errs on the side of
safety.
23,24,29
%ppo FEV
1
After Lobectomy
The value of %ppo FEV
1
is strongly correlated
with the actual postoperative FEV
1
when consider-
ing the number of segments to be removed at
operation.
14,27
Calculating the %ppo FEV
1
by the
number of segments removed is similar to the
method used for perfusion scan:
ppoFEV
1
preoperative FEV
1
(No. of segments
remaining/total No. of segments)
where ppo FEV
1
is expressed as a percentage of
predicted to give %ppo. The lungs have the follow-
ing 19 segments: right upper lobe (3 segments), right
middle lobe (2 segments), right lower lobe (5 seg-
ments), left upper lobe (3 segments), lingual (2
segments), and left lower lobe (4 segments). This
method can also be applied to segmentectomies
because lobectomy does not cause a significantly
greater loss of function when compared to segmen-
tectomy.
30
This same formula may be used to calcu-
late ppo and %ppo Dlco.
Olsen et al
6
suggested a threshold ppo FEV
1
of 0.8 L
as the lower limit for surgical resection. However,
Pate and colleagues
31
found that patients with a
mean ppo FEV
1
of 0.7 L tolerated thoracotomy for
lung cancer resection. This experience might have
reflected resection of less lung tissue than antici-
pated. The main objection to using an absolute value
of ppo FEV
1
as a threshold for operability is that it
might prevent older patients, small stature people,
and females patients, all of whom might tolerate a
lower absolute FEV
1
, from having a potentially
curative lung cancer resection. Consequently, estab-
lishing a threshold for lung function expressed as
%ppo rather than absolute ppo would be desirable.
Case series with small numbers of patients have
shown that perioperative risks increase substantially
when the %ppo FEV
1
is 40% of predicted nor-
mal.
18,23,24,3234
Markos et al
18
reported that three of
six patients with a %ppo FEV
1
40% died in the
perioperative period. Wahi et al
34
found a perioper-
ative mortality rate of 16% in patients with a %ppo
FEV
1
of 41%, vs 3% with those with better
predicted lung function. Pierce and colleagues
23
found that 5 of 13 patients with a %ppo FEV
1
40% died soon after operation, and Bolliger et al
24
reported that 2 of 4 patients with similar lung
function died of respiratory failure perioperatively.
Nakahara et al
35,36
found an especially high postop-
erative mortality rate (6 of 10 patients, 60%) when
the %ppo FEV
1
was 30%.
As a result of the observation by Ferguson et al
17
that the Dlco, expressed as the %ppo, was a strong
predictor of mortality, others have also found that
perioperative risks increase substantially when the
%ppo Dlco 40%.
17,18,23
Pierce et al
23
suggested
that a product of %ppo FEV
1
and %ppo Dlco
1,650 might serve as a more discriminating thresh-
old for perioperative risk assessment. Others have
made a similar observation.
37
Although a %ppo FEV
1
or Dlco 40% indicates
increased risk for perioperative complications, in-
cluding death, from lung cancer resection, these
patients can be successfully operated on. Ribas et al
37
described a selected group of 65 patients who met
these physiologic criteria but still underwent curative
intent lobectomy/wedge resection (n 44) or pneu-
monectomy (n 21). There were only four postop-
erative deaths (6.2% mortality rate), and cardiopul-
monary complications occurred in 31 patients
(47.7%). Although this study indicates that lung
cancer resection can be performed with an accept-
able perioperative risk even in patients with poor
lung function reserve, it is prudent to more thor-
oughly evaluate these patients prior to pulmonary
resection.
Recommendation
8. In patients with lung cancer being considered
for surgical resection, either a %ppo FEV
1
40% or a %ppo Dlco 40% indicate a high
risk for perioperative death and cardiopulmo-
nary complications. These patients should un-
dergo exercise testing preoperatively. Level of
recommendation, B
for surgical resection, either a product of %ppo
FEV
1
and %ppo Dlco 1,650 or a %ppo
FEV
1
30% indicate a very high risk for
perioperative death and cardiopulmonary com-
plications. These patients should be counseled
about nonoperative treatment options for their
lung cancer. Level of evidence, poor; benefit,
substantial; grade of recommendation, C
Exercise Testing
Cardiopulmonary Exercise Testing
Formal cardiopulmonary exercise testing (CPET)
is a sophisticated physiologic testing technique that
includes recording the exercise ECG, heart rate
response to exercise, minute ventilation, and oxygen
uptake per minute. Maximal oxygen consumption
(V
o
2
max) is calculated from this type of exercise test.
Algorithms for the preoperative physiologic assess-
ment of patients being considered for lung cancer
resection have incorporated use of CPET as an
adjunct to estimating the %ppo FEV
1
and Dlco.
10,16
However, there remains the practical difficulty as to
when to recommend CPET, as it is readily acknowl-
edged not to be widely available. We have taken the
view of the BTS recommendations that exercise
testing should be performed if perfusion lung scan-
ning and calculation of %ppo FEV
1
and Dlco
confirms borderline function ( 40%). It is sug-
gested that if an institution is not equipped to
perform CPET, patients in high-risk groups should
be sent to a specialist center for this evaluation.
Numerous studies have examined the relationship
between V
o
2
max and perioperative complications.
Risk for perioperative complications can generally be
stratified by V
o
2
max. Patients with a preoperative
V
o
2
max of 20 mL/kg/min are not at increased risk
of complications or death.
1,18,31,3841
Those patients
with V
o
2
max 10 mL/kg/min have a very high risk
for postoperative complications.
3,10,16,24,32,42,43
Be-
chard and Wetstein
42
reported that 2 of 7 patients
with V
o
2
max 10 mL/kg/min died in the postoper-
ative period; Olsen et al
43
described deaths in 5 of 11
patients; and Holden and colleagues
32
noted deaths
in 2 of 4 patients. V
o
2
max 15 mL/kg/min indicates
an increased risk of perioperative complica-
tions.
1,3,44,45
However, it should be noted that not all
authors agree that perioperative complication rates
can that clearly be stratified by V
o
2
max.
37
In patients with borderline lung function, V
o
2
max
may be helpful in further evaluating the risk for
perioperative complications. Morice et al
40
showed
that in subjects with a ppo FEV
1
33%, eight
patients underwent lobectomy because V
o
2
max
15 mL/kg/min was achieved, and no fatal compli-
cations occurred. Other studies have made similar
observations.
4,31
In patients with both a low %ppo
FEV
1
and a low %ppo Dlco (both 40% predict-
ed), V
o
2
max 15 mL/kg/min indicates a group with
a very high surgical risk.
44
Pulmonary Artery Pressures and DLCO
Measurements of pulmonary arterial pressure dur-
ing exercise have not proven to be helpful in pre-
dicting which patients will acquire perioperative
complications.
37,43
A study by Wang et al
46
found
that measuring Dlco during exercise was a better
predictor of perioperative risk than V
o
2
max, but is a
technically demanding technique and not readily
available.
Stair Climbing and Walking Tests
If CPET is unavailable, another type of exercise
test should be considered. Stair climbing has histor-
ically been used as a surrogate CPET. If patients
were able to climb three flights of stairs, they were
considered suitable candidates for lobectomy. Pneu-
monectomy candidates were expected to be able to
climb five flights of stairs. This approach was found
to correlate with lung function: climbing three flights
reflected an FEV
1
1.7 L and five flights indicated
an FEV
1
2 L.
47
However, stair climbing is not
performed in a standardized manner. The duration
of the test, speed of ascent, number of steps per
flight, height of each step, and criteria for stopping
the test have not been well defined. However, in
general terms, patients who can climb five flights of
stairs will have a V
o
2
max 20 mL/kg/min. Con-
versely, patients who cannot climb one flight of stairs
will have a V
o
2
max 10 mL/kg/min.
48
Other surrogate CPETs are the shuttle walk and
the 6-min walk, but data on the value of these tests
in predicting V
o
2
max are limited.
49
The shuttle walk
requires that patients walk back and forth between
two markers set 10 m apart. The walking speed is
paced by an audio signal and the walking speed is
increased each minute in a graded fashion. The end
of the test occurs when the patient is too breathless
to maintain the required speed. In one study, inabil-
ity to complete 25 shuttles on two occasions sug-
gested a V
o
2
max of 10 mL/kg/min.
50
For the
6-min walk, patients are instructed to walk as far as
possible in the time allotted. Rest during the test is
permissible. Interpretation of the distance walked in
6 min is currently not well standardized.
51
Desaturation
Desaturation during an exercise test has been
associated with increased risk for perioperative com-
plications.
18,23,37,52
Greater than 4% desaturation in-
dicates an increased risk for perioperative complica-
tions.
10
Recommendations
for lung resection, V
o
2
max 10 mL/kg/min
indicates a very high risk for perioperative
death and cardiopulmonary complications.
These patients should be counseled about
nonoperative treatment options for their lung
cancer. Level of evidence, poor; benefit, sub-
stantial; grade of recommendation, C
11. Patients being considered for lung cancer
resection who have V
o
2
max 15 mL/kg/min
and both a %ppo FEV
1
and Dlco 40%
should be considered at very high risk for
perioperative death and cardiopulmonary
complications. These patients should be coun-
seled about nonoperative treatment options
for their lung cancer. Level of evidence, poor;
benefit, substantial; grade of recommenda-
tion, C
resection who walk 25 shuttles on two
shuttle walks or one flight of stairs should
be considered at very high risk for periopera-
tive death and cardiopulmonary complica-
tions. These patients should be counseled
about nonoperative treatment options for
their lung cancer. Level of evidence, poor;
tion, C
Arterial Blood Gas Tensions
Historically, hypercapnea (Paco
2
45 mm Hg)
has been quoted as an exclusion criterion for lung
resection.
16,53,54
This recommendation was made on
the basis of the association of hypercapnea with poor
ventilatory function.
55
The few studies that address
this issue, however, suggest that preoperative hyper-
capnea is not an independent risk factor for in-
creased perioperative complications. Stein et al
56
showed hypercapnea was associated with serious
postoperative respiratory difficulties in five patients;
there were no deaths, despite a Paco
2
45 mm Hg.
In two series of lung cancer patients undergoing
surgery,
57,58
perioperative complications were not
higher in patients with preoperative hypercapnea.
Preoperative hypoxemia, an arterial oxygen satura-
tion (Sao
2
) 90%, has been associated with an
increased risk of postoperative complications.
52
Recommendations
surgery, Paco
2
45 mm Hg is not an inde-
pendent risk factor for increased perioperative
complications. However, further physiologic
testing is advised. Level of evidence, poor;
tion, C
surgery, Sao
2
90% indicates an increased
risk for perioperative complications, and fur-
ther physiologic testing is advised. Level of
recommendation, C
Methods To Reduce Perioperative Risks
LVRS
LVRS for patients with severe emphysema is a
controversial procedure. Some reports document
substantial improvements in lung function, exercise
capability, and quality of life in highly selected
patients with emphysema following LVRS.
59
How-
ever, recently published results from a larger pro-
spective, randomized, controlled trial indicate an
increased mortality rate after LVRS in patients with
either homogenous emphysema or a low Dlco.
60
Case series of patients referred for LVRS indicate
that perhaps 3 to 6% of these patients may have
coexisting lung cancer.
61,62
Anecdotal experience
from these case series suggest that patients with
extremely poor lung function can tolerate combined
LVRS and resection of the lung cancer with an
acceptable mortality rate and surprisingly good post-
operative outcomes.
6167
McKenna et al
61
reported 11 cases of lung cancer
(3%) in their group of 325 patients referred for
LVRS. These 11 patients had an average preopera-
tive FEV
1
of 0.65 L (range of FEV
1
percent pre-
dicted of 12 to 29%). None of these patients would
have been acceptable for lung cancer resection based
on traditional criteria but all underwent combined
LVRS and resection of stage 1 lung cancers, either
with lobectomy or wedge resection. There were no
deaths or major complications; lung function and
exercise capability were improved postoperatively.
There have been other promising reports on the
combination of LVRS and lung cancer resection in
patients with very poor lung function.
6267
Combin-
ing LVRS and lung cancer resection should probably
be limited to those patients with heterogeneous
emphysema, particularly emphysema limited to the
lobe containing the tumor.
67,68
Smoking Cessation
While smoking is strongly associated with lung
cancer, it is also associated with an increased risk of
postoperative complications. However there is little
clinical evidence to suggest that smoking cessation
before surgery is beneficial. One study in cardiac
patients found that cessation of smoking 8 weeks
prior to surgery decreased the perioperative compli-
cation rate; this is an impractical length of time in the
context of surgery for lung cancer.
69
Pulmonary Rehabilitation
As yet, there are no robust data to recommend the
routine use of preoperative pulmonary rehabilitation
for patients with lung cancer.
Recommendation
15. In patients with very poor lung function,
combined LVRS and lung cancer resection
may be considered if emphysema is heteroge-
neous and involves primarily the lobe to be
resected. Level of evidence, poor; benefit,
Summary
Patients with lung cancer often have concomitant
obstructive lung disease and/or atherosclerotic car-
diovascular disease as a consequence of their smok-
ing habit. These diseases may place these patients at
increased risk for perioperative complications, in-
cluding death, after lung cancer resection. A careful
preoperative physiologic assessment will be useful to
identify those patients at increased risk and to enable
an informed decision by the patient about the ap-
propriate therapeutic approach to treating their lung
cancer. This preoperative risk assessment must be
placed in the context that lung cancer surgery is the
most effective currently available treatment for this
disease.
1. Patients with lung cancer should be seen by
physicians interested in the management of
this disease. Level of evidence, fair; benefit,
substantial; grade of recommendation, B
2. Patients with lung cancer should be assessed
by a multidisciplinary team for their suitability
for surgery; there should be liaison between
the chest physician, thoracic surgical team,
and oncologist in all cases prior to surgery.
3. Patients with lung cancer should not be de-
nied lung resection surgery on the grounds of
age alone. Level of evidence, fair; benefit,
4. Patients with lung cancer undergoing surgery
should have a preoperative cardiologic evalu-
ation carried out according to established
guidelines. Level of evidence, fair; benefit,
resection, spirometry should be performed. If
the FEV
1
is 80% predicted normal or 2 L,
the patient is suitable for resection including
pneumonectomy without further evaluation.
If the FEV
1
is 1.5 L, the patient is suitable
for a lobectomy without further evaluation.
resection, if there is evidence of interstitial
lung disease on radiographic studies or undue
dyspnea on exertion, even though the FEV
1
might be adequate, Dlco should be mea-
sured. Level of evidence, fair; benefit, sub-
resection, if either the FEV
1
or Dlco are
80% predicted, postoperative lung function
should be predicted through additional test-
ing. Level of evidence, fair; benefit, substan-
for surgical resection, either %ppo FEV
1
40% or %ppo Dlco 40% indicate a high
risk for perioperative death and cardiopulmo-
nary complications. These patients should un-
dergo exercise testing preoperatively. Level of
recommendation, B
for surgical resection, either a product of
%ppo FEV
1
and %ppo Dlco 1,650 or
%ppo FEV
1
30% indicate a very high risk
for perioperative death and cardiopulmonary
seled about nonoperative treatment options.
for lung resection, V
o
2
max 10 mL/kg/min
indicates a very high risk for perioperative
death and cardiopulmonary complications.
These patients should be counseled about
nonoperative treatment options. Level of evi-
dence, poor; benefit, substantial; grade of
recommendation, C
resection who have V
o
2
max 15 mL/kg/min
and both %ppo FEV
1
and Dlco 40%
resection who walk 25 shuttles on two
shuttle walks or less than one flight of stairs
surgery, Paco
2
45 mm Hg is not an inde-
pendent risk factor for increased perioperative
complications; however, further physiologic
testing is advised. Level of evidence, poor;
tion, C
surgery, Sao
2
90% indicates an increased
risk for perioperative complications, and fur-
ther physiologic testing is advised. Level of
recommendation, C
15. In patients with very poor lung function,
combined LVRS and lung cancer resection
may be considered if emphysema is heteroge-
neous and involves primarily the lobe to be
resected. Level of evidence, poor; benefit,
References
1 Walsh GL, Morice RC, Putnam JB Jr, et al. Resection of lung
cancer is justified in high-risk patients selected by exercise
oxygen consumption. Ann Thorac Surg 1994; 58:704710
2 Damhuis RA, Schutte PR. Resection rates and postoperative
mortality in 7,899 patients with lung cancer. Eur Respir J
1996; 9:710
3 Brutsche MH, Spiliopoulos A, Bolliger CT, et al. Exercise
capacity and extent of resection as predictors of surgical risk
in lung cancer. Eur Respir J 2000; 15:828832
4 Nezu K, Kushibe K, Tojo T, et al. Recovery and limitation of
exercise capacity after lung resection for lung cancer. Chest
1998; 113:15111516
5 Ali MK, Ewer MS, Atallah MR, et al. Regional and overall
pulmonary function changes in lung cancer. Thorac Cardio-
vasc Surg 1983; 86:18
6 Olsen GN, Block AJ, Tobias JA. Prediction of postpneumo-
nectomy pulmonary function using quantitative macroaggre-
gate lung scanning. Chest 1974; 66:1316
7 Muers MF, Haward RA. Management of lung cancer. Thorax
1996; 51:557560
8 Brown JS, Eraut D, Trask C, et al. Age and the treatment of
lung cancer. Thorax 1996; 51:564568
9 Expert Advisory Group. A policy framework for commission-
ing cancer services. London, UK: Department of Health,
1995
10 BTS guidelines: guidelines on the selection of patients with
lung cancer for surgery. Thorax 2001; 56:89108
11 Yellin A, Hill LR, Lieberman Y. Pulmonary resections in
patients over 70 years of age. Isr J Med Sci 1985; 21:833840
12 Eagle KA, Brundage BH, Chaitman BR, et al. Guidelines for
perioperative cardiovascular evaluation for noncardiac sur-
gery: report of the American College of Cardiology/American
Heart Association Task Force on Practice Guidelines. J Am
Coll Cardiol 1996; 27:910948
13 Boushy SF, Billig DM, North LB, et al. Clinical course
related to preoperative and postoperative pulmonary function
in patients with bronchogenic carcinoma. Chest 1971; 59:
383391
14 Wernly JA, DeMeester TR, Kirchner PT, et al. Clinical value
of quantitative ventilation-perfusion lung scans in the surgical
management of bronchogenic carcinoma. Thorac Cardiovasc
Surg 1980; 80:535543
15 Miller JI Jr. Physiologic evaluation of pulmonary function in
the candidate for lung resection. Thorac Cardiovasc Surg
1993; 105:347351
16 Wyser C, Stulz P, Soler M, et al. Prospective evaluation of an
algorithm for the functional assessment of lung resection
candidates. Am J Respir Crit Care Med 1999; 159:14501456
17 Ferguson MK, Little L, Rizzo L, et al. Diffusing capacity
predicts morbidity and mortality after pulmonary resection.
Thorac Cardiovasc Surg 1988; 96:894900
18 Markos J, Mullan BP, Hillman DR, et al. Preoperative
assessment as a predictor of mortality and morbidity after
lung resection. Am Rev Respir Dis 1989; 139:902910
19 Corris PA, Ellis DA, Hawkins T, et al. Use of radionuclide
scanning in the preoperative estimation of pulmonary func-
tion after pneumonectomy. Thorax 1987; 42:285291
20 Ali MK, Mountain CF, Ewer MS, et al. Predicting loss of
pulmonary function after pulmonary resection for broncho-
genic carcinoma. Chest 1980; 77:337342
21 Bria WF, Kanarek DJ, Kazemi H. Prediction of postoperative
pulmonary function following thoracic operations. Thorac
Cardiovasc Surg 1983; 86:186192
22 Giordano A, Calcagni ML, Meduri G, et al. Perfusion lung
scintigraphy for the prediction of postlobectomy residual
pulmonary function. Chest 1997; 111:15421547
23 Pierce RJ, Copland JM, Sharpe K, et al. Preoperative risk
evaluation for lung cancer resection. Am J Respir Crit Care
Med 1994; 150:947955
24 Bolliger CT, Wyser C, Roser H, et al. Lung scanning and
exercise testing for the prediction of postoperative perfor-
mance in lung resection candidates at increased risk for
complications. Chest 1995; 108:341348
25 Wu MT, Chang JM, Chiang AA, et al. Use of quantitative CT
to predict postoperative lung function in patients with lung
cancer. Radiology 1994; 191:257262
26 Wu MT, Pan HB, Chiang AA, et al. Prediction of postoper-
ative lung function in patients with lung cancer. AJR Am J
Roentgenol 2002; 178:667672
27 Zeiher BG, Gross TJ, Kern JA, et al. Predicting postoperative
pulmonary function in patients undergoing lung resection.
Chest 1995; 108:6872
28 Ladurie ML, Ranson-Bitker B. Uncertainties in the expected
value for forced expiratory volume in one second after
surgery. Chest 1986; 90:222228
29 Bolliger C, Jordan P, Soler M, et al. Pulmonary function and
exercise capacity after lung resection. Eur Respir J 1996;
9:415421
30 Takizawa T, Haga M, Yagi N, et al. Pulmonary function after
segmentectomy for small peripheral carcinoma of the lung.
31 Pate P, Tenholder MF, Griffin JP, et al. Preoperative assess-
ment of the high-risk patient for lung resection. Ann Thorac
Surg 1996; 61:14941500
32 Holden DA, Rice TW, Stelmach K, et al. Exercise testing,
6-min walk, and stair climb in the evaluation of patients at
high risk for pulmonary resection. Chest 1992; 102:1774
1779
33 Gass GD, Olsen GN. Preoperative pulmonary function test-
ing to predict postoperative morbidity and mortality. Chest
1986; 89:127135
34 Wahi R, McMurtry MJ, DeCaro LF, et al. Determinants of
perioperative morbidity and mortality after pneumonectomy.
Ann Thorac Surg 1989; 48:3337
35 Nakahara K, Monden Y, Ohno K, et al. A method for
predicting postoperative lung function and its relation to
postoperative complications in patients with lung cancer. Ann
Thorac Surg 1985; 39:260265
36 Nakahara K, Ohno K, Hashimoto J, et al. Prediction of
postoperative respiratory failure in patients undergoing lung
resection for lung cancer. Ann Thorac Surg 1988; 46:549552
37 Ribas J, Diaz O, Barbera JA, et al. Invasive exercise testing in
the evaluation of patients at high-risk for lung resection. Eur
Respir J 1998; 12:14291435
38 Bechard D. Pulmonary function testing. In: LoCicero J III,
ed. Diagnostic procedures in thoracic diseases: chest surgery
clinics. Philadelphia, PA: W.B. Saunders, 1992; 565586
39 Richter Larsen K, Svendsen UG, Milman N, et al. Exercise
testing in the preoperative evaluation of patients with bron-
chogenic carcinoma. Eur Respir J 1997; 10:15591565
40 Morice RC, Peters EJ, Ryan MB, et al. Exercise testing in the
evaluation of patients at high risk for complications from lung
resection. Chest 1992; 101:356361
41 Bolliger CT, Soler M, Stulz P, et al. Evaluation of high-risk
lung resection candidates: pulmonary haemodynamics vs
exercise testing. Respiration 1994; 61:181186
42 Bechard D, Wetstein L. Assessment of exercise oxygen
consumption as preoperative criterion for lung resection. Ann
Thorac Surg 1987; 44:344349
43 Olsen GN, Weiman DS, Bolton JWR, et al. Submaximal
invasive exercise testing and quantitative lung scanning in the
evaluation for tolerance of lung resection. Chest 1989; 95:
267273
44 Bolliger CT, Jordan P, Soler M, et al. Exercise capacity as a
predictor of postoperative complications in lung resection
candidates. Am J Respir Crit Care Med 1995; 151:14721480
45 Smith TP, Kinasewitz GT, Tucker WY, et al. Exercise capac-
ity as a predictor of post-thoracotomy morbidity. Am Rev
Respir Dis 1984; 129:730734
46 Wang JS, Abboud RT, Evans KG, et al. Role of CO diffusing
capacity during exercise in the preoperative evaluation for
lung resection. Am J Respir Crit Care Med 2000; 162:1435
1444
47 Bolton JWR, Weiman DS, Haynes JL, et al. Stair climbing as
an indicator of pulmonary function. Chest 1987; 92:783787
48 Pollock M, Roa J, Benditt J, et al. Estimation of ventilatory
reserve by stair climbing. Chest 1993; 104:13781383
49 Solway S, Brooks D, Lacasse Y, et al. A qualitative systematic
overview of the measurement properties of functional walk
tests used in the cardiorespiratory domain. Chest 2001;
119:256270
50 Singh SJ, Morgan MD, Scott S, et al. Development of a
shuttle walking test of disability in patients with chronic
airway obstruction. Thorax 1992; 47:10191024
51 ATS statement: guidelines for the six-minute walk test. Am J
52 Ninan M, Summers KE, Landreneau RJ, et al. Standardised
exercise oximetry predicts postpneumonectomy outcome.
53 Celli BR. What is the value of preoperative pulmonary
function testing? Med Clin North Am 1993; 77:309325
54 Zibrak JD, ODonnell CR, Marton K. Indications for pulmo-
nary function testing. Ann Intern Med 1990; 112:763771
55 Tisi GM. Preoperative evaluation of pulmonary function. Am
Rev Respir Dis 1979; 119:293310
56 Stein M, Koota G, Simon M. Pulmonary evaluation of surgical
patients. JAMA 1962; 181:765770
57 Kearney DJ, Lee TH, Reilly JJ, et al. Assessment of operative
risk in patients undergoing lung resection: importance of
predicted pulmonary function. Chest 1994; 105:753759
58 Harpole DH, Liptay MJ, DeCamp MM Jr, et al. Prospective
analysis of pneumonectomy: risk factors for major morbidity
and cardiac dysrhythmias. Ann Thorac Surg 1996; 61:977982
59 Geddes D, Davies M, Koyama H, et al. Effect of lung-
volume-reduction surgery in patients with severe emphy-
sema. N Engl J Med 2000; 343:239245
60 National Emphysema Treatment Trial Research Group. Pa-
tients at high risk of death after lung-volume-reduction
surgery. N Engl J Med 2001; 345:10751083
61 McKenna RJ, Fischel RJ, Brenner M, et al. Combined
operations for lung volume reduction surgery and lung
cancer. Chest 1996; 110:885888
62 DeMeester SR, Patterson GA, Sundaresan RS, et al. Lobec-
tomy combined with volume reduction for patients with lung
cancer and advanced emphysema. Thorac Cardiovasc Surg
1998; 115:681688
63 Clark T, Martinez F, Paine R, et al. Lung volume reduction
surgery alters management of pulmonary nodules in patients
with severe COPD. Chest 1997; 112:14941500
64 DeRose JJ, Argenziano M, El-Amir N, et al. Lung reduction
operation and resection of pulmonary nodules in patients with
severe emphysema. Ann Thorac Surg 1998; 65:314318
65 Rozenshtein A, White CS, Austin JHM, et al. Incidental lung
carcinoma detected at CT in patients selected for lung
volume reduction surgery to treat severe pulmonary emphy-
sema. Radiology 1998; 207:487490
66 Allen GM, DeRose JJ. Pulmonary nodule resection during
lung volume reduction surgery. AORN J 1997; 66:808818
67 Edwards JG, Duthie DJR, Waller DA. Lobar volume reduc-
tion surgery: a method of increasing the lung cancer resection
rate in patients with emphysema. Thorax 2001; 56:791795
68 Mentzer SJ, Swanson SJ. Treatment of patients with lung
cancer and severe emphysema. Chest 1999; 116:477S479S
69 Warner MA, Divertie MB, Tinker JH. Preoperative cessation
of smoking and pulmonary complications in coronary artery
bypass patients. Anesthesiology 1984; 60:380383
2003;123;105-114 Chest
for Resectional Surgery
The Physiologic Evaluation of Patients With Lung Cancer Being Considered
& Services
Updated Information
S
http://www.chestjournal.org/cgi/content/full/123/1_suppl/105
References
S#BIBL
free at:
Citations
S#otherarticles
Reprints
2003;123;115-128 Chest
Gilbert Schreiber and Douglas C. McCrory
Lung Cancer: Summary of Published Evidence
Performance Characteristics of Different Modalities for Diagnosis of Suspected
ISSN: 0012-3692.
Performance Characteristics of
Different Modalities for Diagnosis of
Suspected Lung Cancer*
Summary of Published Evidence
Gilbert Schreiber, MD, FCCP; and Douglas C. McCrory, MD, MHS
Study objectives: To determine the test performance characteristics of various modalities for the
diagnosis of suspected lung cancer.
Design, setting, and participants: A systematic search of MEDLINE, HealthStar, and Cochrane
Library databases to July 2001 and print bibliographies was performed to identify studies
comparing the results of sputum cytology, bronchoscopy, transthoracic needle aspirate (TTNA),
or biopsy with histologic reference standard diagnoses among at least 50 patients with suspected
lung cancer.
Measurement and results: For sputum cytology, the pooled specificity was 0.99 and the pooled
sensitivity was 0.66, but sensitivity was higher for central lesions than for peripheral lesions (0.71
vs 0.49, respectively). Studies on bronchoscopic procedures provided data only on diagnostic
yield (sensitivity). The diagnosis of endobronchial disease by bronchoscopy in 30 studies showed
the highest sensitivity for endobronchial biopsy (0.74), followed by cytobrushing (0.59) and
washing (0.48). The sensitivity for all modalities combined was 0.88. Thirty studies reported on
peripheral lesions. Cytobrushing demonstrated the highest sensitivity (0.52), followed by trans-
bronchial biopsy (0.46) and BAL/washing (0.43). The overall sensitivity for all modalities was 0.69.
Peripheral lesions < 2 cm or > 2 cm in diameter showed sensitivities of 0.33 and 0.62,
respectively. Updating a previous meta-analysis with 19 studies revealed a pooled sensitivity of
0.90 for TTNA. A trend toward lower sensitivity was noted for lesions that were < 2 cm in
diameter. The accuracy in differentiating between small cell and non-small cell cytology for the
various diagnostic modalities was 0.98, with individual studies ranging from 0.94 to 1.0. The
average false-positive and false-negative rates were 0.09 and 0.02, respectively.
Conclusions: The sensitivity of bronchoscopy is high for endobronchial disease and poor for
peripheral lesions that are < 2 cm in diameter. The sensitivity of TTNA is excellent for malignant
disease. The distinction between small cell lung cancer and non-small cell lung cancer by cytology
appears to be accurate. (CHEST 2003; 123:115S128S)
Key words: biopsy; bronchoscopy; cytodiagnosis; lung neoplasm; predictive value; sensitivity; specificity; yield
Abbreviations: CI confidence interval; TBNA transbronchial needle aspiration; TTNA transthoracic needle
aspiration
T
his article summarizes and critically examines the
performance characteristics of the various mo-
dalities for the accurate histopathologic diagnosis of
suspected lung cancer. It provides the background
evidence for the Clinical Practice Guidelines on
Lung Cancer that are outlined in this supplement. A
joint panel from the American College of Chest
Physicians assisted in the design, conduct, and de-
velopment of this article.
Lung cancer is usually suspected on the basis of an
abnormal radiographic imaging study, often in con-
junction with symptoms caused by either local or
systemic effects of the tumor. The modality selected
to diagnose a suspected lung cancer is based on the
size and location of the primary tumor in the lung,
the presence of potential metastatic spread, and the
anticipated treatment plan.
The main goals in selecting a specific diagnostic
modality are as follows: (1) to maximize the yield of
the selected procedure for both diagnosis and stag-
*From the Department of Medicine, Duke University Medical
Center, Durham, NC.
Correspondence to: Gilbert Schreiber, MD, FCCP, Department
of Pulmonary Medicine, University of Texas MD Anderson
Cancer Center, Box 403, 1515 Holcombe Blvd, Houston, TX
77030; e-mail: egschreib@mdanderson.org
ing; and (2) to avoid unnecessary invasive tests for
the patient, with special attention to the projected
treatment plan.
Sputum cytology, bronchoscopic techniques,
transthoracic needle biopsies, and surgical biopsy
(resection) define the main modalities employed in
the diagnosis of bronchogenic carcinoma. Positron
emission tomography has emerged as a helpful ad-
junct in both the diagnosis and staging of lung
cancer.
In this article, we focus on techniques for the
histologic and cytologic diagnosis of lung lesions. The
noninvasive evaluation of pulmonary nodules will be
covered elsewhere in this supplement in the article
reporting on single pulmonary nodules. Definitive
diagnostic procedures that are aimed at mediastinal
lymph nodes or extrathoracic tumors will be covered
in the articles on staging.
In several panel discussions, the American College of Chest
Physicians Committee on Clinical Practice Guidelines for Lung
Cancer formulated the following four key questions on the
diagnostic workup of lung cancer that were to be answered by a
comprehensive critical review of the published evidence:
1. What are the performance characteristics (sensitivity and
specificity) for sputum cytology for the diagnosis of lung
cancer with special consideration for the location of the
tumor (central vs peripheral)?
specificity) of flexible bronchoscopy and its ancillary pro-
cedures (biopsy, cytobrushing, washing, transbronchial
needle aspiration [TBNA], and BAL) for the diagnosis of
central (endobronchial), as opposed to peripheral, tumors
and for peripheral lesions 2 cm and 2 cm in diameter?
specificity) for transthoracic needle aspiration (TTNA) as a
diagnostic modality with particular emphasis on the size
and the location of the suspected cancer?
4. What is the diagnostic error rate when differentiating
between non-small cell lung cancer and small cell lung
cancer that is generated by various diagnostic techniques
(bronchoscopy and sputum cytology)?
To address these questions, we conducted a computerized
search of the MEDLINE bibliographic database from 1966 to
July 2001, HealthStar, and the Cochrane Library. We searched
using the terms lung neoplasm, bronchial neoplasm, bronchos-
copy, biopsy, needle, sputum, cytodiagnosis, yield, predictive
value of tests, and sensitivity and specificity. In addition, we
searched the reference lists of included studies, practice guide-
lines, systematic reviews, and meta-analyses.
We selected studies of at least 50 patients with suspected lung
cancer that compared test results with a reference standard
consisting of pathology/histology, definitive cytologic diagnosis, or
radiographic follow-up of at least 1 year. We considered the
following diagnostic tests: sputum cytologic examination (expec-
torated or aspirated, spontaneous or induced); flexible bronchos-
copy (including any of biopsy, brushing, washing, TBNA, or
BAL); and TTNA. Studies were required to report sufficient data
to permit completion of a 2 2 table comparing test results with
a reference standard diagnosis. If too few studies met this
criterion, then we identified studies that described the diagnostic
yield (sensitivity) among patients with lung cancer. When possi-
ble, diagnostic performance was estimated separately for patients
with central (endobronchial) lesions, peripheral lesions 2 cm in
diameter, and peripheral lesions 2 cm in diameter.
Results
Key Question 1: What Are the Performance
Characteristics for Sputum Cytology for the
Diagnosis of Lung Cancer With Special
Consideration for the Location of the Tumor
(Central vs Peripheral)?
We found few studies describing the accuracy of
sputum cytology as a result of our computerized
bibliographic literature search, but we identified
many among the citations of four review articles.
14
Problems With the Published Literature on Spu-
tum Cytology: Studies describing the sensitivity and
specificity of sputum cytology are described in Table
1. Most of the studies involved the identification of
patients from cytology laboratory samples without
regard to the indication for sputum cytology testing.
Several studies described the study population indi-
cations as including the evaluation of suspected lung
cancer and screening in patients with COPD. Few
studies have evaluated both the sensitivity and spec-
ificity of sputum cytology in patients with suspected
lung cancer.
5
Some studies of patients undergoing
bronchoscopy have reported the results of prebron-
choscopy or postbronchoscopy sputum testing; how-
ever, all of these studies (reviewed for key question 1)
were limited to patients with proven lung malignancies
and thus could not describe the test specificity.
Studies of sputum cytology that did attempt to
verify test-negative subjects are reviewed, with the
exception of screening trials of healthy or high-risk
individuals. (These studies are to be covered in the
article on screening elsewhere in this supplement.)
Performance Characteristics of Sputum Cytol-
ogy for the Diagnosis of Suspected Lung Cancer:
Table 1 shows 16 studies.
2,519
Sensitivity ranges
from 0.42 to 0.97, and specificity ranges from 0.68
to 1.0. The pooled sensitivity is 0.66, and the
pooled specificity is 0.99. The single study con-
ducted in patients who were evaluated for sus-
pected lung cancer
5
had a sensitivity of 0.87 and a
specificity of 0.90. After pooling all studies, re-
gardless of the indication for sputum testing, the
false- positive test result rate was 0.09, and the
false-negative test result rate was 0.06.
Some of the bronchoscopy studies included under
key question 1 described the sensitivity of prebron-
choscopy sputum examination. These studies have
the advantage that all patients were suspected of
having lung cancer and, thus, more closely approxi-
mate the population of interest. Table 2 describes
the sensitivity of prebronchoscopy sputum examina-
tion in eight studies.
4,2026
The sensitivity ranges
from 0.10 to 0.74, with an average value of 0.22.
Effect of Location on the Sensitivity of Sputum
Cytology: The effect of location (central vs periph-
eral) of lung nodules or masses on the sensitivity of
sputum cytology has been described in 17 studies
(Table 3).
2,5,1012,18,22,2736
Most studies showed de-
creased sensitivity for peripherally located masses,
but a few showed no such difference. On average,
sensitivity was 0.71 for central lesions and 0.49 for
peripheral lesions.
Bo cking et al
2
have shown that the sensitivity
of sputum cytology for detecting lung cancer is
highly dependent on the number of sputum spec-
imens collected per patient, ranging from approx-
imately 0.68 for a single specimen, to 0.78 for two
specimens, to 0.85 to 0.86 for three or more
specimens.
Sputum Cytology and Associated Clinical Find-
ings: Another approach to association with a positive
sputum diagnosis was described by predictive mod-
eling in a large cohort of patients tested with sputum
cytology.
37
Patient characteristics associated with
positive cytologic diagnosis using sputum were the
following: bloody sputum; low FEV
1
values; large
lung tumors (ie, 2.4 cm); centrally located tumors;
and squamous cell cancers.
Difficulties in Summarizing the Evidence on
Sputum Cytology: Studies of the accuracy of spu-
tum cytology for the diagnosis of lung cancer are
difficult to summarize because of a variety of
methodological problems. The studies show highly
variable estimates of sensitivity and no clear rea-
sons for the variation. There is evidence to suggest
that the number of sputum samples and the
specimen adequacy are strongly related to the
sensitivity of the technique. We have insufficient
detail about these features to determine whether
these factors explain the heterogeneity of the test
accuracy results.
Furthermore, between-study differences in the
threshold for considering cytology positive with
regard to the category of suspicious, and whether
insufficient specimens were excluded or classified as
negative, also may have a profound influence on
calculated sensitivity. These details are likewise un-
available for most studies.
Table 2Sensitivity of Prebronchoscopy Sputum for
Diagnosis of Bronchogenic Carcinoma
Study/Year
Patients,
No.
True-
Positives
False-
Negatives Sensitivity
Sing et al
4
/1997 231 93 138 0.40
Wagner et al
20
/1989 67 17 50 0.25
Schenk et al
21
/1987 75 10 65 0.13
Pilotti et al
22
/1982 129 95 34 0.74
Chaudhary et al
23
/1978 114 28 86 0.25
Chopra et al
24
/1977 52 9 43 0.17
Hattori et al
25
/1971 189 19 170 0.10
Oswald et al
26
/1971 1,598 266 1,332 0.17
Total 2,455 537 1,918 0.22
Table 1Sensitivity and Specificity of Sputum Cytology for Diagnosis of Bronchogenic Carcinoma
Study/Year Patients, No. Indication Sensitivity Specificity False-Positive Rate False-Negative Rate Prevalence
Bo cking et al
2
/1992 1,888 Mixed 0.86 1.00 0.04 0.02 0.12
Kern
6
/1988 1,289 Mixed 0.97 0.68 0.20 0.06 0.57
Risse et al
7
/1985 1,830 Mixed 0.60 0.98 0.11 0.08 0.17
Johnston and Bossen
8
/1981 9,892 Mixed 0.44 1.00 0.03 0.03 0.05
Jay et al
5
/1980 224 Lung mass 0.87 0.90 0.21 0.06 0.31
Yoneyama and Canlas
9
/1978 547 Mixed 0.83 1.00 0.04 0.02 0.12
Gagneten et al
10
/1976 506 Mixed 0.57 0.99 0.01 0.30 0.50
Rosa et al
11
/1973 1,003 Mixed 0.71 1.00 0.01 0.15 0.38
Dahlgren and Lind
12
/1972 121 Mixed 0.42 0.95 0.02 0.76 0.83
Koss et al
13
/1964 1,307 Mixed 0.71 0.98 0.12 0.06 0.17
Hinson and Kuper
14
/1963 528 Mixed 0.60 0.97 0.06 0.24 0.43
Russell et al
15
/1963 3,440 Mixed 0.51 1.00 0.02 0.07 0.13
Allen and Whittlesey
16
/1960 254 Mixed 0.90 1.00 0.00 0.06 0.41
Koss
17
/1958 607 Mixed 0.60 0.98 0.07 0.11 0.24
Spjut et al
18
/1955 4,933 Mixed 0.53 1.00 0.00 0.04 0.09
Liebow et al
19
/1948 108 Mixed 0.43 0.95 0.12 0.33 0.45
Total 28,477 0.66 0.99 0.09 0.06 0.15
Characteristics (Sensitivity and Specificity) of
Flexible Bronchoscopy and Its Ancillary Procedures
(Biopsy, Cytobrushing, Washing, TBNA, and BAL)
for the Diagnosis of Central (Endobronchial), as
Opposed to Peripheral, Tumors and for Peripheral
Lesions 2 cm and 2 cm in Diameter?
A comprehensive literature search on studies pub-
lished since 1970 was performed to determine the
sensitivity of flexible bronchoscopy for the diagnosis
of bronchogenic carcinoma. Studies with 50 pa-
tients and those that reported exclusively on inter-
operator performance variabilities or focused on
technical aspects (eg, needle size or cytology prepa-
ration) were excluded. Forty-four studies met our
inclusion criteria. Only six of these included both
disease-positive and disease-negative patients, and
provided data on test accuracy, positive predictive
value, and negative predictive value. Most of the
studies identified were limited to patients with
pathologically confirmed bronchogenic carcinoma
and provided data only on the diagnostic yield (test
sensitivity). The data were analyzed further with
respect to the diagnosis of central disease with an
endobronchial component and peripheral disease
beyond the segmental level.
Bronchoscopy Sensitivity for Central Disease (Endo-
bronchial Disease): Thirty studies of patients with
central disease were identified (Table 4).
4,2024,26,3860
Endobronchial biopsies provided the highest sensitivity
(0.74; 20 studies), followed by brushings (0.59; 18
studies) and washings (0.48; 12 studies). The interpre-
tation of the sensitivity for bronchoscopic needle aspi-
rates (range, 0.23 to 0.90; average, 0.56; 8 studies) was
limited to fewer studies with large differences in sam-
ple size and inconsistencies in technique (ie, endobron-
chial vs transbronchial biopsy). The overall sensitivity
for all bronchoscopic modalities combined, where re-
ported, was 0.88 for centrally located, endobronchial
disease (14 studies).
Bronchoscopy Sensitivity for Peripheral Disease
(Beyond the Segmental Bronchus): Thirty studies
reported on the sensitivity of flexible bronchoscopy
beyond the visual segmental bronchi (Table
5).
4,22,25,26,38,45,4750,52,53,5673
Brushings provided the
highest sensitivity (0.52; 15 studies), followed by
transbronchial biopsies (0.46; 18 studies) and BAL/
washings (0.43; 13 studies). Although TBNA showed
a high sensitivity (0.67; five studies), the data deserve
cautious interpretation because of the limited num-
ber of studies and the large differences in sample
size. The overall sensitivity for all modalities in the
diagnosis of peripheral disease was 0.69 (12 studies).
Bronchoscopic Sensitivity for the Peripheral Le-
sions 2 cm vs 2 cm in Diameter: Eight studies
were identified that reported on the sensitivity of
bronchoscopy (brush and/or biopsy) for peripheral
lesions with a size 2 cm or 2 cm in diameter
(Table 6).
25,38,5658,7274
The sensitivity for peripheral
Table 3Sensitivity of Sputum Cytology for Diagnosis of Bronchogenic Carcinoma by Location of Lesion
(Central vs Peripheral)*
Study/Year
Central Peripheral
Positive Total Sensitivity Positive Total Sensitivity
Bo cking et al
2
/1992 113 132 0.86 61 72 0.85
Matsuda et al
27
/1990 62 74 0.84 20 31 0.65
Liang
28
/1989 38 49 0.78 80 112 0.71
Kawachi and Shimokata
29
/1985 45 53 0.85 16 28 0.57
Tanaka et al
30
/1985 13 33 0.39 14 54 0.26
Fontana et al
31
/1984 25 41 0.61 7 50 0.14
Ng and Horak
32
/1983 136 160 0.85 17 41 0.41
Pilotti et al
22
/1982 55 91 0.60 58 108 0.54
Clee and Sinclair
33
/1981 127 229 0.55 26 80 0.33
Jay et al
5
/1980 15 31 0.48 14 27 0.52
Droese et al
34
/1978 23 47 0.49 42 100 0.42
Gagneten et al
10
/1976 32 46 0.70 27 59 0.46
Rosa et al
11
/1973 155 188 0.82 53 110 0.48
Dahlgren and Lind
12
/1972 21 37 0.57 5 29 0.17
Umiker
35
/1969 41 64 0.64 22 36 0.61
Kern et al
36
/1968 37 51 0.73 20 46 0.43
Spjut et al
18
/1955 139 199 0.70 13 27 0.48
Total 1,077 1,525 0.71 495 1,010 0.49
*Adapted from Table X in Sing et al, 1997.
4
lesions 2 cm in diameter was 0.33. Peripheral
tumors with a diameter 2 cm resulted in a sensi-
tivity of 0.62. Six studies
23,24,59,65,67,71
reported on the
sensitivity of postbronchoscopy sputa as an adjunct
to the previously mentioned bronchoscopic tech-
niques, which was 0.35 (Table 7).
Characteristics (Sensitivity and Specificity) for
TTNA as Diagnostic Modality, With Particular
Emphasis on the Size and the Location of the
Suspected Cancer?
A meta-analysis by Lacasse et al
75
focused on the
performance characteristics of TBNA or biopsy for
the diagnosis of localized pulmonary lesions. The
study encompassed a comprehensive search (up to
1995) of English-language reports on the use of
needle aspiration or biopsy for the evaluation of
solitary or multiple pulmonary lesions.
At least 90% of the study populations had to have
parenchymal pulmonary lesions as opposed to medi-
astinal, hilar, or pleural lesions. All diagnoses were
verified by surgical biopsy, biopsy of an adjacent site
with tumor involvement, culture results, or clinical
follow-up for at least 1 year. Cytology alone, even
when confirmed by another site, was not accepted as
a reference standard. At least 90% of patients in each
study had a histologic reference standard diagnosis.
Diagnostic accuracy data were abstracted in a 2 6
table and were analyzed according to predefined
thresholds.
We updated the literature search and meta-
analysis using the same inclusion criteria, except that
we required each study to have a minimum of 50
subjects. In our reanalysis, we considered only the
following cut point: definite malignancy or suspicion
for malignancy as test-positive, and all other test
results, including nondiagnostic, benign, nonspecific,
and specific benign diagnoses, as test-negative (this
corresponds to cut point b in the published meta-
analysis).
We identified 19 studies that had not been included
Table 4Sensitivity of Flexible Bronchoscopic Diagnostic Procedures for Central Bronchogenic Carcinoma*
Study/Year Patients, No.
Sensitivity
All Methods Endobronchial Biopsy Brush Washing EBNA/TBNA
Baaklini et al
38
/2000 22 0.82
Bungay et al
39
/2000 24 0.92
Dasgupta et al
40
/1999 32 0.97 0.78
Govert et al
41
/1999 57 0.95 0.74 0.63 0.82
McLean et al
42
/1998 71 0.82
Bilaceroglu et al
43
/1997 68 0.96 0.66 0.90
Sing et al
4
/1997 53 0.64
Govert et al
44
/1996 177 0.85 0.81 0.48 0.43
Castella et al
45
/1995 39 0.87
Utz et al
46
/1993 88 0.36
Buccheri et al
47
/1991 708 0.80 0.35 0.31
Popp et al
48
/1991 99 0.93 0.79
Mak et al
49
/1990 125 0.87 0.76 0.52 0.50
Gay and Brutine
50
/1989 53 0.23
Saita et al
51
/1990 105 0.48 0.30
Wagner et al
20
/1989 72 0.67 0.58 0.39 0.35 0.36
Schenk et al
21
/1987 91 0.71 0.56 0.40 0.29 0.45
Cox et al
52
/1984 33 0.94 0.84 0.83 0.76
Lam et al
53
/1983 329 0.94 0.82 0.74 0.76
Zisholtz and Eisenberg
54
/1983 51 0.73 0.67 0.65 0.44
Gellert et al
55
/1982 218 0.78
Pilotti et al
22
/1982 286 0.78
McDougall and Cortese
56
/1981 16 0.50 0.23
Radke et al
57
/1979 15 0.87
Chaudhary et al
23
/1978 95 0.76 0.53 0.78
Chopra et al
24
/1977 51 0.66 0.72 0.51
Stringfield et al
58
/1977 78 0.85
Kvale et al
59
/1976 71 0.71 0.77 0.63
Zavala
60
/1975 193 0.94 0.97 0.93
Oswald et al
26
/1971 434 0.61
Summary 3,754 0.88 0.74 0.59 0.48 0.56
*EBNA endobronchial needle aspiration.
No. of patients represents the maximum number included in sensitivity calculations for any one method.
in the previously published meta-analysis. Seven stud-
ies
7682
were published in 1996 or before and had been
considered and excluded by Lacasse et al
75
Twelve
reports had not been considered by Lacasse et al,
75
11
published since 1996,
8393
and one published in 1995.
94
We excluded five studies
9599
with 50 subjects that
were included in the published meta-analysis.
Performance Characteristics of TTNA: We ana-
lyzed the additional 19 reports, compared our these
results with those of the published meta-analysis,
and reanalyzed all the studies
74,7694,100139
with
particular attention to lesion size, location, and im-
aging technology (Table 8). The pooled sensitivity
was 0.90 (95% confidence interval [CI], 0.88 to
Table 6Sensitivity of Flexible Bronchoscopic Diagnosis of Bronchogenic Carcinoma by Size of Lesion*
Study/Year
2 cm Lesion 2 cm Lesion
Patients, No. Pos Neg Sens Patients, No. Pos Neg Sens
Baaklini et al
38
/2000 16 4 12 0.25 135 93 42 0.69
Gasparini et al
74
/1995 195 82 113 0.42 300 169 131 0.56
Naidich et al
72
/1988 15 4 11 0.27 46 26 20 0.57
Wallace and Deutsch
73
/1982 65 3 62 0.05 78 24 54 0.31
56
/1981 9 1 8 0.11 36 21 15 0.58
Radke et al
57
/1979 21 6 15 0.29 76 49 27 0.64
Stringfield et al
58
/1977 3 1 2 0.33 26 13 13 0.50
Hattori et al
25
/1971 17 13 4 0.76 182 150 32 0.82
Summary 341 114 227 0.33 879 545 334 0.62
*Neg negatives; Pos positives; Sens sensitivity.
Table 5Sensitivity of Flexible Bronchoscopic Diagnostic Procedures for Peripheral Bronchogenic Carcinoma
Study/Year Patients,* No.
Sensitivity
All Methods Transbronchial Biopsy Brushing BAL TBNA
Baaklini et al
38
/2000 129 0.61
Gasparini et al
61
/1999 480 0.76 0.50 0.70
Reichenberger et al
62
/1999 103 0.39 0.36 0.28 0.47
Aristizabal et al
63
/1998 64 0.34
Bilaceroglu et al
64
/1998 92 0.64
Wongsurakiat et al
65
/1998 30 0.50 0.17 0.47
Sing et al
4
/1997 22 0.22
Castella et al
45
/1995 45 0.69
Debeljak et al
66
/1994 39 0.77 0.59 0.36
de Gracia et al
67
/1993 55 0.33
Torrington and Kern
68
/1993 91 0.20
Pirozynski
69
/1992 145 0.33 0.30 0.65 0.58
Buccheri et al
47
/1991 337 0.75 0.44 0.33
Popp et al
48
/1991 87 0.80 0.83
Mak et al
49
/1990 63 0.56 0.37 0.29 0.38
Rennard et al
70
/1990 730 0.47
Gay and Brutinel
50
/1989 20 0.65
Mori et al
71
/1989 85 0.84 0.84 0.42
Naidich et al
72
/1988 65 0.48
Cox et al
52
/1984 22 0.36 0.29 0.22 0.36
Lam et al
53
/1983 155 0.86 0.61 0.52 0.52
Pilotti et al
22
/1982 84 0.29
Wallace and Deutsch
73
/1982 143 0.19
56
/1981 130 0.62 0.48 0.36 0.36
Radke et al
57
/1979 82 0.51
Stringfield et al
58
/1977 29 0.48
Kvale et al
59
/1976 29 0.27 0.21 0.12
Zavala
60
/1975 137 0.71 0.69 0.70
Hattori et al
25
/1971 208 0.83
Oswald et al
26
/1971 435 0.28
Summary 4,136 0.69 0.46 0.52 0.43 0.67
*No. of patients represents the maximum number included in sensitivity calculations for any one method.
0.92). Individual study estimates ranged from 0.62 to
0.99 (Fig 1). The pooled estimate was not impor-
tantly different, regardless of whether it was based
on the 19 additional studies that had not been
reviewed by Lacasse et al,
75
which included 50
subjects, or the combination of these two categories.
There was little difference in specificity for any
group of studies analyzed.
Performance Characteristics for TTNA by Lesion
Size: A total of seven studies described a lower bound-
ary or gave results limited to large lesions; however, the
boundary varied. Studies described the accuracy in
lesions 1.5 cm in diameter,
76,139
2 cm in diame-
ter,
80,84,100,136
and 3 cm in diameter.
87
These studies
range in date from 1967 to 2000 and used varied
imaging technologies for the localization of lung nod-
ules. A total of five studies reported results for small
lesions, which were defined as 1.5 cm in diame-
ter,
76,91
2 cm in diameter,
84,100
and 3 cm in
diameter.
92
Because of the great differences in date
and technique, we believe that the most valid data
come from two studies
76,84
in which biopsy speci-
mens from small and large lesions were obtained and
that described results separately for each type of
lesion. In another study of this type,
100
the rate of
inadequate specimens was 10% and was signifi-
cantly greater than the rate for large lesions. Fur-
thermore, these patients were not subject to the
reference standard; therefore, this study was not
used for this analysis.
100
In one of the included studies,
84
the sensitivity
was 0.95 (95% CI, 0.89 to 0.98) for lesions 2 cm
in diameter, and 0.91 (95% CI, 0.79 to 0.97) for
lesions 2 cm in diameter. In the other study,
76
sensitivity was 0.94 (95% CI, 0.84 to 0.98) for
lesions 1.5 cm, and 0.78 (95% CI, 0.56 to 0.92)
for lesions 1.5 cm. In neither study was the
difference between sensitivity in small and large
lesions statistically significant. However, both
studies, each of which used CT guidance, sug-
gested a trend toward lower sensitivity.
Performance Characteristics for TTNA by Loca-
tion of the Lesion: Overall, only a few studies de-
scribed the test performance data (sensitivity and
specificity) according to location of lesion, so that we
had limited data with which to address the question
of differences in test performance based on lesion
location. It is possible that some studies describing
only the yield (sensitivity) would provide data by
location of lesion.
Performance Characteristics for TTNA by Imag-
ing Technique (CT Scan vs Fluoroscopy): We com-
pared the pooled sensitivity and specificity for stud-
ies using CT guidance vs those using fluoroscopy
guidance in order to investigate the potential reasons
for the between-study differences in accuracy. Most
of the new studies used CT scanning to guide TTNA,
and, although initially hypothesized, Lacasse et al
75
did not find any differences in test operating char-
acteristics between CT and fluoroscopic guidance of
TTNA in their original meta-analysis. However, with
substantially more data from CT-guided TTNA stud-
ies in our analysis, we found that studies using CT
guidance showed higher sensitivity than those using
fluoroscopy guidance. Using a random-effects
model, the pooled sensitivities were 0.92 (95% CI,
0.90 to 0.94) and 0.88 (95% CI, 0.85 to 0.90) for
studies of CT-guided and fluoroscopy-guided TTNA,
respectively. Significant differences in sensitivity
based on imaging technology may relate to the issue
of lesion size. As CT scanning has been used to guide
TTNA, the size of the nodules for which biopsy has
been attempted has decreased, with the lower limit
of the size of the lesion for some recently published
studies as low as 0.5 cm. Pooled specificity was
identical for the two technologies (0.97; 95% CI,
0.96 to 0.98).
Performance Characteristics by Needle Type
(Aspiration, Aspiration Biopsy, or Cutting Needle):
Twenty studies used cutting-type needles that were
designed to remove a tissue sample for histologic
analysis, and in 12 studies this was the only sample
on which the diagnosis was based (Table 8). The
remaining studies used either fine-needle aspiration
cytology exclusively (n 19) or fine-needle aspira-
tion with cytologic analysis of the aspirate and histo-
logic analysis of the tissue aspirated (n 21). In one
study, the needle type was not described. Controlling
for needle type did not decrease heterogeneity
among studies in the meta-analysis by Lacasse et al,
75
nor did sensitivity or specificity estimates vary by
needle type in this analysis.
Two studies
77,94
reported direct comparisons be-
tween aspiration cytology and cutting needle biopsy
histologic diagnoses. Both studies found that trans-
Table 7Sensitivity of Postbronchoscopy Sputum for
Diagnosis of Bronchogenic Carcinoma
Study/Year
Patients,
No.
Postbronchoscopy Sputum
Positive Negative Sensitivity
Wongsurakiat et al
65
/1998 26 2 24 0.08
de Gracia et al
67
/1993 43 13 30 0.30
Mori et al
71
/1989 81 17 64 0.21
Chaudhary et al
23
/1978 114 58 56 0.51
Chopra et al
24
/1977 51 24 27 0.48
Kvale et al
59
/1976 22 3 19 0.14
Summary 337 117 220 0.35
Table 8Sensitivity and Specificity of Transthoracic Needle Aspiration and/or Biopsy for Diagnosis of Peripheral
Bronchogenic Carcinoma*
Study/Year
Patients,
No.
Type of
Needle
Radiologic
Assistance Sensitivity Specificity
False-Positive
Rate
False-Negative
Rate Prevalence
Lopez Hanninen et al
83
/2001 79 C CT 0.96 1.00 0 0.06 0.63
Laurent et al
84
/2000 202 C CT 0.94 1.00 0 0.18 0.80
Hirose et al
85
/2000 50 C CT 0.83 1.00 0 0.19 0.58
Charig and Phillips
86
/2000 185 C CT 0.93 1.00 0 0.48 0.93
Swischuk et al
87
/1998 612 C Fluo, CT 0.96 0.99 0 0.13 0.76
Lucidarme et al
88
/1998 89 C CT 0.93 1.00 0 0.26 0.84
Larscheid et al
89
/1998 130 A, C CT 0.91 1.00 0 0.26 0.80
Yankelevitz et al
90
/1997 114 A CT 0.94 1.00 0 0.16 0.76
Westcott et al
91
/1997 62 A, C Fluo, CT 0.93 1.00 0 0.12 0.67
Santambrogio et al
92
/1997 220 A CT 0.93 0.99 0.01 0.11 0.64
Cattelani et al
93
/1997 119 A CT 0.93 1.00 0 0.13 0.67
Li et al
76
/1996 97 A CT 0.89 1.00 0 0.43 0.88
Klein et al
77
/1996 129 A, C CT 0.95 1.00 0 0.08 0.64
Milman et al
78
/1995 103 A Fluo 0.69 1.00 0 0.49 0.76
Bo cking et al
94
/1995 371 A, C CT 0.99 0.94 0.02 0.04 0.79
Zakowski et al
79
/1992 176 A Fluo, CT 0.84 1.00 0 0.47 0.84
Yang et al
80
/1992 120 A US 0.62 1.00 0.00 0.63 0.82
Cristallini et al
81
/1992 390 A, B Fluo, CT 0.94 0.99 0.00 0.16 0.77
Calhoun et al
82
/1986 197 A Fluo 0.87 1.00 0.00 0.35 0.81
Knudsen et al
101
/1996 128 A US 0.95 0.95 0.02 0.09 0.68
Gasparini et al
74
/1995 589 A, C Fluo, CT 0.93 0.99 0.00 0.15 0.72
Garcia Rio et al
102
/1994 84 A CT 0.84 1.00 0.00 0.39 0.80
Burbank et al
103
/1994 60 C CT 0.95 1.00 0.00 0.11 0.72
Targhetta et al
104
/1993 64 B US 0.91 1.00 0.00 0.31 0.83
Grode et al
105
/1993 219 A, B, C Fluo 0.89 1.00 0.00 0.31 0.80
Collins et al
106
/1992 129 B, C Fluo, CT 0.94 1.00 0.00 0.39 0.91
Veale et al
107
/1988 95 A Fluo 0.71 1.00 0.00 0.49 0.77
Veale et al
107
/1988 100 A Fluo 0.84 1.00 0.00 0.52 0.87
Simpson et al
108
/1988 227 B Fluo 0.82 1.00 0.00 0.73 0.93
Lovett et al
109
/1988 92 A Fluo 0.90 1.00 0.00 0.38 0.86
Levine et al
110
/1988 58 NR Fluo 0.71 1.00 0.00 0.30 0.60
Balslov et al
111
/1988 284 C Fluo 0.78 1.00 0.00 0.37 0.73
Weisbrod et al
112
/1987 133 C Fluo 0.78 1.00 0.00 0.36 0.71
Stanley et al
113
/1987 440 A Fluo, CT 0.97 0.97 0.01 0.09 0.73
Winning et al
114
/1986 165 A Fluo 0.77 1.00 0.00 0.43 0.76
Nahman et al
115
/1985 120 B Fluo 0.98 0.94 0.01 0.11 0.86
Lees et al
116
/1985 86 A, B Fluo, CT, US 0.85 1.00 0.00 0.42 0.83
Greene et al
117
/1985 150 B Fluo 0.97 1.00 0.00 0.13 0.81
Crosby et al
118
/1985 180 A Fluo, CT, US 0.82 1.00 0.00 0.69 0.93
Stevens and Jackman
119
/1984 348 A, B, C Fluo 0.92 0.99 0.00 0.13 0.64
Harrison et al
120
/1984 89 C Fluo 0.96 1.00 0.00 0.14 0.78
McEvoy et al
121
/1983 81 C Fluo 0.87 1.00 0.00 0.45 0.86
Johnson et al
122
/1983 200 A, B Fluo, CT 0.95 0.98 0.01 0.09 0.68
Vine et al
123
/1982 91 C Fluo 0.87 1.00 0.00 0.22 0.69
Samuelsson et al
100
/1982 367 A Fluo 0.97 0.96 0.02 0.06 0.67
Pilotti et al
124
/1982 130 A Fluo 0.92 0.93 0.01 0.39 0.88
Jamieson et al
125
/1981 82 A, B Fluo 0.94 1.00 0.00 0.19 0.80
Allison and Hemingway
126
/1981 147 B Fluo 0.89 1.00 0.00 0.15 0.62
Westcott
127
/1980 400 B Fluo 0.98 0.94 0.02 0.05 0.73
Taft et al
128
/1980 100 B Fluo 0.83 0.95 0.01 0.42 0.80
Poe and Tobin
129
/1980 95 B Fluo 0.90 0.94 0.01 0.32 0.81
Pak et al
130
/1981 52 A, B Fluo 0.98 0.00 0.18 1.00 0.83
Flower and Verney
131
/1979 282 B Fluo 0.87 0.96 0.02 0.25 0.72
Sagel et al
132
/1978 1153 B Fluo 0.96 0.99 0.00 0.13 0.78
Lalli et al
133
/1978 1204 B Fluo 0.85 0.99 0.00 0.36 0.78
House and Thomson
134
/1977 88 B Fluo 0.96 0.97 0.02 0.06 0.65
Francis
135
/1977 244 B Fluo 0.82 0.95 0.03 0.29 0.68
Pavy et al
136
/1974 59 B Fluo 0.86 1.00 0.00 0.54 0.89
Stevens et al
137
/1968 100 B Fluo 0.90 0.95 0.03 0.14 0.62
Nasiell
138
/1967 144 B Fluo 0.72 1.00 0.00 0.29 0.60
King and Russell
139
/1967 59 A Fluo 0.88 1.00 0.00 0.35 0.81
0.90 0.97
Summary (0.880.92) (0.960.98)
*A aspiration needle; B aspiration biopsy needle; C cutting biopsy needle; Fluo fluoroscopy; NR not reported; US ultrasound.
False-positive rate is 1 minus the positive predictive value of the test. It is highly dependent on the prevalence of disease.
False-negative rate is defined here as 1 minus the negative predictive value of the test. It is highly dependent on the prevalence of disease.
thoracic needle core biopsy compared with fine-
needle aspirate showed similar sensitivity for malig-
nancy (86% vs 92%, respectively
77
; and 98% vs
98.4%, respectively
94
) and better ability to determine
a specific diagnosis for nonmalignant lesions (100% vs
44%, respectively
77
; and 100% vs 50%, respectively
94
).
Key Question 4: What Is the Diagnostic Error Rate
When Differentiating Between Non-small Cell
Lung Cancer and Small Cell Lung Cancer
Generated by Various Diagnostic Techniques
(Bronchoscopy and Sputum Cytology)?
We systematically reviewed the studies that we
selected for reviews of the diagnostic accuracy of
TTNA and bronchoscopy to find data on the differ-
ences in diagnosis between small cell and non-small
cell lung cancer based on the cytologic vs histologic
diagnoses. We also included studies describing series
of patients with lung cancer, and correlating cyto-
logic and histologic diagnoses that had been identi-
fied through previous reviews.
1,2,4,140
We considered
sputum cytology, TTNA cytology, bronchoscopic
washings, BAL, and brushings.
We attempted to construct a 2 2 table correlat-
ing cytologic diagnosis of small cell lung cancer or
non-small cell lung cancer with a histologic diagnosis
based on the same categories. All histologic types of
non-small cell lung cancer were considered to be the
same. We limited consideration only to patients with
lung cancer who had both a cytologic and histologic
diagnosis.
Table 9 summarizes 21 studies,
7,20,22,26,89,91,101,105,
109,113,122,128,131,136,141147
some addressing several di-
agnostic modalities (TTNA, 14 studies; expectorated
sputum, 5 studies; bronchoscopy brush sample, 2
studies; and bronchoscopy TBNA, 4 studies). These
studies showed that the overall accuracy of diagnos-
ing small cell lung cancer vs non-small cell lung
cancer is 0.98, with individual studies ranging from
0.94 to 1.0. The proportion of small cell lung cancer
ranged from 0.03 to 0.25 among these studies. When
small cell lung cancer was diagnosed based on
cytology, the true diagnosis was non-small cell lung
cancer in about 9% of such cases (the false-positive
test rate for cytologic diagnosis of small cell lung
cancer, on average, was 0.09, with individual studies
ranging from 0 to 0.33). When non-small cell lung
cancer was diagnosed based on cytology, the true
diagnosis was small cell lung cancer in about 2% of
cases. (The false-negative test rate for diagnosing
non-small cell lung cancer was 0.02, with individual
studies ranging from 0 to 0.07.)
Discussion
The available techniques for diagnosing lung can-
cer are well-supported in the medical literature. The
majority of the data on flexible bronchoscopy for
the diagnosis of suspected lung cancer defines only
the diagnostic yield (sensitivity). The sensitivity for
endobronchial disease is high, especially for biopsies
and brushings. The sensitivity is lower for peripheral
lesions, with cytobrushing showing the highest sen-
sitivity, followed by transbronchial biopsies and
BAL/washing. Flexible bronchoscopy has a poor
sensitivity for peripheral lesions 2 cm in diameter.
The sensitivity of TTNA is excellent, and our
expanded analyses did not reveal significant differ-
ences from a recently published meta-analysis.
There is a trend toward lower sensitivity for smaller
lesions (ie, 2 cm in diameter). Studies using CT
guidance had a significantly higher sensitivity than
those using fluoroscopy guidance.
Studies on the accuracy of sputum cytology for the
diagnosis of lung cancer are difficult to summarize
because of a variety of methodological problems, and
suggest that the number of sputum samples and the
specimen adequacy are strongly related to the sen-
sitivity of the technique.
The distinction between small cell lung cancer and
non-small cell lung cancer on the basis of cytology
Figure 1. Summary receiver operating characteristic curve for
studies of TTNA or biopsy for the diagnosis of peripheral
bronchogenic carcinoma. Gray ovals indicate sensitivity and
specificity estimates from individual studies. The black curve
indicates the summary receiver operating characteristic curve. A
black X indicates an estimate of sensitivity and specificity from
separate meta-analyses of sensitivity and specificity using a
random-effects model (upper) or a fixed-effects model (lower).
Reprinted with permission by Sing et al.
4
appears to be highly accurate, although a cytologic
diagnosis of non-small cell lung cancer is more
reliable than a cytologic diagnosis of small cell lung
cancer, with average misclassification rates of 2%
and 9%, respectively.
Important limitations of the literature include the
fact that few studies describe the outcomes of eval-
uation in all patients who are suspected of having
lung cancer. This problem is particularly apparent in
studies of flexible bronchoscopy. Although false-
positive diagnoses are rare, nondiagnostic results are
not uncommon and lead to the use of other tech-
niques or procedures for diagnosis.
References
1 Benbassat J, Regev A, Slater PE. Predictive value of sputum
cytology (erratum appears in Thorax 1987; 42:640). Thorax
1987; 42:165172
2 Bo cking A, Biesterfeld S, Chatelain R, et al. Diagnosis of
bronchial carcinoma on sections of paraffin-embedded spu-
tum: sensitivity and specificity of an alternative to routine
cytology. Acta Cytol 1992; 36:3747
3 Mehta AC, Marty JJ, Lee FY. Sputum cytology. Clin Chest
Med 1993; 14:6985
4 Sing A, Freudenberg N, Kortsik C, et al. Comparison of the
sensitivity of sputum and brush cytology in the diagnosis of
lung carcinomas. Acta Cytol 1997; 41:399408
5 Jay SJ, Wehr K, Nicholson DP, et al. Diagnostic sensitivity
and specificity of pulmonary cytology: comparison of tech-
niques used in conjunction with flexible fiber optic bron-
choscopy. Acta Cytol 1980; 24:304312
6 Kern WH. The diagnostic accuracy of sputum and urine
cytology. Acta Cytol 1988; 32:651654
7 Risse EK, vant Hof MA, Laurini RN, et al. Sputum cytology
by the Saccomanno method in diagnosing lung malignancy.
Diagn Cytopathol 1985; 1:286291
8 Johnston WW, Bossen EH. Ten years of respiratory cytopa-
thology at Duke University Medical Center: I. The cyto-
pathologic diagnosis of lung cancer during the years 1970 to
1974, noting the significance of specimen number and type.
Acta Cytol 1981; 25:103107
9 Yoneyama T, Canlas MS. From exfoliative to diagnostic
cytology: a statistical evaluation of pulmonary cytology. Acta
Cytol 1978; 22:158161
10 Gagneten CB, Geller CE, Del Carmen Saenz M. Diagnosis
of bronchogenic carcinoma through the cytologic examina-
tion of sputum, with special reference to tumor typing. Acta
Cytol 1976; 20:530536
11 Rosa UW, Prolla JC, Gastal ED. Cytology in diagnosis of
cancer affecting the lung: results in 1,000 consecutive
patients. Chest 1973; 63:203207
12 Dahlgren SE, Lind B. Comparison between diagnostic
results obtained by transthoracic needle biopsy and by
sputum cytology. Acta Cytol 1972; 16:5358
13 Koss LG, Melamed MR, Goodner JT. Pulmonary cytology: a
brief survey of diagnostic results from July 1st 1952 until
December 31st 1960. Acta Cytol 1964; 8:104113
14 Hinson KFW, Kuper SWA. The diagnosis of lung cancer by
examination of sputum. Thorax 1963; 18:350353
15 Russell WO, Neidhardt HW, Mountain CE, et al. Cyto-
diagnosis of lung cancer. Acta Cytol 1963; 7:144
16 Allen WB, Whittlesey P. The results of the experimental use
Table 9Accuracy of Cytology for Distinguishing Between Small Cell Lung Cancer and Non-small Cell Lung
Cancer (Histology Gold Standard)
Study/Year Patients, No. Technique Accuracy False-Positive Rate False-Negative Rate Prevalence
Pilotti et al
22
/1982 252 Brush 0.96 0.00 0.04 0.15
Matsuda et al
141
/1986 443 Brush 0.94 0.11 0.04 0.24
Oswald et al
26
/1971 476 Sputum 0.97 0.21 0.01 0.08
Payne et al
142
/1981 656 Sputum 0.99 0.08 0.01 0.07
Clee et al
143
/1982 140 Sputum 0.98 0.00 0.02 0.15
Pilotti et al
144
/1982 400 Sputum 0.97 0.12 0.02 0.12
Risse et al
7
/1985 143 Sputum 0.97 0.03 0.03 0.24
Payne et al
142
/1981 126 TBNA 0.98 0.00 0.03 0.08
Wagner et al
20
/1989 18 TBNA 0.94 0.00 0.10 0.50
Clee et al
143
/1982 33 TBNA/brush 1.00 0.00 0.00 0.12
Clee et al
143
/1982 50 TBNA/brush 0.98 0.00 0.02 0.18
Pavy et al
136
/1974 17 TTNA 0.94 0.00 0.07 0.24
Flower and Verney
131
/1979 77 TTNA 0.97 0.50 0.00 0.03
Taft et al
128
/1980 33 TTNA 1.00 0.00 0.00 0.06
Payne et al
142
/1981 65 TTNA 0.98 0.33 0.00 0.03
Johnson et al
122
/1983 200 TTNA 0.98 0.00 0.03 0.15
Johnston
145
/1984 1,015 TTNA 0.98 0.12 0.01 0.09
Zaman et al
146
/1986 1,209 TTNA 0.98 0.09 0.01 0.10
Young et al
147
/1987 72 TTNA 0.99 0.00 0.01 0.03
Stanley et al
113
/1987 323 TTNA 0.99 0.04 0.00 0.10
Lovett et al
109
/1988 61 TTNA 1.00 0.00 0.00 0.07
Grode et al
105
/1993 224 TTNA 1.00 0.00 0.00 0.10
Knudsen et al
101
/1996 80 TTNA 0.99 0.25 0.00 0.04
Westcott et al
91
/1997 62 TTNA 1.00 0.00 0.00 0.06
Larscheid et al
89
/1998 130 TTNA 1.00 0.00 0.00 0.25
Total 6,305 0.98 0.09 0.02 0.12
of sulfur dioxide in the production of material for cell studies
in lung cancer. Ann Intern Med 1960; 52:326329
17 Koss LG. Cytological diagnosis of lung cancer. Acta Pathol
Jpn 1958; 8:99111
18 Spjut HJ, Fier DJ, Ackermann LV. Exfoliative cytology and
pulmonary cancer. J Thorac Surg 1955; 30:90107
19 Liebow AA, Lindskog GF, Bloomer WE. Cytological studies
of sputum and bronchial secretions in the diagnosis of
cancer of the lung. Cancer 1948; 1:223233
20 Wagner ED, Ramzy I, Greenberg SD, et al. Transbronchial
fine-needle aspiration: reliability and limitations. Am J Clin
Pathol 1989; 92:3641
21 Schenk DA, Bryan CL, Bower JH, et al. Transbronchial
needle aspiration in the diagnosis of bronchogenic carci-
noma. Chest 1987; 92:8385
22 Pilotti S, Rilke F, Gribaudi G, et al. Cytologic diagnosis of
pulmonary carcinoma on bronchoscopic brushing material.
Acta Cytol 1982; 26:655660
23 Chaudhary BA, Yoneda K, Burki NK. Fiberoptic bronchos-
copy: comparison of procedures used in the diagnosis of lung
cancer. Thorac Cardiovasc Surg 1978; 76:3337
24 Chopra SK, Genovesi MG, Simmons DH, et al. Fiberoptic
bronchoscopy in the diagnosis of lung cancer comparison of
pre and post-bronchoscopy sputa, washings, brushings and
biopsies. Acta Cytol 1977; 21:524527
25 Hattori S, Matsuda M, Nishihara H, et al. Early diagnosis of
small peripheral lung cancer: cytologic diagnosis of very
fresh cancer cells obtained by the TV-brushing technique.
Acta Cytol 1971; 15:460467
26 Oswald NC, Hinson KF, Canti G, et al. The diagnosis of
primary lung cancer with special reference to sputum
cytology. Thorax 1971; 26:623627
27 Matsuda M, Horai T, Doi O, et al. Diagnosis of squamous-
cell carcinoma of the lung by sputum cytology: with special
reference to correlation of diagnostic accuracy with size and
proximal extent of resected tumor. Diagn Cytopathol 1990;
6:248251
28 Liang XM. Accuracy of cytologic diagnosis and cytotyping of
sputum in primary lung cancer: analysis of 161 cases. J Surg
Oncol 1989; 40:107111
29 Kawachi H, Shimokata K. Factors affecting the rate of
positivity of sputum cytology in lung cancer. Jpn J Clin
Oncol 1985; 15:451456
30 Tanaka T, Yamamoto M, Tamura T, et al. Cytologic and
histologic correlation in primary lung cancer: a study of 154
cases with resectable tumors. Acta Cytol 1985; 29:4956
31 Fontana RS, Sanderson DR, Taylor WF, et al. Early lung
cancer detection: results of the initial (prevalence) radiologic
and cytologic screening in the Mayo Clinic study. Am Rev
Respir Dis 1984; 130:561565
32 Ng AB, Horak GC. Factors significant in the diagnostic
accuracy of lung cytology in bronchial washing and sputum
samples: II. Sputum samples. Acta Cytol 1983; 27:397402
33 Clee MD, Sinclair DJ. Assessment of factors influencing the
result of sputum cytology in bronchial carcinoma. Thorax
1981; 36:143146
34 Droese M, Bayer E, Prauer H. Cytological studies of sputum
and bronchial secretions (in German). Dtsch Med Wochen-
schr 1978; 103:244248
35 Umiker WO. Diagnosis of bronchogenic carcinoma: an
evaluation of pulmonary cytology, bronchoscopy and scalene
lymph node biopsy. Dis Chest 1969; 37:8290
36 Kern WH, Markovits BA, Lindesmith GG. The prognostic
significance of sputum cytology. Acta Cytol 1968; 12:430
432
37 Risse EK, vant Hof MA, Vooijs GP. Relationship between
patient characteristics and the sputum cytologic diagnosis of
lung cancer. Acta Cytol 1987; 31:159165
38 Baaklini WA, Reinoso MA, Gorin AB, et al. Diagnostic yield
of fiberoptic bronchoscopy in evaluating solitary pulmonary
nodules. Chest 2000; 117:10491054
39 Bungay HK, Pal CR, Davies CW, et al. An evaluation of
computed tomography as an aid to diagnosis in patients
undergoing bronchoscopy for suspected bronchial carci-
noma. Clin Radiol 2000; 55:554560
40 Dasgupta A, Jain P, Minai OA, et al. Utility of transbronchial
needle aspiration in the diagnosis of endobronchial lesions.
Chest 1999; 115:12371241
41 Govert JA, Dodd LG, Kussin PS, et al. A prospective
comparison of fiberoptic transbronchial needle aspiration
and bronchial biopsy for bronchoscopically visible lung
carcinoma. Cancer 1999; 87:129134
42 McLean AN, Semple PA, Franklin DH, et al. The Scottish
multi-centre prospective study of bronchoscopy for bron-
chial carcinoma and suggested audit standards. Respir Med
1998; 92:11101115
43 Bilaceroglu S, Gunel O, Cagirici U, et al. Comparison of
endobronchial needle aspiration with forceps and brush
biopsies in the diagnosis of endobronchial lung cancer.
Monaldi Arch Chest Dis 1997; 52:1317
44 Govert JA, Kopita JM, Matchar D, et al. Cost-effectiveness
of collecting routine cytologic specimens during fiberoptic
bronchoscopy for endoscopically visible lung tumor. Chest
1996; 109:451456
45 Castella J, Buj J, Puzo C, et al. Diagnosis and staging of
bronchogenic carcinoma by transtracheal and transbronchial
needle aspiration. Ann Oncol 1995; 6:S21S24
46 Utz JP, Patel AM, Edell ES. The role of transcarinal needle
aspiration in the staging of bronchogenic carcinoma. Chest
1993; 104:10121016
47 Buccheri G, Barberis P, Delfino MS. Diagnostic, morpho-
logic, and histopathologic correlates in bronchogenic carci-
noma: a review of 1,045 bronchoscopic examinations. Chest
1991; 99:809814
48 Popp W, Rauscher H, Ritschka L, et al. Diagnostic sensitiv-
ity of different techniques in the diagnosis of lung tumors
with the flexible fiberoptic bronchoscope: comparison of
brush biopsy, imprint cytology of forceps biopsy, and histol-
ogy of forceps biopsy. Cancer 1991; 67:7275
49 Mak VH, Johnston ID, Hetzel MR, et al. Value of washings
and brushings at fibreoptic bronchoscopy in the diagnosis of
lung cancer. Thorax 1990; 45:373376
50 Gay PC, Brutinel WM. Transbronchial needle aspiration in
the practice of bronchoscopy. Mayo Clin Proc 1989; 64:158
162
51 Saita S, Tanzillo A, Riscica C, et al. Bronchial brushing and
biopsy: a comparative evaluation in diagnosing visible bron-
chial lesions. Eur J Cardiothorac Surg 1990; 4:270272
52 Cox ID, Bagg LR, Russell NJ, et al. Relationship of radio-
logic position to the diagnostic yield of fiberoptic bronchos-
copy in bronchial carcinoma. Chest 1984; 85:519522
53 Lam WK, So SY, Hsu C, et al. Fibreoptic bronchoscopy in
the diagnosis of bronchial cancer: comparison of washings,
brushings and biopsies in central and peripheral tumours.
Clin Oncol (R Coll Radiol) 1983; 9:3542
54 Zisholtz BM, Eisenberg H. Lung cancer cell type as a
determinant of bronchoscopy yield. Chest 1983; 84:428430
55 Gellert AR, Rudd RM, Sinha G, et al. Fibreoptic bronchos-
copy: effect of multiple bronchial biopsies on diagnostic
yield in bronchial carcinoma. Thorax 1982; 37:684687
56 McDougall JC, Cortese DA. Transbronchoscopic lung bi-
opsy for localized pulmonary disease. Semin Respir Med
1981; 3:3034
57 Radke JR, Conway WA, Eyler WR, et al. Diagnostic accu-
racy in peripheral lung lesions: factors predicting success
with flexible fiberoptic bronchoscopy. Chest 1979; 76:176
179
58 Stringfield JT, Markowitz DJ, Bentz RR, et al. The effect of
tumor size and location on diagnosis by fiberoptic bronchos-
copy. Chest 1977; 72:474476
59 Kvale PA, Bode FR, Kini S. Diagnostic accuracy in lung
cancer: comparison of techniques used in association with
flexible fiberoptic bronchoscopy. Chest 1976; 69:752757
60 Zavala DC. Diagnostic fiberoptic bronchoscopy: techniques
and results of biopsy in 600 patients. Chest 1975; 68:1219
61 Gasparini S, Zuccatosta L, Zitti P, et al. Integration of
TBNA, and TCNA in the diagnosis of peripheral lung
nodules: influence on staging. Ann Ital Chir 1999; 70:851
855
62 Reichenberger F, Weber J, Tamm M, et al. The value of
transbronchial needle aspiration in the diagnosis of periph-
eral pulmonary lesions. Chest 1999; 116:704708
63 Aristizabal JF, Young KR, Nath H. Can chest CT decrease
the use of preoperative bronchoscopy in the evaluation of
suspected bronchogenic carcinoma? Chest 1998; 113:1244
1249
64 Bilaceroglu S, Kumcuoglu Z, Alper H, et al. CT bronchus
sign-guided bronchoscopic multiple diagnostic procedures
in carcinomatous solitary pulmonary nodules and masses.
Respiration 1998; 65:4955
65 Wongsurakiat P, Wongbunnate S, Dejsomritrutai W, et al.
Diagnostic value of bronchoalveolar lavage and postbron-
choscopic sputum cytology in peripheral lung cancer. Respi-
rology 1998; 3:131137
66 Debeljak A, Mermolja M, Sorli J, et al. Bronchoalveolar
lavage in the diagnosis of peripheral primary and secondary
malignant lung tumors. Respiration 1994; 61:226230
67 de Gracia J, Bravo C, Miravitlles M, et al. Diagnostic value
of bronchoalveolar lavage in peripheral lung cancer. Am Rev
Respir Dis 1993; 147:649652
68 Torrington KG, Kern JD. The utility of fiberoptic bronchos-
copy in the evaluation of the solitary pulmonary nodule.
Chest 1993; 104:10211024
69 Pirozynski M. Bronchoalveolar lavage in the diagnosis of
peripheral, primary lung cancer. Chest 1992; 102:372374
70 Rennard SI, Albera C, Carratu L, et al. Clinical guidelines
and indications for bronchoalveolar lavage (BAL): pulmo-
nary malignancies. Eur Respir J 1990; 3:956957, 961969
71 Mori K, Yanase N, Kaneko M, et al. Diagnosis of peripheral
lung cancer in cases of tumors 2 cm or less in size. Chest
1989; 95:304308
72 Naidich DP, Sussman R, Kutcher WL, et al. Solitary
pulmonary nodules: CT-bronchoscopic correlation. Chest
1988; 93:595598
73 Wallace JM, Deutsch AL. Flexible fiberoptic bronchoscopy
and percutaneous needle lung aspiration for evaluating the
solitary pulmonary nodule. Chest 1982; 81:665671
74 Gasparini S, Ferretti M, Secchi EB, et al. Integration of
transbronchial and percutaneous approach in the diagnosis
of peripheral pulmonary nodules or masses: experience with
1,027 consecutive cases. Chest 1995; 108:131137
75 Lacasse Y, Wong E, Guyatt GH, et al. Transthoracic needle
aspiration biopsy for the diagnosis of localised pulmonary
lesions: a meta-analysis. Thorax 1999; 54:884893
76 Li H, Boiselle PM, Shepard JO, et al. Diagnostic accuracy
and safety of CT-guided percutaneous needle aspiration
biopsy of the lung: comparison of small and large pulmonary
nodules. AJR Am J Roentgenol 1996; 167:105109
77 Klein JS, Salomon G, Stewart EA. Transthoracic needle
biopsy with a coaxially placed 20-gauge automated cutting
needle: results in 122 patients. Radiology 1996; 198:715720
78 Milman N, Faurschou P, Grode G. Diagnostic yield of
transthoracic needle aspiration biopsy following negative
fiberoptic bronchoscopy in 103 patients with peripheral
circumscribed pulmonary lesions. Respiration 1995; 62:13
79 Zakowski MF, Gatscha RM, Zaman MB. Negative predic-
tive value of pulmonary fine needle aspiration cytology. Acta
Cytol 1992; 36:283286
80 Yang PC, Lee YC, Yu CJ, et al. Ultrasonographically guided
biopsy of thoracic tumors: a comparison of large-bore
cutting biopsy with fine-needle aspiration. Cancer 1992;
69:25532560
81 Cristallini EG, Ascani S, Farabi R, et al. Fine needle
aspiration biopsy in the diagnosis of intrathoracic masses.
Acta Cytol 1992; 36:416422
82 Calhoun P, Feldman PS, Armstrong P, et al. The clinical
outcome of needle aspirations of the lung when cancer is not
diagnosed. Ann Thorac Surg 1986; 41:592596
83 Lopez Hanninen E, Vogl TJ, Ricke J, et al. CT-guided
percutaneous core biopsies of pulmonary lesions: diagnostic
accuracy, complications and therapeutic impact. Acta Radiol
2001; 42:151155
84 Laurent F, Latrabe V, Vergier B, et al. CT-guided transtho-
racic needle biopsy of pulmonary nodules smaller than 20
mm: results with an automated 20-gauge coaxial cutting
needle. Clin Radiol 2000; 55:281287
85 Hirose T, Mori K, Machida S, et al. Computed tomographic
fluoroscopy-guided transthoracic needle biopsy for diagnosis
of pulmonary nodules. Jpn J Clin Oncol 2000; 30:259262
86 Charig MJ, Phillips AJ. CT-guided cutting needle biopsy of
lung lesions: safety and efficacy of an out-patient service.
Clin Radiol 2000; 55:964969
87 Swischuk JL, Castaneda F, Patel JC, et al. Percutaneous
transthoracic needle biopsy of the lung: review of 612
lesions. J Vasc Interv Radiol 1998; 9:347352
88 Lucidarme O, Howarth N, Finet JF, et al. Intrapulmonary
lesions: percutaneous automated biopsy with a detachable,
18-gauge, coaxial cutting needle. Radiology 1998; 207:759
765
89 Larscheid RC, Thorpe PE, Scott WJ. Percutaneous trans-
thoracic needle aspiration biopsy: a comprehensive review of
its current role in the diagnosis and treatment of lung
tumors. Chest 1998; 114:704709
90 Yankelevitz DF, Henschke CI, Koizumi JH, et al. CT-
guided transthoracic needle biopsy of small solitary pulmo-
nary nodules. Clin Imaging 1997; 21:107110
91 Westcott JL, Rao N, Colley DP. Transthoracic needle biopsy
of small pulmonary nodules. Radiology 1997; 202:97103
92 Santambrogio L, Nosotti M, Bellaviti N, et al. CT-guided
fine-needle aspiration cytology of solitary pulmonary nod-
ules: a prospective, randomized study of immediate cyto-
logic evaluation. Chest 1997; 112:423425
93 Cattelani L, Campodonico F, Rusca M, et al. CT-guided
transthoracic needle biopsy in the diagnosis of chest tu-
mours. J Cardiovasc Surg (Torino) 1997; 38:539542
94 Bo cking A, Klose KC, Kyll HJ, et al. Cytologic vs histologic
evaluation of needle biopsy of the lung, hilum and medias-
tinum: sensitivity, specificity and typing accuracy. Acta Cytol
1995; 39:463471
95 Baker RR, Stitik FP, Marsh BR. The clinical assessment of
selected patients with bronchogenic carcinoma. Ann Thorac
Surg 1975; 20:520528
96 Costrini AM. Transthoracic needle aspiration biopsy of focal
pulmonary lesions: a community hospital experience. South
Med J 1986; 79:2123
97 Meyer JE, Gandbhir LH, Milner LB, et al. Percutaneous
aspiration biopsy of nodular lung lesions. Thorac Cardiovasc
Surg 1977; 73:787791
98 Yazdi HM, MacDonald LL, Hickey NM. Thoracic fine
needle aspiration biopsy vs fine needle cutting biopsy: a
comparative study of 40 patients. Acta Cytol 1988; 32:635
640
99 Yuan A, Yang PC, Chang DB, et al. Ultrasound-guided
aspiration biopsy of small peripheral pulmonary nodules.
Chest 1992; 101:926930
100 Samuelsson L, Albrechtsson U, Tylen U. Fine-needle biopsy
of chest lesions. Radiologe 1982; 22:493496
101 Knudsen DU, Nielsen SM, Hariri J, et al. Ultrasonographi-
cally guided fine-needle aspiration biopsy of intrathoracic
tumors. Acta Radiol 1996; 37:327331
102 Garcia Rio F, Diaz Lobato S, Pino JM, et al. Value of
CT-guided fine needle aspiration in solitary pulmonary
nodules with negative fiberoptic bronchoscopy. Acta Radiol
1994; 35:478480
103 Burbank F, Kaye K, Belville J, et al. Image-guided auto-
mated core biopsies of the breast, chest, abdomen, and
pelvis. Radiology 1994; 191:165171
104 Targhetta R, Bourgeois JM, Marty-Double C, et al. Periph-
eral pulmonary lesions: ultrasonic features and ultrasonically
guided fine needle aspiration biopsy. J Ultrasound Med
1993; 12:369374
105 Grode G, Faurschou P, Milman N. Percutaneous transtho-
racic fine-needle lung biopsy with 3 different needles: a
retrospective study of results and complications in 224
patients. Respiration 1993; 60:284288
106 Collins CD, Breatnach E, Nath PH. Percutaneous needle
biopsy of lung nodules following failed bronchoscopic bi-
opsy. Eur J Radiol 1992; 15:4953
107 Veale D, Gilmartin JJ, Sumerling MD, et al. Prospective
evaluation of fine needle aspiration in the diagnosis of lung
cancer. Thorax 1988; 43:540544
108 Simpson RW, Johnson DA, Wold LE, et al. Transthoracic
needle aspiration biopsy: review of 233 cases. Acta Cytol
1988; 32:101104
109 Lovett JV, Manalo PB, Barcia TC, et al. Diagnosis of
pulmonary masses by fine-needle aspiration. Am J Surg
1988; 156:441445
110 Levine MS, Weiss JM, Harrell JH, et al. Transthoracic
needle aspiration biopsy following negative fiberoptic bron-
choscopy in solitary pulmonary nodules. Chest 1988; 93:
11521155
111 Balslov S, Vestbo J, Viskum KA. Value of Tru-cut lung
biopsy in focal and diffuse lung disease. Thorax 1988;
43:147150
112 Weisbrod GL, Herman SJ, Tao LC. Preliminary experience
with a dual cutting edge needle in thoracic percutaneous
fine-needle aspiration biopsy. Radiology 1987; 163:7578
113 Stanley JH, Fish GD, Andriole JG, et al. Lung lesions:
cytologic diagnosis by fine-needle biopsy. Radiology 1987;
162:389391
114 Winning AJ, McIvor J, Seed WA, et al. Interpretation of
negative results in fine needle aspiration of discrete pulmo-
nary lesions. Thorax 1986; 41:875879
115 Nahman BJ, Van Aman ME, McLemore WE, et al. Use of
the Rotex needle in percutaneous biopsy of pulmonary
malignancy. AJR Am J Roentgenol 1985; 145:9799
116 Lees WR, Hall-Craggs MA, Manhire A. Five years experi-
ence of fine-needle aspiration biopsy: 454 consecutive cases.
Clin Radiol 1985; 36:517520
117 Greene R, Szyfelbein WM, Isler RJ, et al. Supplementary
tissue-core histology from fine-needle transthoracic aspira-
tion biopsy. AJR Am J Roentgenol 1985; 144:787792
118 Crosby JH, Hager B, Hoeg K. Transthoracic fine-needle
aspiration: experience in a cancer center. Cancer 1985;
56:25042507
119 Stevens GM, Jackman RJ. Outpatient needle biopsy of the
lung: its safety and utility. Radiology 1984; 151:301304
120 Harrison BD, Thorpe RS, Kitchener PG, et al. Percutaneous
Trucut lung biopsy in the diagnosis of localised pulmonary
lesions. Thorax 1984; 39:493499
121 McEvoy RD, Begley MD, Antic R. Percutaneous biopsy of
intrapulmonary mass lesions: experience with a disposable
cutting needle. Cancer 1983; 51:23212326
122 Johnson RD, Gobien RP, Valicenti JF Jr. Current status of
radiologically directed pulmonary thin needle aspiration
biopsy: an analysis of 200 consecutive biopsies and review of
the literature. Ann Clin Lab Sci 1983; 13:225239
123 Vine HS, Kasdon EJ, Simon M. Percutaneous lung biopsy
using the Lee needle and a track-obliterating technique.
Radiology 1982; 144:921922
124 Pilotti S, Rilke F, Gribaudi G, et al. Fine needle aspiration
biopsy cytology of primary and metastatic pulmonary tu-
mors. Acta Cytol 1982; 26:661666
125 Jamieson WR, Suen KC, Hicken P, et al. Reliability of
percutaneous needle aspiration biopsy for diagnosis of bron-
chogenic carcinoma. Cancer Detect Prev 1981; 4:331336
126 Allison DJ, Hemingway AP. Percutaneous needle biopsy of
the lung. BMJ 1981; 282:875878
127 Westcott JL. Direct percutaneous needle aspiration of lo-
calized pulmonary lesions: result in 422 patients. Radiology
1980; 137:3135
128 Taft PD, Szyfelbein WM, Greene R. A study of variability in
cytologic diagnoses based on pulmonary aspiration speci-
mens. Am J Clin Pathol 1980; 73:3640
129 Poe RH, Tobin RE. Sensitivity and specificity of needle
biopsy in lung malignancy. Am Rev Respir Dis 1980;
122:725729
130 Pak HY, Yokota S, Teplitz RL, et al. Rapid staining tech-
niques employed in fine needle aspirations of the lung. Acta
Cytol 1981; 25:178184
131 Flower CD, Verney GI. Percutaneous needle biopsy of
thoracic lesions: an evaluation of 300 biopsies. Clin Radiol
1979; 30:215218
132 Sagel SS, Ferguson TB, Forrest JV, et al. Percutaneous
transthoracic aspiration needle biopsy. Ann Thorac Surg
1978; 26:399405
133 Lalli AF, McCormack LJ, Zelch M, et al. Aspiration biopsies
of chest lesions. Radiology 1978; 127:3540
134 House AJ, Thomson KR. Evaluation of a new transthoracic
needle for biopsy of benign and malignant lung lesions. AJR
Am J Roentgenol 1977; 129:215220
135 Francis D. Aspiration biopsies from diagnostically difficult
pulmonary lesions: a consecutive case material. Acta Pathol
Microbiol Scand 1977; 85A:235239
136 Pavy RD, Antic R, Begley M. Percutaneous aspiration
biopsy of discrete lung lesions. Cancer 1974; 34:21092117
137 Stevens GM, Weigen JF, Lillington GA. Needle aspiration
biopsy of localized pulmonary lesions with amplified fluoro-
scopic guidance. Am J Roentgenol Radium Ther Nucl Med
1968; 103:561571
138 Nasiell M. Diagnosis of lung cancer by aspiration biopsy and
a comparison between this method and exfoliative cytology.
Acta Cytol 1967; 11:114119
139 King EB, Russell WM. Needle aspiration biopsy of the lung:
technique and cytologic morphology. Acta Cytol 1967; 11:
319324
140 Detterbeck FC, Rivera MP. Clinical presentation and diag-
nosis. In: Detterbeck FC, Rivera MP, Socinski MA, et al,
eds. Diagnosis and treatment of lung cancer: an evidence-
based guide for the practicing clinician. Philadelphia, PA:
WB Saunders, 2001; 4572
141 Matsuda M, Horai T, Nakamura S, et al. Bronchial brushing
and bronchial biopsy: comparison of diagnostic accuracy and
cell typing reliability in lung cancer. Thorax 1986; 41:475478
142 Payne CR, Hadfield JW, Stovin PG, et al. Diagnostic
accuracy of cytology and biopsy in primary bronchial carci-
noma. J Clin Pathol 1981; 34:773778
143 Clee MD, Duguid HL, Sinclair DJ. Accuracy of morpholog-
ical diagnosis of lung cancer in a department of respiratory
medicine. J Clin Pathol 1982; 35:414419
144 Pilotti S, Rilke F, Gribaudi G, et al. Sputum cytology for the
diagnosis of carcinoma of the lung. Acta Cytol 1982; 26:649
654
145 Johnston WW. Percutaneous fine needle aspiration biopsy of
the lung: a study of 1,015 patients. Acta Cytol 1984;
28:218224
146 Zaman MB, Hajdu SI, Melamed MR, et al. Transthoracic
aspiration cytology of pulmonary lesions. Semin Diagn
Pathol 1986; 3:176187
147 Young GP, Young I, Cowan DF, et al. The reliability of
fine-needle aspiration biopsy in the diagnosis of deep
lesions of the lung and mediastinum: experience with 250
cases using a modified technique. Diagn Cytopathol 1987;
3:17
2003;123;115-128 Chest
Gilbert Schreiber and Douglas C. McCrory
Lung Cancer: Summary of Published Evidence
Performance Characteristics of Different Modalities for Diagnosis of Suspected
& Services
Updated Information
S
References
S#BIBL
free at:
Citations
S#otherarticles
Reprints
2003;123;129-136 Chest
M. Patricia Rivera, Frank Detterbeck and Atul C. Mehta
Diagnosis of Lung Cancer: The Guidelines
ISSN: 0012-3692.
Diagnosis of Lung Cancer*
The Guidelines
M. Patricia Rivera, MD, FCCP; Frank Detterbeck, MD, FCCP; and
Atul C. Mehta, MD, FCCP
Lung cancer is usually suspected in individuals who have abnormal chest radiograph findings or
have symptoms caused by either local or systemic effects of the tumor. The method of diagnosis
of suspected lung cancer depends on the type of lung cancer (ie, small cell lung cancer or
non-small cell lung cancer), the size and location of the primary tumor, the presence of
metastasis, and the overall clinical status of the patient. Achieving a diagnosis and staging are
usually done in concert because the most efficient way to make a diagnosis often is dictated by the
stage of the cancer. The best sequence of studies and interventions in a particular patient involves
careful judgment of the probable reliability of a number of presumptive diagnostic issues, so as
to maximize the sensitivity and to avoid performing multiple or unnecessary invasive procedures.
In this article, we consider all manner of clinical presentations of lung cancer in light of currently
available diagnostic procedures. Published data supporting a particular diagnostic approach is
weighed based on the quality of the benefit as well as the estimated net benefit. Recommenda-
tions are graded in terms of strength to provide clinicians with guidance as to the most efficient
and approach to the diagnosis of lung cancer in individual patients.
(CHEST 2003; 123:129S136S)
Key words: bronchoscopy; lung neoplasm; sensitivity; specificity; sputum cytology; transthoracic needle aspiration
Abbreviations: FDG
18
F-2-fluoro-2-deoxy-D-glucose; FN false negative; FNA fine needle aspirate;
NSCLC non-small cell lung cancer; PET positron emission tomography; SCLC small cell lung cancer;
TBNA transbronchial needle aspiration; TTNA transthoracic needle aspiration
T
he radiographic findings and clinical presenta-
tion usually allow a presumptive differentiation
between small cell lung cancer (SCLC) and non-
SCLC (NSCLC). Massive lymphadenopathy and di-
rect mediastinal invasion are well-recognized phe-
nomena in SCLC.
1,2
A mass in, or adjacent to, the
hilum is a particular characteristic of SCLC and is
seen in about 78% of cases.
1,2
Not infrequently,
patients with SCLC present with paraneoplastic
syndromes.
3
These include the syndrome of inappro-
priate antidiuretic hormone, ectopic adrenocortico-
trophic hormone production, and the Lambert-
Eaton syndrome. If SCLC is suspected, the diagnosis
should be achieved by whatever means is easiest (ie,
sputum cytology, thoracentesis if an accessible pleu-
ral effusion is present, fine needle aspirate [FNA] of
a supraclavicular node or metastatic site, bronchos-
copy with or without transbronchial needle aspira-
tion [TBNA] of mediastinal nodes or submucosal
process). If the diagnosis of SCLC is established on
a biopsy of the primary lesion, the distinction be-
tween limited or extensive disease then is made
radiographically.
In patients who are suspected of having NSCLC,
the method of achieving a diagnosis is usually dic-
tated by the presumed stage of the disease. Patients
with suspected lung cancer who present with a
pleural effusion should undergo thoracentesis first in
order to differentiate between a malignant effusion
(ie, one due to malignant involvement of the pleura)
and a paramalignant effusion (ie, one due to other
factors such as lymphatic blockade, atelectasis, or
hypoproteinemia). Distinction between the two is
important because the finding of malignant cells in
the pleural fluid alters the stage and treatment of the
particular patient. Because pleural metastases are
more common in the visceral pleura
4
and tend to be
focal when there is involvement of the parietal
pleura, pleural fluid cytology is a more sensitive
diagnostic test than percutaneous pleural biopsy, the
latter being a blind sampling procedure.
57
When
*From the Departments of Medicine (Dr. Rivera) and Surgery
(Dr. Detterbeck), The University of North Carolina at Chapel
Hill, Chapel Hill, NC; and the Department of Medicine (Dr.
Mehta), The Cleveland Clinic Foundation, Cleveland OH.
Correspondence to: M. Patricia Rivera, MD, FCCP, Assistant
Professor of Medicine, University of North Carolina at Chapel
Hill, 420 Burnett-Womack Blvd, CB No. 7020, Chapel Hill, NC
27599; e-mail: mprivera@med.unc.edu
three separate pleural fluid specimens from a patient
with malignant pleural disease are submitted to an
experienced cytologist, one should expect a positive
diagnosis in about 80% of patients.
7,8
Percutaneous,
closed pleural biopsy is reported
6
to be diagnostic for
malignancy in about 50% of cases. Thoracoscopic
biopsy of the pleura is safe and can provide a
definitive diagnosis with a high degree of accuracy
and minimal risk to the patient.
9,10
The reported
sensitivity rate is 0.80 to 1, the specificity rate is 1,
and the negative predictive value is 0.93.
9,1113
False-
negative test results are more common with me-
sothelioma than with primary lung carcinoma.
11
Patients with metastatic NSCLC (stage IV disease)
usually present with constitutional symptoms (ie,
fatigue and weight loss), organ-specific symptoms
(ie, bone pain and neurologic symptoms), and/or
abnormal laboratory findings (ie, anemia, elevated
alkaline phosphatase levels, and/or elevated liver
enzyme levels). In many of these patients, a FNA or
a needle biopsy of a site of metastasis represents the
most efficient way both to make a diagnosis and to
confirm the stage. In some cases, however, the
metastatic site may be technically difficult to biopsy.
If metastatic disease can be predicted with a high
degree of accuracy on the basis of radiographic
findings (ie, multiple brain, liver, or bone lesions), it
may be more efficient to achieve a diagnosis of the
primary lung lesion by whatever method is easiest for
the patient (ie, sputum cytology, bronchoscopy, or
transthoracic needle aspiration [TTNA]). This deci-
sion must be made by weighing the technical con-
siderations involved in each approach as well as the
reliability of diagnosing an extrathoracic lesion as a
site of metastasis based on radiographic appearances
alone (see the article on clinical/noninvasive staging
elsewhere in this supplement). A joint decision
among the radiologist, the pulmonologist, and the
medical or radiation oncologist is the desirable ap-
proach.
NSCLC can present with extensive infiltration of
the mediastinum. In such patients, the diagnosis
should be achieved by the method that has the most
favorable risk/benefit ratio. Bronchoscopy with
TBNA for cytologic or histologic examination of
mediastinal lymph nodes has been shown to be a safe
procedure.
1417
Technical aspects that are frequently
emphasized to be important in achieving a high
success rate include accurate preparation of the
specimen, rapid on-site evaluation by a cytopatholo-
gist, and using the larger 19-gauge needles, which
provide better tissue samples for histologic evalua-
tion.
18,19
The overall sensitivity of TBNA is 0.76, and
the overall specificity is 0.96.
1422
(The reader is
referred to the article on invasive clinical staging of
non-small cell lung cancer elsewhere in this supple-
ment for a more detailed review on the performance
characteristics of TBNA for staging the mediasti-
num.) The negative predictive value of TBNA (0.71)
is not high enough to obviate the need for a further
confirmation of negative results. Mediastinoscopy is
warranted in patients with nondiagnostic results.
CT-guided TTNA of mediastinal masses or nodes
(ie, nodes 1.5 cm in size) can be performed
safely.
23
The role of TTNA in patients with extensive
mediastinal disease (defined as such extensive inva-
sion of the mediastinum by the tumor that discrete
lymph nodes can no longer be discerned) is usually
to confirm SCLC, or in patients with NSCLC who
are not surgical candidates because of the extent of
mediastinal disease.
In the case of a small (ie, 3 cm), solitary,
peripheral lung lesion that is suspicious for lung
cancer in a patient who appears to have early-stage
disease and is a surgical candidate, the diagnostic
dilemma generally centers around whether or not to
obtain a biopsy to confirm the diagnosis of cancer
before surgical resection is carried out. When the
lesion is moderately to highly suspicious for lung
cancer, an excisional biopsy performed via thoracos-
copy has a much higher sensitivity than TTNA and is
the most definitive method of diagnosing a periph-
eral lung nodule. TTNA has no role in patients with
a solitary lesion that is moderately or highly suspi-
cious for lung carcinoma who appear to have early
stage-disease and are candidates for surgical resec-
tion. (The reader is referred to the article on solitary
pulmonary nodules elsewhere in this supplement for
a more detailed review on the diagnostic approach to
the solitary pulmonary nodule.)
General Approach to Diagnosis
Recommendations
1. In patients suspected of having SCLC based on
the radiographic and clinical findings, the diag-
nosis should be obtained by whatever method is
easiest (ie, sputum cytology, thoracentesis,
FNA, or bronchoscopy, including TBNA), as
dictated by the patients presentation. Level of
evidence, fair; benefit, moderate; grade of rec-
ommendation, B
2. In patients suspected of having lung cancer
who have an accessible pleural effusion, a
definitive diagnosis of the pleural effusion via
thoracentesis should be made first. Level of
recommendation, B
3. In a patient with an accessible pleural effusion,
if pleural fluid cytology is negative (after at least
two thoracentesis procedures), thoracoscopy
should be performed as the next step. Level of
ommendation, B
who have a solitary extrathoracic site suspicious
of a metastasis, the diagnosis should be ob-
tained by a fine-needle aspiration or biopsy of
the distant site. Level of evidence, fair; benefit,
who present with lesions in multiple distant
sites that are typical for metastases but in whom
biopsy of a metastatic site would be technically
difficult, a diagnosis of the primary lung lesion
should be obtained by whatever method is
easiest and safest (ie, sputum cytology, bron-
choscopy, or TTNA). Level of evidence, poor;
benefit, moderate; grade of recommendation, C
who have no known distant metastases or pleu-
ral effusions but have extensive infiltration of
the mediastinum based on radiographic studies,
a diagnosis should be obtained from the medi-
astinal tissue by whatever method is most effi-
cacious (ie, bronchoscopy with TBNA, TTNA,
or mediastinoscopy) as dictated by the location
of the tumor. Level of evidence, fair; benefit,
7. A patient with a solitary peripheral lesion that is
even moderately suspicious for lung cancer,
who appears to have early-stage disease (ie,
negative findings of chest CT scan and
18
F-2-
fluoro-2-deoxy-D-glucose [FDG] positron
emission tomography [PET] scan of the medi-
astinum) and is a surgical candidate, should
undergo excisional biopsy and subsequent lo-
bectomy if a resectable lung cancer is con-
firmed. Level of evidence, poor; benefit, sub-
Diagnosis of Primary Tumor
A variety of techniques (eg, sputum cytology,
flexible bronchoscopy, and TTNA) are available as
methods of achieving a definitive diagnosis. The
selection of the most appropriate test is best per-
formed in a multidisciplinary fashion. Four key
questions on the diagnostic workup of patients with
lung cancer were formulated and answered by a
comprehensive review of the published literature
performed by the Duke University Center for Clin-
ical Health Policy Research. (Please refer to a pre-
vious article in this supplement entitled: Perfor-
mance Characteristics of Different Modalities for
Diagnosis of Suspected Lung Cancer: Summary of
Published Evidence, which hereafter is referred to
as the background article, for the detailed review and
tables.) The following section summarizes the results
of the data and provides recommendations based on
a critical review of the published evidence.
Sputum Cytology
What Are the Performance Characteristics of Spu-
tum Cytology for the Diagnosis of Lung Cancer With
Special Consideration for the Location of the Tumor?
Sputum cytology is the least invasive means of
obtaining a diagnosis in a patient who is suspected of
having lung cancer. The diagnostic accuracy of spu-
tum cytology is dependent on rigorous specimen
sampling (at least three specimens) and preservation
techniques, as well as on the location and size of the
tumor (ie, central vs peripheral). Unfortunately,
many institutions do not have an established pro-
gram for sputum collection and processing, and
therefore sputum analysis has a much lower sensitiv-
ity than that indicated in the data presented here
(which come from institutions with well-established
sputum analysis programs). Sputum cytology is par-
ticularly useful in patients who present with centrally
located tumors (ie, SCLC or squamous call carci-
noma) and in those who present with hemoptysis.
The sampling of sputum specimens should certainly
be the first step in a patient who presents with a
central lesion with or without radiographic evidence
of metastatic disease, in whom a semi-invasive pro-
cedure such as bronchoscopy or TTNA might pose a
higher risk. A systematic literature review by the
Duke University Center for Clinical Health Policy
Research found 16 studies that provided data on the
performance characteristics of sputum cytology for
the diagnosis of suspected lung cancer (see Table 1
in the background article). The results of these
studies revealed the average sensitivity and specific-
ity rate of sputum cytology to be 0.66 and 0.99,
respectively. Eight studies provided data on the
sensitivity of prebronchoscopy sputum analysis in
patients who were suspected of having lung cancer
(see evidence Table 2 in the background article) and
revealed a prebronchoscopy sputum sensitivity rang-
ing from 0.10 to 0.74, with an average sensitivity of
0.22. Seventeen studies that addressed the effect of
location of the lesion on the sensitivity of sputum
cytology were identified (see Table 3 in the back-
ground article). The sensitivity of sputum cytology
for central lesions is 0.71 and decreases to 0.49 for
peripheral lesions.
Recommendations
8. In patients with a central lesion who present
with or without hemoptysis, sputum cytology
(at least three specimens) is a reasonable first
step (in centers with a formal program directed
at the acquisition, handling, and interpretation
of sputum samples) in the diagnostic workup.
9. In patients with a peripheral lesion that is
suspicious for lung cancer, sputum cytology (in
centers with a formal program directed at the
acquisition, handling, and interpretation of spu-
tum samples) may confirm the diagnosis of lung
cancer. However, further testing to diagnose
definitively a peripheral lung lesion must follow
a negative result on sputum cytology. Level of
ommendation, B
Flexible Bronchoscopy
What Are the Performance Characteristics of Flex-
ible Bronchoscopy and Its Ancillary Procedures for
the Diagnosis of Central Tumors (Endobronchial), as
Opposed to Peripheral Tumors, and to Peripheral
Tumors 2 cm and 2 cm in Size? Flexible
bronchoscopy with its attendant procedures is a
valuable diagnostic procedure in the workup of a
patient who is suspected of having lung cancer. A
comprehensive literature search conducted by the
Research revealed 44 studies, each with 50 pa-
tients, that reported on the sensitivity of flexible
bronchoscopy for the diagnosis of lung cancer. The
decision about whether to perform a diagnostic
bronchoscopy for a lesion that is suspicious for lung
cancer largely depends on the location of the lesion
(ie, central vs peripheral). Central lesions can
present as an exophytic endobronchial mass, submu-
cosal spread, or a peribronchial tumor causing ex-
trinsic compression. Thirty of the 44 studies that
were found reported on the sensitivity of bronchos-
copy for central, endobronchial lesions (see Table 4
in the background article). Among a total of 3,754
patients, the overall sensitivity of flexible bronchos-
copy was 0.88. Direct forceps biopsy of visible
central lesions is the technique used most frequently,
and the sensitivity of this test by itself was 0.74. At
least three forceps biopsies of the visible lesion are
recommended. The sensitivity from washings and
brushings is somewhat lower (0.48 and 0.59, respec-
tively), but these tests often are combined with
forceps biopsies. The addition of TBNA to obtain
cytology or histology samples when there is submu-
cosal tumor spread or peribronchial tumor causing
extrinsic compression increases the sensitivity of
bronchoscopy.
24,25
Peripheral lesions are defined in most studies as
lesions that are not visible in the main or lobar
airways, and thus it is not surprising that the sensi-
tivity of flexible bronchoscopy for diagnosing periph-
eral lung cancers is lower than that for central
lesions. Thirty of the 44 studies also reported on the
sensitivity of flexible bronchoscopy for peripheral
lesions (see Table 5 in the background article).
Among a total of 4,136 patients, the average sensi-
tivity of flexible bronchoscopy for peripheral lesions
was 0.69. A few points must be made in order to
interpret the results of bronchoscopy in the diagnosis
of peripheral lung cancers. First, most of the studies
used fluoroscopy routinely for peripheral lesions,
which increases the reported sensitivity of bronchos-
copy.
26
Second, the number of transbronchial biopsy
samples obtained is important, with a sensitivity of
0.45 for one sample and 0.70 for six samples being
reported in one study.
27
The sensitivity of bronchoscopy for peripheral
lesions is most affected by the size of the lesion.
Eight studies that reported on the sensitivity of
flexible bronchoscopy based on the size of the
peripheral lesion were identified by the Duke Uni-
versity Center for Clinical Health Policy Research
(see Table 6 in the background article). In 879
patients with lesions that were 2 cm in size, the
sensitivity of bronchoscopy was 0.62, whereas in 341
patients with lesions 2 cm in size, the sensitivity
was only 0.33. The sensitivity of bronchoscopy is
reported to be higher if a CT scan shows a bronchus
extending to the lesion (0.60 vs 0.25, respectively).
28,29
The false-negative (FN) rate for bronchoscopy
results has not yet been defined. Most clinicians
would pursue the diagnosis further in the case of a
nondiagnostic bronchoscopy of a visible endobron-
chial abnormality. The FN rate can be estimated to
be fairly high in the case of peripheral lesions,
especially smaller ones, because of the relative low
sensitivity in this setting. Bronchoscopy has an im-
portant role in the diagnosis of benign conditions,
but the chance of finding a benign condition in a
patient who is clinically suspected of having lung
cancer is only 1%.
30
Recommendations
10. In a patient with a central lesion, bronchos-
copy is the most sensitive way to confirm a
diagnosis of cancer. Level of evidence, fair;
benefit, substantial; grade of recommen-
dation, B
11. In a patient with a central lesion that is
suspicious for lung cancer, further testing to
definitively rule out cancer must follow a
nonspecific benign result on bronchoscopy.
12. In a patient with a small (ie, 2 cm) periph-
eral lesion, the sensitivity of bronchoscopy is
low. Therefore, a nonspecific result on bron-
choscopy of a peripheral lesion that is suspi-
cious for lung cancer requires further testing
to definitively rule out cancer. Level of evi-
recommendation, A
TTNA
What Are the Performance Characteristics for
TTNA as a Diagnostic Modality With Particular
Emphasis on the Size and Location of the Suspected
Cancer? A meta-analysis of 47 studies and an addi-
tional 19 studies focusing on the performance char-
acteristics of TBNA or biopsy for the diagnosis of
localized pulmonary lesions were analyzed by the
Research. Five studies with 50 patients included
in the meta-analysis were excluded from the final
analysis (see Table 8 in the background article).
Among 11,279 patients, the overall sensitivity and
specificity of TTNA for diagnosing peripheral lung
cancers were 0.90 and 0.97, respectively. The differ-
ence between the sensitivity of TTNA in lesions that
were 2 cm in size and those that were 2 cm in
size was not statistically different (0.95 and 0.91,
respectively).
TTNA of a peripheral lung lesion can be per-
formed under either fluoroscopic or CT scan guid-
ance. Studies using CT scan guidance show a higher
sensitivity (0.92) than those using fluoroscopy guid-
ance (0.88). For peripheral lung lesions, the sensi-
tivity of TTNA is higher than that of bronchoscopy.
In patients who have lung cancer, TTNA has approx-
imately a 90% chance of providing confirmation of
the diagnosis. Furthermore, given the false-positive
rate of 0.01 to 0.02, a TTNA result that is positive for
cancer is reliable. On the other hand, the FN test
result rate of TTNA is high (range, 0.20 to 0.30).
31
Thus, TTNA is generally not useful in ruling out
cancer. As such, TTNA has no role in patients who
have lesions that are even moderately suspicious for
lung cancer, and who appear to have early-stage
disease and are candidates for surgical resection.
Although a test that could reliably rule out lung
cancer might be useful in this setting, the high FN
rate of TTNA makes reliance on a negative result
untenable.
Establishing a specific benign diagnosis such as
tuberculosis, fungal infection, or hamartoma, using
TTNA is quite valuable, particularly in patients in
whom the clinical and radiologic findings strongly
suggest a benign diagnosis. In such cases, a specific
benign diagnosis using TTNA further decreased the
risk of missing a cancer.
PET scanning using FDG has proven to be an
excellent modality for evaluating solitary pulmonary
nodules. In a recently published meta-analysis
32
of
the available data on FDG-PET scanning, the aver-
age sensitivity and specificity of FDG-PET scanning
for detecting a malignancy were reported to be 0.97
and 0.78, respectively. Like any test, PET scanning
has some limitations. The current generation of PET
scanners can miss lesions that are 1 cm in size,
3234
and FN results can occur when dealing with carci-
noid tumors or bronchoalveolar carcinomas.
32,34,35
False-positive results may be seen with certain in-
flammatory or infectious lesions such as tuberculo-
mas, histoplasmomas, and rheumatoid nodules.
32,34
(The reader is referred to the article on solitary
pulmonary nodule elsewhere in this supplement for
a more detailed discussion of FDG-PET scanning in
the evaluation of the solitary pulmonary nodules.)
Recommendations
13. In the case of a peripheral lung lesion, TTNA
has a much higher sensitivity than bronchos-
copy. It is the procedure of choice for con-
firming the diagnosis of lung cancer in pa-
tients in whom it is indicated (ie, patients for
whom preoperative therapy is planned or sur-
gery is not feasible). Level of evidence, good;
dation, A
14. A nonspecific result using TTNA of a lesion
that is suspicious of being a lung cancer carries
a high FN rate, and therefore further testing
to establish a definitive diagnosis should be
pursued. Level of evidence, good; benefit,
15. In patients with a lesion that is even moder-
ately suspicious for lung cancer, who appear to
have limited disease (ie, negative findings of
chest CT scan and FDG-PET scan of the
mediastinum) and are surgical candidates,
TTNA has no role (unless preoperative ther-
apy is planned). These patients should un-
dergo excisional biopsy and subsequent lobec-
tomy if a lung cancer is confirmed. Level of
recommendation, B
Cell Type Accuracy
What Is the Diagnostic Error When Differentiat-
ing Between NSCLC and SCLC Generated by Var-
ious Diagnostic Techniques (Bronchoscopy, TTNA,
and Sputum Cytology)? In a patient with lung
cancer, distinguishing between SCLC and NSCLC is
of paramount importance as each of these cancers is
treated in a radically different manner. The distinc-
tion between SCLC and NSCLC on sputum cytology,
TTNA cytology, and bronchoscopic washings, brush-
ings, and BAL cytology is quite reliable. Twenty-one
studies with a total of 6,305 patients were systematically
reviewed for the analysis of the diagnostic accuracy
of various techniques in differentiating NSCLC from
SCLC (see Table 9 in the background article).
Indeed, the chance that a preoperative diagnosis of
NSCLC is in error (ie, that the tumor is actually
SCLC) is 0.02 (range, 0.01 to 0.07). On the other
hand, the error rate of a diagnosis of SCLC (ie, that
the tumor is actually NSCLC) is on average 0.09,
with individual studies ranging from 0 to 0.33.
Recommendations
16. In a patient who is suspected of having lung
cancer, the diagnosis of NSCLC that is made
on cytology (ie, with sputum, TTNA speci-
mens, or bronchoscopic specimens) is highly
reliable and can be accepted with a high degree
of certainty. Level of evidence, good; benefit,
17. The possibility of an erroneous diagnosis of
SCLC on a cytology specimen must be kept in
mind if the clinical presentation or clinical
course is not consistent with that of SCLC. In
such a case, further testing to establish a
definitive cell type should be pursued. Level
of evidence, good; benefit, substantial; grade
of recommendation, A
Conclusion
A variety of techniques is available to assist the
clinician in achieving a definitive diagnosis of lung
cancer. Selection of the most appropriate test is best
done in a multidisciplinary fashion with input from a
pulmonologist, chest radiologist, and thoracic sur-
geon. Furthermore, the most appropriate test is
usually determined by the type of lung cancer (ie,
SCLC or NSCLC), the size and location of the
tumor, and the presumed stage of the cancer.
A diagnosis should be obtained by whatever
method is easiest in patients who are presumed to
have SCLC or who have very clear evidence of
advanced NSCLC (ie, large pleural effusions or
metastatic disease). Sputum cytology is a reasonable
first step in patients with central lesions, but diag-
nostic accuracy depends of the rigorous acquisition,
handling, and interpretation of samples. Flexible
bronchoscopy is the most useful test for central
lesions, whereas in the case of peripheral lesions, the
sensitivity of TTNA is higher than that of bron-
choscopy.
General Approach to Diagnosis
1. In patients suspected of having SCLC based on
the radiographic and clinical findings, the diag-
nosis should be obtained by whatever method is
easiest (ie, sputum cytology, FNA, and bron-
choscopy, including TBNA), as dictated by the
patients presentation. Level of evidence, fair;
benefit, moderate; grade of recommendation, B
who have an accessible pleural effusion, a
definitive diagnosis of the pleural effusion via
thoracentesis should be made first. Level of
recommendation, B
3. In a patient with an accessible pleural effusion,
if the results of pleural fluid cytology are neg-
ative (after at least two thoracentesis proce-
dures), thoracoscopy should be performed as
the next step. Level of evidence, fair; benefit,
4. In patients who are suspected of having lung
cancer and who have a solitary extrathoracic
site that is suspicious of metastasis, the diagno-
sis should be obtained by a fine-needle aspira-
tion or biopsy of the distant site. Level of
recommendation, B
cancer and who present with lesions in multiple
distant sites that are typical for metastases, but
in whom biopsy of a metastatic site would be
technically difficult, a diagnosis of the primary
lung lesion should be obtained by whatever
method is easiest and safest (ie, sputum cytol-
ogy, bronchoscopy, or TTNA). Level of evi-
dence, poor; benefit, moderate; grade of rec-
ommendation, C
cancer and who have no known distant metas-
tases or pleural effusions but have extensive
infiltration of the mediastinum based on radio-
graphic studies, a diagnosis should be obtained
from the mediastinal tissue by whatever
method is most efficacious (ie, bronchoscopy
with TBNA, TTNA, or mediastinoscopy) as
dictated by the location of the tumor. Level of
ommendation, B
7. A patient with a solitary peripheral lesion that is
even moderately suspicious for lung cancer,
who appears to have early-stage disease (ie,
negative findings on a chest CT of the medias-
tinum) and is a surgical candidate, should un-
dergo excisional biopsy and subsequent lobec-
tomy if a resectable lung cancer is confirmed.
Diagnosis of Primary Tumor
8. In patients with a central lesion who present
with or without hemoptysis, sputum cytology
(at least three specimens) is a reasonable first
step (in centers with a formal program di-
rected at the acquisition, handling, and inter-
pretation of sputum samples) in the diagnostic
workup. Level of evidence, fair; benefit, sub-
9. In patients with a peripheral lesion that is
suspicious for lung cancer, sputum cytology
(in centers with a formal program directed at
the acquisition, handling, and interpretation
of sputum samples) may confirm the diagnosis
of lung cancer. However, further testing to
diagnose definitively a peripheral lung lesion
must follow a negative result on sputum cy-
tology. Level of evidence, fair; benefit, mod-
erate; grade of recommendation, B
10. In a patient with a central lesion, bronchos-
copy is the most sensitive way to confirm
a diagnosis of cancer. Level of evidence,
fair; benefit, substantial; grade of recom-
mendation, B
11. In a patient with a central lesion that is
suspicious for lung cancer, further testing to
definitively rule out cancer must follow a
nonspecific benign result on bronchoscopy.
12. In a patient with a small (ie, 2 cm) periph-
eral lesion, the sensitivity of bronchoscopy is
low. Therefore, a nonspecific result on bron-
choscopy of a peripheral lesion that is suspi-
cious for lung cancer requires further testing
to definitively rule out cancer. Level of evi-
recommendation, A
13. In the case of a peripheral lung lesion, TTNA
has a much higher sensitivity than bronchos-
copy. It is the procedure of choice for con-
firming the diagnosis of lung cancer in pa-
tients in whom it is indicated (ie, those in
whom preoperative therapy is planned or sur-
gery is not feasible). Level of evidence, good;
dation, A
14. A nonspecific result on TTNA of a lesion that
is suspicious of being a lung cancer carries a
high FN rate, and therefore further testing to
establish a definitive diagnosis should be pur-
sued. Level of evidence, good; benefit, sub-
stantial; grade of recommendation, A
15. In patients with a lesion that is even moder-
ately suspicious for lung cancer who appear to
have limited disease and are surgical candi-
dates, TTNA has no role (unless preoperative
therapy is planned). These patients should
undergo excisional biopsy and subsequent lo-
bectomy if a lung cancer is confirmed. Level
recommendation, B
16. In a patient who is suspected of having lung
cancer, the diagnosis of NSCLC made on
cytology (ie, from sputum, TTNA specimens,
or bronchoscopic specimens) is highly reliable
and can be accepted with a high degree of
certainty. Level of evidence, good; benefit,
17. The possibility of an erroneous diagnosis of
SCLC on a cytology specimen must be kept in
mind if the clinical presentation or clinical
course is not consistent with that of SCLC. In
such a case, further testing to establish a
definitive cell type should be pursued. Level
References
1 Foster BB, Muller NL, Miller RR, et al. Neuroendocrine
carcinomas of the lung: clinical, radiologic and pathologic
correlation. Radiology 1989; 170:441445
2 Pearlberg JL, Sandler MA, Lewis JW. Small-cell Broncho-
genic carcinoma: CT evaluation. Radiology 1988; 150:265
268
3 Patel AM, Davila DG, Peters SG. Paraneoplastic syndromes
associated with lung cancer. Mayo Clin Proc 1993; 68:278
287
4 Rodriguez-Panadero F, Bordens Naranjo F, Lopez-Mejias J.
Pleural metastatic tumors and effusions: frequency and
pathogenic mechanisms in a post-mortem series. Eur Respir
J 1989; 2:366369
5 Sahn SA. Malignancy metastatic to the pleura. Clin Chest
Med 1998; 19:351361
6 Prakash UBS, Reinman HM. Comparison of needle biopsy
with cytologic analysis for the evaluation of pleural effusion:
analysis of 414 cases. Mayo Clin Proc 1985; 60:158164
7 Johnson WW. The malignant pleural effusion: a review of
cytopathologic diagnoses of 584 specimens for 472 consecu-
tive patients. Cancer 1985; 56; 905909
8 Sears D, Hadju SI. The cytologic diagnosis of malignant
neoplasms in pleural and peritoneal effusions. Acta Cytol
1987; 31:8792
9 Boutin C, Viallat JR, Cargnino P, et al. Thoracoscopy in
malignant pleural effusions. Am Rev Respir Dis 1981; 124:
588592
10 Rusch VW. Thoracoscopy under regional anesthesia for the
diagnosis and management of pleural disease. Am J Surg
1987; 154:274278
11 Menzies R, Charbonneau M. Thoracoscopy for the diagnosis
of pleural disease. Ann Intern Med 1991; 114:271276
12 Page RD, Jeffrey RR, Donnelly RJ. Thoracoscopy: a review of
121 consecutive surgical procedures. Ann Thorac Surg 1989;
48:6668
13 Harris RJ, Kavuru MA, Mehta AC, et al. The impact of
thoracoscopy on the management of pleural disease. Chest
1995; 107:845852
1993; 104:10121016
15 Schenk DA, Bower JH, Bryan CL, et al. Transbronchial
needle aspiration staging of bronchogenic carcinoma. Am Rev
Respir Dis 1986; 134:146148
16 Schenk DA, Strollo PJ, Pickard JS. The utility of the Wang
18-gauge transbronchial histology needle in the staging of
bronchogenic carcinoma. Chest 1989; 96:272274
17 Harrow EM, Abi-Saleh W, Blum J, et al. The utility of trans-
bronchial needle aspiration in the staging of bronchogenic
carcinoma. Am J Respir Crit Care Med 2000; 161:601607
18 Davenport RD. Rapid on-site evaluation of transbronchial
aspirates. Chest 1990; 98:5961
19 Schenk DA, Chambers SL, Derdak S, et al. Comparison of the
Wang 19 gauge and 22 gauge needles in the mediastinal staging
of lung cancer. Am Rev Respir Dis 1993; 147:12511258
20 Vansteenkiste J, Lacquet LM, Demedts M, et al. Transcarinal
needle aspiration biopsy in the staging of lung cancer. Eur
Respir J 1994; 7:265268
21 Ratto GB, Mereu C, Motta G. The prognostic significance of
preoperative assessment of mediastinal lymph nodes in pa-
tients with lung cancer. Chest 1988; 93:807813
22 Garspestad E, Goldberg S, Herth F, et al. CT fluoroscopy
guidance for transbronchial needle aspiration: an experience
in 35 patients. Chest 2001; 119:329332
23 Protopapas Z, Westcott JL. Transthoracic needle biopsy of
mediastinal lymph nodes for staging lung cancer and other
cancers. Radiology 1996; 199:489496
24 Dasgupta A, Jain P, Sandur S, et al. Utility of transbron-
chial needle aspiration in the diagnosis of endobronchial
lesions [abstract]. Am J Respir Crit Care Med 1998;
157:A703
25 Shure D, Fedullo PF. Transbronchial needle aspiration in the
diagnosis of submucosal and peribronchial bronchogenic
carcinoma. Chest 1985; 88:4951
26 Cox ID, Bagg LR, Russell NJ, et al. Relationship of radiologic
position to the diagnostic yield of fiberoptic bronchoscopy in
bronchial carcinoma. Chest 1984; 85:519522
27 Popovich J Jr, Kvale PA, Eichenhorn MS, et al. Diagnostic
accuracy of multiple biopsies from flexible Fiberoptic bron-
choscopy. Am Rev Respir Dis 1982; 125:521523
28 Naidich DP, Sussman R, Kutcher WL, et al. Solitary pulmo-
nary nodules: CT bronchoscopic correlation. Chest 1988;
93:595598
29 Gaeta M, Pandolfo I, Volta S, et al. Bronchus sign on CT in
peripheral carcinoma of the lung: value in predicting results
of transbronchial biopsy. AJR Am J Roentgenol 1991; 157:
11811185
30 Rolston KVI, Rodriguez S, Dhilakia N, et al. Pulmonary
infections mimicking cancer: a retrospective, three-year re-
view. Support Care Cancer 1997; 5:9095
31 Zarbo RJ, Fenoglio-Preiser CM. Institutional database for
comparison of performance in lung fine-needle aspiration
cytology: a College of American Pathologists Q-Probe study
of 5264 cases with histologic correlation. Arch Pathol Lab
Med 1992; 116:463470
32 Gould MK, Maclean CC, Kuschner WG, et al. Accuracy of
positron emission tomography for diagnosis of pulmonary
nodules and mass lesions: a meta-analysis. JAMA 2001;
285:914924
33 Knight SB, Delbeke D, Stewart JR, et al. Evaluation of
pulmonary lesions with FDG-PET: comparison of findings in
patients with and without a history of prior malignancy. Chest
1996; 109:982988
34 Guhlmann A, Storck M, Kotzerke J, et al. Lymph node staging in
non-small cell lung cancer: evaluation by FDG positron emission
tomography (PET). Thorax 2997; 52:438441
35 Gupta NC, Maloof J, Gunel E. Probability of malignancy in
solitary pulmonary nodules using fluorine-18-FDG and PET.
J Nucl Med 1996; 37:943948
2003;123;129-136 Chest
M. Patricia Rivera, Frank Detterbeck and Atul C. Mehta
Diagnosis of Lung Cancer: The Guidelines
& Services
Updated Information
S
References
S#BIBL
free at:
Reprints
2003;123;137-146 Chest
Eric M. Toloza, Linda Harpole and Douglas C. McCrory
Evidence
Noninvasive Staging of Non-small Cell Lung Cancer: A Review of the Current
ISSN: 0012-3692.
Noninvasive Staging of Non-small Cell
Lung Cancer*
A Review of the Current Evidence
Eric M. Toloza, MD, PhD; Linda Harpole, MD, MPH; and
Douglas C. McCrory, MD, MHS
Study objectives: To determine the test performance characteristics of CT scanning, positron
emission tomography (PET) scanning, MRI, and endoscopic ultrasound (EUS) for staging the
mediastinum, and to evaluate the accuracy of the clinical evaluation (ie, symptoms, physical
findings, or routine blood test results) for predicting metastatic disease in patients in whom
non-small cell lung cancer or small cell lung cancer is diagnosed.
Design, setting, and participants: Systematic searches of MEDLINE, HealthStar, and Cochrane
Library databases to July 2001, and of print bibliographies. Studies evaluating the staging results
of CT scanning, PET scanning, MRI, or EUS, with either tissue histologic confirmation or
long-term clinical follow-up, were included. The performance of the clinical evaluation was
compared against the results of brain and abdominal CT scans and radionuclide bone scans.
Measurement and results: Pooled sensitivities and specificities for staging the mediastinum were
as follows: for CT scanning: sensitivity, 0.57 (95% confidence interval [CI], 0.49 to 0.66);
specificity, 0.82 (95% CI, 0.77 to 0.86); for PET scanning: sensitivity, 0.84 (95% CI, 0.78 to 0.89);
specificity, 0.89 (95%CI, 0.83 to 0.93); and for EUS: sensitivity, 0.78 (95%CI, 0.61 to 0.89); specificity,
0.71 (95% CI, 0.56 to 0.82). For the evaluation of brain metastases, the summary estimate of the
negative predictive value (NPV) of the clinical neurologic evaluation was 0.94 (95% CI, 0.91
to 0.96). For detecting adrenal and/or liver metastases, the summary NPV of the clinical
evaluation was 0.95 (95% CI, 0.93 to 0.96), and for detecting bone metastases, it was 0.90
(95% CI, 0.86 to 0.93).
Conclusions: PET scanning is more accurate than CT scanning or EUS for detecting
mediastinal metastases. The NPVs of the clinical evaluations for brain, abdominal, and bone
metastases are > 90%, suggesting that routinely imaging asymptomatic lung cancer patients
may not be necessary. However, more definitive prospective studies that better define the
patient population and improved reference standards are necessary to more accurately assess
the true NPV of the clinical evaluation. (CHEST 2003; 123:137S146S)
Key words: CT; lung neoplasms; meta-analysis; predictive value of tests; radiograph; sensitivity; specificity
Abbreviations: CI confidence interval; EUS endoscopic ultrasound; NPV negative predictive value;
NSCLC non-small cell lung cancer; PET positron emission tomography; PPV positive predictive value;
ROC receiver operating characteristic; SCLC small cell lung cancer
E
valuation of patients with suspected lung cancer
includes both a diagnosis of the primary tumor
and an evaluation of the extent of spread to regional
or distant lymph nodes or to other structures. The
current system for staging lung cancer is based on
the TNM classification.
1,2
The staging of lung cancer
patients not only provides important prognostic in-
formation with regard to survival, but also guides the
decision-making process with regard to the choosing
optimal treatment modality.
Mediastinal lymph node involvement is found in
26% of newly diagnosed lung cancer patients, and
extrathoracic metastases are found in 49%.
3
In pa-
tients with non-small cell lung cancer (NSCLC),
those with mediastinal lymph node involvement are
classified as having stage III disease, and those with
extrathoracic metastases as having stage IV disease.
For patients with small cell lung cancer (SCLC),
both types of patients would be classified as having
*From the Departments of Surgery (Dr. Toloza) and Medicine
(Drs. Harpole and McCrory), Duke University Medical Center,
Durham, NC.
Correspondence to: Eric M. Toloza, MD, PhD, Duke Thoracic
Oncology, DUMC Box 3048, Durham, NC 27710; e-mail:
toloz001@mc.duke.edu
extensive disease. The remainder of newly diagnosed
lung cancer patients who present without mediasti-
nal lymph node or extrathoracic metastases are said
to have stage I or II NSCLC or limited-stage SCLC.
The evaluation of a patient who presents with
newly suspected lung cancer includes a clinical
evaluation to assess for the presence or absence of
extrathoracic metastatic disease and an imaging pro-
cedure to evaluate the mediastinum. As the most
common sites of lung cancer metastases are the
brain, bones, liver, and adrenal glands, the clinical
evaluation ordinarily includes an assessment for
headache, focal neurologic deficits, seizures, or
changes in personality that may suggest brain metas-
tases; bone pain or pathologic fractures that may
suggest bone metastases; and abdominal or flank
pain, or serum enzyme levels that may suggest liver
metastases. Adrenal metastases are usually asymp-
tomatic. Noninvasive imaging of the mediastinum
may include the use of CT scanning, MRI, endo-
scopic ultrasound (EUS), or positron emission to-
mography (PET) scanning.
The purpose of the present study is twofold, as follows: (1) to
assess the performance characteristics of noninvasive imaging
procedures for staging the mediastinum; and (2) to assess the
negative predictive value (NPV) of the clinical evaluation for
predicting extrathoracic metastases. The NPV is the probability
that a negative result correctly indicates the absence of metastatic
disease.
Key Questions
The following two key questions were identified:
1. What are the sensitivities and specificities of CT scanning,
MRI, EUS, and PET scanning for detecting malignant
mediastinal lymph node involvement in lung cancer pa-
tients?
2. How accurate is the clinical evaluation (ie, symptoms,
physical findings, or routine blood tests) for predicting
metastatic disease among patients in whom lung cancer (ie,
NSCLC or SCLC) has been diagnosed?
Search Strategy
Computerized searches of the MEDLINE bibliographic data-
base (January 1991 to July 2001), HealthStar, and the Cochrane
Library were performed. The decision to limit the search to the
past 10 years was based on the evolution of technology after
consultation with diagnostic radiologists at Duke University
Medical Center. Key words used for the search included lung
neoplasm, bronchial neoplasm, neoplasm staging, neoplasm me-
tastasis, lymphatic metastasis, CT, mediastinum radiography,
emission-CT, adrenal gland neoplasms, and sensitivity and spec-
ificity. In addition, we searched the reference lists of included
studies, selected textbooks, practice guidelines, systematic re-
views, and meta-analyses in order to ensure that all relevant
studies were identified. Only articles that had been published in
English were considered.
To address question 2, all articles described in a previously
published meta-analysis evaluating tests for extrathoracic metas-
tases also were included.
4
Selection Criteria
Titles and abstracts, and the full text of all articles passing the
title-and-abstract screen, were evaluated independently by at
least two of the authors for inclusion or exclusion based on the
following five criteria: (1) publication in a peer-reviewed journal;
(2) study size of 20 patients (except for studies involving CT
scan evaluation of the mediastinum, for which 50 patients were
required); (3) patient group not included in a subsequent update
of the study; (4) histologic or cytologic confirmation of medias-
tinal nodes or extrathoracic sites in addition to the primary tumor;
and (5) availability of the raw data needed to calculate indepen-
dently the sensitivity, specificity, NPV, and positive predictive
value (PPV) of CT scanning, PET scanning, MRI, or EUS
(question 1), or, for question 2, the raw data needed to calculate
the NPV of the clinical evaluation. This last criterion was most
important when analyzing studies that did not originally differ-
entiate between hilar nodal involvement and mediastinal nodal
involvement. Only articles meeting all five of these criteria were
included.
Data Abstraction
Data were abstracted from included studies and tabulated
separately by type of noninvasive test (ie, CT scan, PET scan,
MRI, and EUS). Studies that evaluated the mediastinum by CT
or PET scan were tabulated both by correlation to nodal station
and by correlation to patient. Some studies described both hilar
nodal involvement (stage N1) and mediastinal nodal involvement
(stages N2 and N3). We considered only mediastinal nodal
involvement (stages N2 and N3) as being disease-positive in
analyzing the studies, which is consistent with the 1997 revision
of the lung cancer staging system.
1
Patients with clinical or
histopathologic stage IV disease were excluded from calculations
of operating characteristics if the nodal stage was not described.
Either tissue histologic confirmation or, if that was unavailable,
long-term (ie, 1 year) clinical outcome was utilized as the
reference or gold standard by which imaging data were com-
pared. If neither tissue pathologic confirmation nor clinical
outcome was available, then the patient was deemed to be not
evaluable and was excluded from further statistical analysis.
Data from studies that described the presence or absence of
symptoms or signs of metastatic disease on clinical evaluation (ie,
symptoms, physical examination findings, and laboratory evalua-
tion results) were abstracted, with the presence or absence of
extrathoracic metastases tabulated separately by the site of
metastatic disease (ie, brain, abdomen, and bone). Positive
findings on neuroimaging studies (ie, CT scan, MRI, or PET
scan), abdominal CT scans, and radionuclide bone scans were
used as references by which clinical evaluations for brain,
abdominal, and bone metastases, respectively, were compared.
Statistical Analysis
Sensitivity is defined as the percentage of people with the
disease who are detected by the test. It is calculated as the
number of true-positive results divided by the sum of the number
of true-positive results and false-negative results. Specificity is
defined as the percentage of people without the disease who
were correctly labeled by the test as not having the disease. It is
calculated as the number of true-negative results divided by the
sum of true-negative results and false-positive results. The PPV is
defined as the likelihood that a patient with a positive test result
actually has the disease. It is calculated as the number of
true-positive results divided by the sum of true-positive results
and false-positive results. The NPV is defined as the likelihood
that a patient with a negative test result actually does not have the
disease. It is calculated as the number or true-negative results
divided by the sum of true-negative results and false-negative
results. Both the PPV and the NPV vary with the prevalence of
disease, which is the frequency of disease in the population. The
prevalence can be calculated as the number of patients with
either true-positive or false-negative results divided by the total
number of patients.
Summary sensitivity and specificity, and their respective con-
fidence intervals (CIs), were calculated using statistical software
for the meta-analysis of diagnosis tests (Meta-Test; New England
Cochrane Center; Boston, MA).
5
For studies that included
patients with a positive and negative clinical evaluation, the
sensitivity, specificity, and PPV of the evaluation for metastatic
disease were calculated. As most studies of clinical evaluation
included only asymptomatic patients, the primary outcome in
these studies was the NPV.
We calculated NPVs using statistical software (FAST*PRO),
6
which tests individual study estimates for homogeneity and
calculates a summary estimate of the NPV. Summary receiver
operating characteristic (ROC) curves were generated for studies
that provided information about patients with both positive and
negative clinical evaluation results and for studies with tissue
confirmation of disease.
7
As with conventional ROC curves, a
summary ROC curve closer to the upper left-hand corner of the
graph indicates better overall diagnostic test performance. To
compare the summary ROC curves, a t test comparing the
intercepts was performed using a Student t test.
Results
Although some studies analyzed results by nodal
station, which may increase the correlation between
imaging abnormalities and pathology, clinical deci-
sions are made based on the accurate staging of
multiple lymph-node stations in the mediastinum.
We therefore present results at the patient level, not
at the nodal-station level.
Mediastinal Staging by CT Scan
Twenty studies,
827
with 3,438 evaluable patients,
evaluated the accuracy of standard CT scanning for
staging the mediastinum (Table 1). All but three
studies
16,17,27
used a 10-mm short-axis diameter as
the criterion for nodal positivity. The individual
studies are plotted by sensitivity and 1 specificity
in Figure 1, which also displays the summary ROC
curve. The pooled sensitivity of CT scanning was
0.57 (95% CI, 0.49 to 0.66), and the pooled speci-
ficity was 0.82 (95% CI, 0.77 to 0.86). However,
there was marked heterogeneity in the sensitivity
and specificity of the individual studies. The overall
PPV and NPV of CT scanning for a patient were 0.56
(range, 0.26 to 0.84) and 0.83 (range, 0.63 to 0.93),
respectively. The overall prevalence of mediastinal
disease among the patients was 28% (range, 18 to
50%).
Mediastinal Staging by PET Scan
Staging the mediastinum by PET scanning was
assessed by 18 studies
8,11,13,14,1618,21,2837
that in-
cluded 1,045 evaluable patients from total of 1,166
enrolled patients (Table 2). In all studies, the imag-
ing technique used
18
F-2-fluoro-2-deoxy-D-glucose
Table 1Accuracy of CT Scanning for Staging the Mediastinum in Lung Cancer Patients*
Study/Year Patients, No. CT Scan Technique Sensitivity Specificity PPV NPV Prevalence
Dunagan et al
8
/2001 72 Contrast 0.50 0.87 0.56 0.84 0.25
Kamiyoshihara et al
9
/2001 546 Contrast 0.33 0.90 0.46 0.84 0.20
Osada et al
10
/2001 335 Contrast 0.56 0.93 0.77 0.83 0.30
Pieterman et al
11
/2000 102 Contrast 0.75 0.66 0.50 0.85 0.31
Takamochi et al
12
/2000 401 Contrast 0.30 0.82 0.30 0.83 0.20
Marom et al
13
/1999 79 Contrast 0.59 0.86 0.84 0.63 0.56
Saunders et al
14
/1999 84 NR 0.20 0.90 0.30 0.84 0.18
Suzuki et al
15
/1999 440 Contrast 0.33 0.92 0.56 0.82 0.23
Vansteenkiste et al
16
/1998 68 Contrast 0.75 0.63 0.58 0.78 0.41
Vansteenkiste et al
17
/1998 56 Contrast 0.86 0.79 0.80 0.85 0.50
Bury et al
18
/1997 64 Contrast 0.79 0.84 0.58 0.93 0.22
Gdeedo et al
19
/1997 100 Contrast 0.63 0.57 0.41 0.76 0.32
Buccheri et al
20
/1996 80 Contrast 0.64 0.74 0.48 0.84 0.28
Bury et al
21
/1996 53 Contrast 0.71 0.81 0.63 0.85 0.32
Aaby et al
22
/1995 57 NR 0.72 0.91 0.86 0.81 0.44
Primack et al
23
/1994 159 Contrast 0.63 0.86 0.73 0.79 0.38
Yokoi et al
24
/1994 113 Contrast 0.62 0.80 0.61 0.81 0.33
McLoud et al
25
/1992 143 Contrast 0.64 0.62 0.44 0.79 0.31
Jolly et al
26
/1991 336 Contrast 0.71 0.86 0.69 0.87 0.30
Cole et al
27
/1993 150 NR 0.26 0.81 0.26 0.81 0.21
Summary 3,438 0.57 0.82 0.56 0.83 0.28
*NR not reported.
as the positron emitter. The individual studies are
plotted by sensitivity and 1 specificity in Figure 2,
which also displays the summary ROC curve. As
compared with the CT scanning summary ROC
curve, the PET scanning summary ROC curve was
significantly more accurate (t 19; p 0.001). The
pooled sensitivity was 0.84 (95% CI, 0.78 to 0.89),
and the pooled specificity was 0.89 (95% CI, 0.83 to
0.93). The heterogeneity was less than that found
among the studies of CT scanning. The overall PPV
and NPV were 0.79 (range, 0.40 to 1.00) and 0.93
(range, 0.75 to 1.00), respectively. The overall prev-
alence of mediastinal disease was 32% (range, 5
to 56%).
Mediastinal Staging by CT and PET Scans
Combined
Three studies,
16,17,31
including 152 patients total,
evaluated the mediastinum with both CT and PET
scanning. The sensitivity of the combined tests
ranged from 0.78 to 0.93, and the specificity ranged
from 0.82 to 0.95. The PPV ranged from 83 to 93%,
and the NPV ranged from 88 to 95%. The prevalence
of mediastinal disease ranged from 32 to 50%.
Mediastinal Staging by MRI Scan
One study
38
of 20 patients assessed the accuracy of
MRI for staging the mediastinum with and without
gadolinium enhancement. With gadolinium, the sen-
sitivity was 1.0, the specificity was 0.91, the PPV was
0.96, and the NPV was 1.0. Without gadolinium, the
sensitivity dropped to 0.63, the specificity improved
to 1.0, the PPV improved to 1.0, and the NPV fell
to 0.55.
Table 2Accuracy of PET Scanning for Staging the Mediastinum in Lung Cancer Patients
Study/Year Patients, No. Sensitivity Specificity PPV NPV Prevalence
Dunagan et al
8
/2001 81 0.52 0.88 0.61 0.84 0.26
Farrell et al
28
/2000 84 1.00 0.93 0.40 1.00 0.05
Liewold et al
29
/2000 78 0.93 0.78 0.69 0.95 0.35
Pieterman et al
11
/2000 102 0.91 0.86 0.74 0.95 0.31
Roberts et al
30
/2000 100 0.88 0.91 0.75 0.96 0.24
Magnani et al
31
/1999 28 0.67 0.84 0.67 0.84 0.32
Marom et al
13
/1999 79 0.73 0.94 0.85 0.88 0.56
Saunders et al
14
/1999 84 0.71 0.97 0.86 0.93 0.20
Vansteenkiste et al
16
/1998 68 0.93 0.95 0.93 0.95 0.41
Vansteenkiste et al
17
/1998 56 0.86 0.43 0.60 0.75 0.50
Bury et al
18
/1997 64 0.86 1.0 1.0 0.96 0.22
Guhlmann et al
32
/1997 32 0.87 1.0 1.0 0.89 0.47
Steinert et al
33
/1997 47 0.92 0.97 0.92 0.97 0.28
Bury et al
21
/1996 30 0.88 0.86 0.88 0.86 0.53
Sazon et al
34
/1996 32 1.00 1.00 1.00 1.00 0.50
Scott et al
35
/1996 27 1.00 1.00 1.00 1.00 0.33
Chin et al
36
/1995 30 0.78 0.81 0.64 0.89 0.30
Wahl et al
37
/1994 23 0.82 0.75 0.75 0.82 0.48
Summary 1,045 0.84 0.89 0.79 0.93 0.32
Figure 1. Summary ROC curve for imaging mediastinal lymph
nodes that are 1 cm diameter with standard CT scanning. The
gray ellipses indicate individual study estimates of sensitivity and
1 specificity. A study showing the highest accuracy will appear
in the top left corner of the graph. The size of the gray ellipse is
proportional to the statistical weight of the study in the analysis
(roughly proportional to its size). The X indicates the summary
point estimates for sensitivity and 1 specificity from meta-
analyses of sensitivity and specificity, independent of one an-
other. The outlined box indicates 95% CIs about the summary
sensitivity and 1 specificity estimates. The dark line represents
the summary ROC curve.
Mediastinal Staging by EUS
Five studies
3943
assessed the use of EUS to stage
the mediastinum in 163 evaluable patients (Table 3).
The size of the lymph node and evidence of necrosis
(ie, echo heterogeneity) or capsular disruption (ie,
irregularity of shape) provided criteria with which to
determine the presence of metastases. The pooled
sensitivity of EUS was 0.78 (95% CI, 0.61 to 0.89),
and the pooled specificity of EUS was 0.71 (95% CI,
0.56 to 0.82). The overall PPV was 75% (range, 38 to
100%), and the overall NPV was 79% (range, 57 to
89%). The overall prevalence of mediastinal involve-
ment in these patients was 50% (range, 25 to 76%).
Detection of Brain Metastases by Clinical
Evaluation
Seventeen studies
10,4459
were included that evalu-
ated the ability of clinical evaluation to detect brain
metastases in 1,784 patients (Table 4). Eight stud-
ies
46,48,5156
provided information on patients with both
positive and negative clinical evaluations, while nine
studies
10,44,45,47,49,50,5759
included only patients with a
negative clinical evaluation findings. One study
44
used
both CT scanning and MRI as the reference standard
in two separate analyses. Only the results from the CT
scan analysis are considered here.
The NPV of the clinical neurologic evaluation ranged
from0.79 to 1.00, but there was significant heterogeneity.
The calculated summary estimate for the NPV was 0.94
(95%CI, 0.91 to 0.96). For studies that evaluated patients
with both positive and negative clinical evaluations
(n 8), the pooled sensitivity was 0.76 (95% CI, 0.64 to
0.84), and the pooled specificity was 0.87 (95%CI, 0.74 to
0.94). The PPVwas 0.54, with a range of 0.21 to 1.00. The
overall prevalence of brain metastases in these patients
was 13% (range, 0 to 32%).
Detection of Abdominal Metastases by Clinical
Evaluation
Twelve studies
4850,6068
assessed the ability of the
clinical evaluation to detect adrenal and/or liver
metastases in 1,201 patients (Table 5). Three of the
studies
60,61,64
evaluated both symptomatic (including
abnormal liver function tests as the only symptom) and
asymptomatic patients, while nine studies
4850,62,63,6568
included only negative clinical evaluation patients.
There was heterogeneity in the NPV, which ranged
from 0.82 to 1.0. The calculated summary estimate was
0.95 (95% CI, 0.93 to 0.96). For studies that evaluated
patients who had a negative and a positive clinical
evaluation (n 3), the pooled sensitivity and specificity
of clinical abdominal evaluation for detecting abdomi-
nal metastases were 0.92 (95% CI, 0.82 to 0.97) and
0.49 (95% CI, 0.25 to 0.74), respectively. The PPV
ranged from 0.20 to 0.59, with a mean of 0.32. The
overall prevalence of abdominal metastases was 10%
(range, 0 to 40%).
Detection of Bone Metastases by Clinical
Evaluation
Seven studies
48,50,6973
assessed the ability of clinical
evaluation to detect bone metastases in 633 patients
(Table 6). All but two studies
50,71
included both symp-
Figure 2. Summary ROC curve for imaging mediastinal lymph
nodes that are 1 cm diameter with PET using
18
F-2-fluoro-2-
deoxy-D-glucose. The gray ellipses indicate individual study
estimates of sensitivity and 1 specificity. The size of the gray
ellipse is proportional to the weight of the study in the analysis.
The X indicates the summary point estimates for sensitivity and
1 specificity from meta-analyses of sensitivity and specificity,
independent of one another. The outlined box indicates 95% CIs
about the summary sensitivity and 1 specificity estimates. The
dark line represents the summary ROC curve.
Table 3Accuracy of EUS Without Biopsy for Staging the Mediastinum in Lung Cancer Patients
Study/Year Patients, No. Sensitivity Specificity PPV NPV Prevalence
Wiersema et al
39
/2001 29 0.86 0.57 0.86 0.57 0.76
Gress et al
40
/1997 45 0.86 0.83 0.83 0.86 0.49
Potepan et al
41
/1996 30 0.45 0.58 0.38 0.65 0.37
Silvestri et al
42
/1996 27 0.89 1.00 1.00 0.82 0.67
Schuder et al
43
/1991 32 0.75 0.71 0.46 0.89 0.25
Summary 163 0.78 0.71 0.75 0.79 0.50
tomatic and asymptomatic patients. The pooled sensi-
tivity and specificity of clinical bone evaluation for
detecting bone metastases were 0.87 (95% CI, 0.79 to
0.93) and 0.67 (95% CI, 0.40 to 0.88), respectively.
There was heterogeneity in the NPV, with a calculated
summary estimate of 0.90 (95% CI, 0.86 to 0.93) and a
range of 0.70 to 1.0. The mean PPV was 0.36 (range,
0.16 to 0.90). The overall prevalence of bone metasta-
ses was 20% (range, 8 to 34%).
Discussion
Patients presenting with newly diagnosed lung
cancer need to be assessed for potential mediastinal
Table 5Utility of the Clinical Evaluation in Detecting Abdominal Metastases Using CT Scanning
as the Reference Standard*
Study/Year Organ Scanned Patients, No. Routine Scan? Sensitivity Specificity PPV NPV Prevalence
Miralles et al
60
/1993 Liver 71 No 0.94 0.65 0.44 0.97 0.23
Silvestri et al
61
/1992 Adrenal 173 No 1.00 0.27 0.20 1.00 0.15
Ettinghausen and Burt
62
/1991 Adrenal 246 NR 0.98 0.02
Salvatierra et al
48
/1990 Adrenal 146 Yes 0.92 0.08
Grant et al
49
/1988 Liver, adrenal 114 Yes 0.92 0.08
Whittlesey
63
/1988 Adrenal 180 Yes 0.97 0.03
Mirvis et al
64
/1987 Liver, adrenal 72 Yes 0.90 0.58 0.59 0.89 0.40
Osada et al
50
/1987 Liver, adrenal 47 No 1.00 0.00
Heavey et al
65
/1986 Adrenal 31 Yes, stage 1 0.97 0.03
Pearlberg et al
66
/1985 Liver, adrenal 23 Probably no 1.00 0.00
Chapman et al
67
/1984 Adrenal 14 Yes 0.86 0.14
Nielsen et al
68
/1982 Adrenal 84 Yes 0.82 0.18
Summary 1,201 0.92 0.49 0.32 0.95 0.10
*See Table 1 for abbreviation not used in the text.
Not included by Silvestri et al.
4
PPV could not be estimated because study evaluated with CT scanning only those patients in whom the clinical examination findings were
negative.
Table 4Utility of the Clinical Evaluation in Detecting Brain Metastases Using Neuroimaging (CT Scanning, MRI,
and PET Scanning) as the Reference Standard*
Study/Year Examination Patients, No. Routine Scan? Sensitivity Specificity PPV NPV Prevalence
Osada et al
10
/2001 Neuro 91 cT1T2, N2 0.98 0.02
Yokoi et al
44
/1999 Neuro 155 Yes; CT 0.99 0.01
Cole et al
45
/1994 Neuro 42 No 1.00 0.00
Habets et al
46
/1992 Neuro 54 Yes 1.00 0.98 0.75 1.00 0.06
Kormas et al
47
/1992 Screening 157 N2 only 0.97 0.03
Salvatierra et al
48
/1990 Expanded 146 Adenocarcinoma
and large cell
only
0.79 0.91 0.58 0.97 0.13
Grant et al
49
/1988 Screening 114 Yes 0.91 0.09
Osada et al
50
/1987 Screening 56 No 1.00 0.00
Crane et al
51
/1984 Neuro 145 Yes 0.65 0.98 0.88 0.94 0.16
Hooper et al
52
/1984 Expanded 89 No 1.00 0.38 0.26 1.00 0.18
Levitan et al
53
/1984 Neuro 55 Yes 0.73 1.00 1.00 0.91 0.27
Mintz et al
54
/1984 Neuro 66 Yes 0.38 0.81 0.21 0.90 0.12
Tarver et al
55
/1984 Neuro 323 Adenocarcinoma
and SCLC
only
0.83 0.78 0.64 0.91 0.32
Johnson et al
56
/1983 Neuro 84 No 0.83 0.81 0.42 0.97 0.14
Jennings et al
57
/1980 Screening 102 NR 0.79 0.21
Butler et al
58
/1979 Screening 55 Yes 0.95 0.05
Jacobs et al
59
/1977 Screening 50 Yes 0.94 0.06
Summary 1,784 0.76 0.87 0.54 0.94 0.13
*Neuro neurological; NR not reported.
Not included by Silvestri et al.
4
PPV could not be estimated because study evaluated with neuroimaging only those patients in whom the clinical examination findings were
negative.
and extrathoracic metastases. When selecting a test
for staging the disease of a patient with known or
suspected lung cancer, two issues need to be consid-
ered. First, one needs to select a test that will assess
the patient for metastatic disease. This test should
have a high sensitivity and specificity. Second, one
needs to be able to interpret accurately the test
results for an individual patient. For example, what is
the likelihood that a negative test result means a
patient is free of metastatic disease or, conversely,
that a positive test confirms metastatic disease? This
issue is addressed by the PPVs and NPVs of a test.
However, the NPV and PPV of a test are affected by
the prevalence of disease. Thus, patients may be
approached differently depending on ones initial
degree of suspicion regarding mediastinal or ex-
trathoracic involvement. A patient suspected of hav-
ing metastases would require a test with high sensi-
tivity to confirm the diagnosis. In contrast, a patient
suspected of having early-stage disease would re-
quire a test with high specificity to rule out advanced
disease. Determining the most expeditious and cost-
effective method of doing this has been the focus of
considerable research.
4,7478
The present meta-analysis suggests that PET scan-
ning of the mediastinum is more sensitive and
specific than either CT scanning or EUS alone. A
comparison of the summary ROC curves demon-
strates greater accuracy for PET scanning than for
CT scanning, with a negative PET scan providing a
90% certainty of the absence of positive medias-
tinal lymph node metastases. These findings are
compatible with those of Dwamena et al,
77
who also
demonstrated the superiority of PET scanning to CT
scanning for detecting mediastinal nodal metastases.
Other studies
11
have demonstrated improved sensi-
tivity and specificity for PET scanning vs CT scan-
ning and have shown that 60% of patients have a
resultant change in their clinical stage when using
PET scanning vs CT scanning for staging. Addition-
ally, a randomized trial comparing a conventional
workup (ie, invasive diagnostic and therapeutic pro-
cedures) alone vs PET scanning followed by a con-
ventional workup identified more patients in the
PET scan group with mediastinal and distant metas-
tases, and subsequently fewer patients who under-
went futile thoracotomies.
79
EUS without biopsy is
limited in that it cannot image all nodal stations, and
its sensitivity and specificity are lower than those for
PET scanning
Our meta-analysis of CT scanning demonstrated
that the accuracy of CT scanning for mediastinal
staging has not improved over the past decade,
despite improvements in CT scan resolution. The
sensitivity and specificity of CT scanning in our
analysis were not significantly better than those in a
meta-analysis of studies performed between 1980
and 1988,
80
with sensitivities of 57% vs 79%, respec-
tively, and specificities of 82% vs 78%, respectively.
Although only three studies,
16,17,31
which had
small sample sizes, have evaluated the performance
characteristics of combined CT and PET scanning
for mediastinal staging, the combination of the two
imaging tests may achieve greater staging accuracy
than either test alone. A cost-effectiveness analysis
utilizing PET scanning for all patients who had
node-negative CT scan results was encouraging,
demonstrating that the cost of PET scanning was
nearly compensated for by the more appropriate
selection of patients for beneficial surgery.
78
Future
studies should include the evaluation of a combina-
tion of CT and PET scanning in prospective clinical
trials. Furthermore, the ongoing development of
fusion scanners that combine CT and PET warrants
clinical trials and may provide an optimal method for
noninvasive clinical staging.
The present meta-analysis also updates the meta-
analysis performed by Silvestri et al
4
As with his
study, we found the calculated summary estimates
for the NPV to be 90% for the clinical evaluation
of asymptomatic patients for brain, abdominal, or
bone metastases. These findings are consistent with
Table 6Utility of the Clinical Evaluation in Detecting Bone Metastases Using Radionuclide Bone Scanning as the
Reference Standard
Study/Year Patients, No. Histology Routine Scan? Sensitivity Specificity PPV NPV Prevalence
Michel et al
69
/1991 110 NSCLC No 1.00 0.54 0.16 1.00 0.08
Tornyos et al
70
/1991 50 NSCLC Yes 0.88 0.30 0.39 0.83 0.34
Salvatierra et al
48
/1990 146 NSCLC No 0.79 0.88 0.50 0.97 0.13
Osada et al
50
/1987 66 NSCLC Yes * 0.70 0.30
Turner and Haggith
71
/1981 55 NSCLC/SCLC No * 0.84 0.16
Hooper et al
72
/1978 155 NSCLC/SCLC No 0.90 0.40 0.36 0.92 0.27
Ramsdell et al
73
/1977 51 NSCLC No 0.90 0.98 0.90 0.98 0.20
Summary 633 0.87 0.67 0.36 0.90 0.20
*PPV could not be estimated because study evaluated with bone scanning only those patients in whom the clinical examination findings were
negative.
the findings of a retrospective analysis
75
of 755
patients with clinical stage T1-2N0, which found only
five sites of silent metastasis after extensive imaging
for extrathoracic disease in all presenting patients.
However, within the studies combined in our meta-
analyses, there was considerable heterogeneity with
respect to their individual NPV. The greatest range
was found for the evaluation for bone metastases, in
which the NPVs ranged from 0.7 to 1.0. Some of the
heterogeneity may be due to the imperfect reference
standard, in which abnormal bone scan findings were
not followed up with histologic confirmation of
disease. For example, in the study in which 30% of
patients had a false-negative clinical evaluation,
50
only 4 of 17 patients with abnormal bone scan
findings received confirmations of lung cancer that
was metastatic to bone. There was also significant
variation in the stage and cell type of disease in the
patients who were evaluated in the included studies.
For the evaluation of abdominal metastases, the
NPVs also ranged broadly between 0.82 and 1.0. In
this case, (1) the studies did not control for stage
and/or size of the primary tumor, (2) the reference
standard was suboptimal, as abnormal findings on
CT scanning were not always confirmed with biopsy,
and (3) patient populations varied between studies.
For the evaluation of brain metastases, with the
exception of one very poorly designed study,
57
the
NPV, although heterogeneous, ranged more nar-
rowly between 0.91 and 1.0.
In contrast, although the PPV varied among the
studies included in this meta-analysis of the clinical
evaluation for detecting metastatic disease, the re-
sultant following conclusion remains the same: fur-
ther diagnostic testing is necessary to confirm the
presence of disease in patients with abnormal find-
ings on clinical evaluation.
A separate issue that was not evaluated by this
meta-analysis is the role that whole-body PET scan-
ning may play in the workup of asymptomatic pa-
tients for metastatic disease. A recent randomized
trial
79
suggests that the addition of PET scanning to
a conventional workup identified more patients with
distant metastases among those with suspected
NSCLC.
There are limitations to the present meta-analysis.
No attempt was made to obtain unpublished data or
non-English-language studies. An evaluation of per-
formance characteristics by cell type and/or stage
was not possible, but such data, if available, may be
useful for assessing the optimal staging criteria for
different tumor types.
In summary, the clinical evaluation of a newly
diagnosed lung cancer patient suggests that PET is
more accurate for detecting mediastinal metastases
than the standard assessment with CT chest scan-
ning. Future studies should be performed to deter-
mine the role of combined PET and CT scanning or
to delineate which patients should undergo PET
scanning after undergoing a CT scan. Future tech-
nologic advancements, such as the development of a
single-modality scanner that performs the role of
both a CT and PET scanner, may provide a more
accurate and cost-effective approach. However, this
would need to be validated in prospective clinical
trials.
This update of the meta-analysis performed by
Silvestri et al
4
suggests that the clinical evaluation of
patients who present without symptoms of meta-
static disease may obviate the need for brain, abdom-
inal, and bone imaging to detect distant metastases.
However, more definitive prospective studies that
clearly define the patient population, including cell
type and stage, that address the issue of patient
referral bias, and that provide improved reference
standards with histologic confirmation of positive
findings are necessary to more accurately assess the
true NPV of the clinical examination. Moreover, the
role of whole-body PET scanning for evaluating
mediastinal and extrathoracic metastases remains to
be studied.
References
2 Fleming ID, Cooper JS, Henson DE, et al. AJCC cancer
staging handbook. 5th ed. Philadelphia, PA: Lippincott Wil-
liams & Wilkins, 1998
3 Jemal A, Thomas A, Murray T, et al. Cancer statistics, 2002.
CA Cancer J Clin 2002; 52:2347
for detecting metastatic lung cancer: a meta-analysis. Am J
5 Lau J. Meta-test, version 0.6. Boston, MA: New England
Medical Center, 1997
6 Hasselblad V, Eddy DM. FAST*PRO: software for meta-
analysis by the confidence profile method, version 1.8. Bos-
ton, MA: Academic Press, 1992
7 Moses LE, Shapiro D, Littenberg B. Combining independent
studies of a diagnostic test into a summary ROC curve:
data-analytic approaches and some additional considerations.
Stat Med 1993; 12:12931316
8 Dunagan DP, Chin R, McCain TW, et al. Staging by positron
emission tomography predicts survival in patients with non-
small cell lung cancer. Chest 2001; 119:333339
9 Kamiyoshihara M, Kawashima O, Ishikawa S, et al. Medias-
tinal lymph node evaluation by computed tomographic scan
in lung cancer. J Cardiovasc Surg (Torino) 2001; 42:119124
10 Osada H, Kojima K, Tsukada H, et al. Cost-effectiveness
associated with the diagnosis and staging of non-small-cell
lung cancer. Jpn J Thorac Cardiovasc Surg 2001; 49:110
11 Pieterman RM, van Putten JW, Meuzelaar JJ, et al. Preoper-
ative staging of non-small-cell lung cancer with positron-
emission tomography. N Engl J Med 2000; 343:254261
12 Takamochi K, Nagai K, Yoshida J, et al. The role of computed
tomographic scanning in diagnosing mediastinal node in-
volvement in non-small cell lung cancer. Thorac Cardiovasc
Surg 2000; 119:11351140
13 Marom EM, McAdams HP, Erasmus JJ, et al. Staging
non-small cell lung cancer with whole-body PET. Radiology
1999; 212:803809
14 Saunders CA, Dussek JE, ODoherty MJ, et al. Evaluation of
fluorine-18-fluorodeoxyglucose whole body positron emission
tomography imaging in the staging of lung cancer. Ann
Thorac Surg 1999; 67:790797
15 Suzuki K, Nagai K, Yoshida J, et al. Clinical predictors of N2
disease in the setting of a negative computed tomographic
scan in patients with lung cancer. Thorac Cardiovasc Surg
1999; 117:593598
16 Vansteenkiste JF, Stroobants SG, De Leyn PR, et al. Lymph
node staging in non-small-cell lung cancer with FDG-PET
scan: a prospective study on 690 lymph node stations from 68
patients. J Clin Oncol 1998; 16:21422149
17 Vansteenkiste JF, Stroobants SG, Dupont PJ, et al. FDG-
PET scan in potentially operable non-small cell lung cancer:
do anatometabolic PET-CT fusion images improve the local-
isation of regional lymph node metastases?; the Leuven Lung
Cancer Group. Eur J Nucl Med 1998; 25:14951501
18 Bury T, Dowlati A, Paulus P, et al. Whole-body 18FDG
positron emission tomography in the staging of non-small cell
lung cancer. Eur Respir J 1997; 10:25292534
19 Gdeedo A, Van Schil P, Corthouts B, et al. Prospective
evaluation of computed tomography and mediastinoscopy in
mediastinal lymph node staging. Eur Respir J 1997; 10:1547
1551
20 Buccheri G, Biggi A, Ferrigno D, et al. Anti-CEA immu-
noscintigraphy and computed tomographic scanning in the
preoperative evaluation of mediastinal lymph nodes in lung
cancer. Thorax 1996; 51:359363
21 Bury T, Paulus P, Dowlati A, et al. Staging of the mediasti-
num: value of positron emission tomography imaging in
non-small cell lung cancer. Eur Respir J 1996; 9:25602564
22 Aaby C, Kristensen S, Nielsen SM. Mediastinal staging of
non-small-cell lung cancer: computed tomography and cervi-
cal mediastinoscopy. ORL J Otorhinolaryngol Relat Spec
1995; 57:279285
23 Primack SL, Lee KS, Logan PM, et al. Bronchogenic carci-
noma: utility of CT in the evaluation of patients with sus-
pected lesions. Radiology 1994; 193:795800
24 Yokoi K, Okuyama A, Mori K, et al. Mediastinal lymph node
metastasis from lung cancer: evaluation with Tl-201 SPECT;
comparison with CT. Radiology 1994; 192:813817
25 McLoud TC, Bourgouin PM, Greenberg RW, et al. Broncho-
genic carcinoma: analysis of staging in the mediastinum with
CT by correlative lymph node mapping and sampling. Radi-
ology 1992; 182:319323
26 Jolly PC, Hutchinson CH, Detterbeck F, et al. Routine
computed tomographic scans, selective mediastinoscopy, and
other factors in evaluation of lung cancer. Thorac Cardiovasc
Surg 1991; 102:266270
27 Cole PH, Roszkowski A, Firouz-Abadi A, et al. Computerised
tomography does not predict N2 disease in patients with lung
cancer. Aust N Z J Med 1993; 23:688691
28 Farrell MA, McAdams HP, Herndon JE, et al. Non-small cell
lung cancer: FDG PET for nodal staging in patients with
stage I disease. Radiology 2000; 215:886890
29 Liewald F, Grosse S, Storck M, et al. How useful is positron
emission tomography for lymphnode staging in non-small-cell
lung cancer? Thorac Cardiovasc Surg 2000; 48:9396
30 Roberts PF, Follette DM, von Haag D, et al. Factors
associated with false-positive staging of lung cancer by
positron emission tomography. Ann Thorac Surg 2000; 70:
11541159
31 Magnani P, Carretta A, Rizzo G, et al. FDG/PET, and spiral
CT image fusion for mediastinal lymph node assessment of
non-small cell lung cancer patients. J Cardiovasc Surg
(Torino) 1999; 40:741748
32 Guhlmann A, Storck M, Kotzerke J, et al. Lymph node
staging in non-small cell lung cancer: evaluation by
[18F]FDG positron emission tomography (PET). Thorax
1997; 52:438441
33 Steinert HC, Hauser M, Allemann F, et al. Non-small cell
lung cancer: nodal staging with FDG PET vs CT with
correlative lymph node mapping and sampling. Radiology
1997; 202:441446
34 Sazon DA, Santiago SM, Soo Hoo GW, et al. Fluorodeoxy-
glucose-positron emission tomography in the detection and
staging of lung cancer. Am J Respir Crit Care Med 1996;
153:417421
35 Scott WJ, Gobar LS, Terry JD, et al. Mediastinal lymph node
staging of non-small-cell lung cancer: a prospective compar-
ison of computed tomography and positron emission tomog-
raphy. Thorac Cardiovasc Surg 1996; 111:642648
36 Chin R Jr, Ward R, Keyes JW, et al. Mediastinal staging of
non-small-cell lung cancer with positron emission tomogra-
phy. Am J Respir Crit Care Med 1995; 152:20902096
37 Wahl RL, Quint LE, Greenough RL, et al. Staging of
mediastinal non-small cell lung cancer with FDG PET, CT,
and fusion images: preliminary prospective evaluation. Radi-
ology 1994; 191:371377
38 Crisci R, Di Cesare E, Lupattelli L, et al. MR study of N2
disease in lung cancer: contrast-enhanced method using
gadolinium-DTPA. Eur J Cardiothorac Surg 1997; 11:214
217
39 Wiersema MJ, Vazquez-Sequeiros E, Wiersema LM. Evalu-
ation of mediastinal lymphadenopathy with endoscopic US-
guided fine-needle aspiration biopsy. Radiology 2001; 219:
252257
40 Gress FG, Savides TJ, Sandler A, et al. Endoscopic ultra-
sonography, fine-needle aspiration biopsy guided by endo-
scopic ultrasonography, and computed tomography in the
preoperative staging of non-small-cell lung cancer: a compar-
ison study. Ann Intern Med 1997; 127:604612
41 Potepan P, Meroni E, Spagnoli I, et al. Non-small-cell lung
cancer: detection of mediastinal lymph node metastases by
endoscopic ultrasound and CT. Eur Radiol 1996; 6:1924
42 Silvestri GA, Hoffman BJ, Bhutani MS, et al. Endoscopic
ultrasound with fine-needle aspiration in the diagnosis and
staging of lung cancer. Ann Thorac Surg 1996; 61:14411445
43 Schuder G, Isringhaus H, Kubale B, et al. Endoscopic
ultrasonography of the mediastinum in the diagnosis of
bronchial carcinoma. Thorac Cardiovasc Surg 1991; 39:299
303
44 Yokoi K, Kamiya N, Matsuguma H, et al. Detection of brain
metastasis in potentially operable non-small cell lung cancer:
a comparison of CT and MRI. Chest 1999; 115:714719
45 Cole FH Jr, Thomas JE, Wilcox AB, et al. Cerebral imaging
in the asymptomatic preoperative bronchogenic carcinoma
patient: is it worthwhile? Ann Thorac Surg 1994; 57:838840
46 Habets JM, van Oosterhout AG, ten Velde GP, et al. Diag-
nostic value of CT in the detection of brain metastasis in small
cell lung cancer patients. J Belge Radiol 1992; 75:179181
47 Kormas P, Bradshaw JR, Jeyasingham K. Preoperative com-
puted tomography of the brain in non-small cell bronchogenic
carcinoma. Thorax 1992; 47:106108
48 Salvatierra A, Baamonde C, Llamas JM, et al. Extrathoracic
staging of bronchogenic carcinoma. Chest 1990; 97:1052
1058
49 Grant D, Edwards D, Goldstraw P. Computed tomography of
the brain, chest, and abdomen in the preoperative assessment
of non-small cell lung cancer. Thorax 1988; 43:883886
50 Osada H, Nakajima Y, Taira Y, et al. The role of mediastinal
and multi-organ CT scans in staging presumable surgical
candidates with non-small-cell lung cancer. Jpn J Surg 1987;
17:362368
51 Crane JM, Nelson MJ, Ihde DC, et al. A comparison of
computed tomography and radionuclide scanning for detec-
tion of brain metastases in small cell lung cancer. J Clin Oncol
1984; 2:10171024
52 Hooper RG, Tenholder MF, Underwood GH, et al. Com-
puted tomographic scanning of the brain in initial staging of
53 Levitan N, Hong WK, Byrne RE, et al. Role of computerized
cranial tomography in the staging of small cell carcinoma of
the lung. Cancer Treat Rep 1984; 68:13751377
54 Mintz BJ, Tuhrim S, Alexander S, et al. Intracranial metas-
tases in the initial staging of bronchogenic carcinoma. Chest
1984; 86:850853
55 Tarver RD, Richmond BD, Klatte EC. Cerebral metastases
from lung carcinoma: neurological and CT correlation: work
in progress. Radiology 1984; 153:689692
56 Johnson DH, Windham WW, Allen JH, et al. Limited value of
CT brain scans in the staging of small cell lung cancer. AJR
Am J Roentgenol 1983; 140:3740
57 Jennings EC, Aungst CW, Yatco R. Asymptomatic patients
with primary carcinoma; computerized axial tomography
study. N Y State J Med 1980; 80:10961098
58 Butler AR, Leo JS, Lin JP, et al. The value of routine cranial
computed tomography in neurologically intact patients with
primary carcinoma of the lung. Radiology 1979; 131:399401
59 Jacobs L, Kinkel WR, Vincent RG. Silent brain metastasis
from lung carcinoma determined by computerized tomogra-
phy. Arch Neurol 1977; 34:690693
60 Miralles TG, Gosabez F, de Lera J, et al. Percutaneous fine
needle aspiration biopsy cytology of the liver for staging small
cell lung carcinoma: comparison with other methods. Acta
Cytol 1993; 37:499502
102:17481751
62 Ettinghausen SE, Burt ME. Prospective evaluation of unilat-
eral adrenal masses in patients with operable non-small-cell
63 Whittlesey D. Prospective computed tomographic scanning
in the staging of bronchogenic cancer. Thorac Cardiovasc
Surg 1988; 95:876882
64 Mirvis SE, Whitley NO, Aisner J, et al. Abdominal CT in the
staging of small-cell carcinoma of the lung: incidence of
metastases and effect on prognosis. AJR Am J Roentgenol
1987; 148:845847
65 Heavey LR, Glazer GM, Gross BH, et al. The role of CT in
staging radiographic T1N0M0 lung cancer. AJR Am J Roent-
genol 1986; 146:285290
66 Pearlberg JL, Sandler MA, Beute GH, et al. T1N0M0
bronchogenic carcinoma: assessment by CT. Radiology 1985;
157:187190
67 Chapman GS, Kumar D, Redmond J 3rd, et al. Upper
abdominal computerized tomography scanning in staging
non-small cell lung carcinoma. Cancer 1984; 54:15411543
68 Nielsen ME, Heaston DK, Dunnick NR, et al. Preoperative
CT evaluation of adrenal glands in non-small cell broncho-
genic carcinoma. AJR Am J Roentgenol 1982; 139:317320
69 Michel F, Soler M, Imhof E, et al. Initial staging of non-small
cell lung cancer: value of routine radioisotope bone scanning.
Thorax 1991; 46:469473
70 Tornyos K, Garcia O, Karr B, et al. A correlation study of
bone scanning with clinical and laboratory findings in the
staging of nonsmall-cell lung cancer. Clin Nucl Med 1991;
16:107109
71 Turner P, Haggith JW. Preoperative radionuclide scanning in
bronchogenic carcinoma. Br J Dis Chest 1981; 75:291294
Respir Dis 1978; 118:279286
73 Ramsdell JW, Peters RM, Taylor AT Jr, et al. Multiorgan
scans for staging lung cancer: correlation with clinical evalu-
ation. Thorac Cardiovasc Surg 1977; 73:653659
74 Canadian Lung Oncology Group. Investigating extrathoracic
metastatic disease in patients with apparently operable lung
cancer. Ann Thorac Surg 2001; 71:425434
75 Tanaka K, Kubota K, Kodama T, et al. Extrathoracic staging
is not necessary for non-small-cell lung cancer with clinical
stage T1-2N0. Ann Thorac Surg 1999; 68:10391042
76 Hillers TK, Sauve MD, Guyatt GH. Analysis of published
studies on the detection of extrathoracic metastases in pa-
tients presumed to have operable non-small cell lung cancer.
Thorax 1994; 49:1419
77 Dwamena BA, Sonnad SS, Angobaldo JO, et al. Metastases
from non-small cell lung cancer: mediastinal staging in the
1990s; meta-analytic comparison of PET and CT. Radiology
1999; 213:530536
78 Dietlein M, Weber K, Gandjour A, et al. Cost-effectiveness of
FDG-PET for the management of potentially operable non-
small cell lung cancer: priority for a PET-based strategy after
nodal-negative CT results. Eur J Nucl Med 2000; 27:1598
1609
79 van Tinteren H, Hoekstra OS, Smit EF, et al. Effectiveness of
positron emission tomography in the preoperative assessment
of patients with suspected non-small-cell lung cancer: the
PLUS multicentre randomised trial. Lancet 2002; 359:1388
1393
80 Dales RE, Stark RM, Raman S. Computed tomography to
stage lung cancer: approaching a controversy using meta-
analysis. Am Rev Respir Dis 1990; 141:10961101
2003;123;137-146 Chest
Eric M. Toloza, Linda Harpole and Douglas C. McCrory
Evidence
Noninvasive Staging of Non-small Cell Lung Cancer: A Review of the Current
& Services
Updated Information
S
References
S#BIBL
free at:
Citations
S#otherarticles
Reprints
2003;123;147-156 Chest
Detterbeck
Gerard A. Silvestri, Lynn T. Tanoue, Mitchell L. Margolis, John Barker and Frank
The Noninvasive Staging of Non-small Cell Lung Cancer: The Guidelines
ISSN: 0012-3692.
The Noninvasive Staging of Non-small
Cell Lung Cancer*
The Guidelines
Gerard A. Silvestri, MD, FCCP; Lynn T. Tanoue, MD;
Mitchell L. Margolis, MD, FCCP; John Barker, MD; and
Frank Detterbeck, MD, FCCP
Correctly staging lung cancer is extremely important because the treatment options and the
prognosis differ significantly by stage. Several noninvasive imaging studies are available to aid in
identifying disease both within and outside of the chest. Chest CT scanning is useful in providing
anatomic detail that better identifies the location of the tumor, its proximity to local structures,
and whether or not lymph nodes in the mediastinum are enlarged. Unfortunately, the accuracy
of chest CT scanning in differentiating benign from malignant lymph nodes in the mediastinum
is unacceptably low. Whole-body positron emission tomography (PET) scanning provides func-
tional information on tissue activity and has much better sensitivity and specificity than chest CT
scanning for staging lung cancer in the mediastinum. In addition, metastatic disease can be
detected by PET scan. Still, positive findings of PET scans can occur from nonmalignant
etiologies (eg, infections), so that tissue sampling to confirm the suspected malignancy must be
performed. The clinical evaluation tool, which is composed of a thorough history and physical
examination, remains the best predictor of metastatic disease. If the findings from the clinical
evaluation are negative, then imaging studies such as a CT scan of the head, a bone scan, or an
abdominal CT scan are unnecessary, and the search for metastatic disease is complete. If signs,
symptoms, or findings from the physical examination suggest the presence of malignancy, then
sequential imaging, starting with the most appropriate study based on the clues obtained by the
clinical evaluation, should be performed. Abnormalities detected by all of the aforementioned
imaging studies are not always cancer. Unless overwhelming evidence of metastatic disease is
present on an imaging study, in situations in which it will make a difference in treatment, all
abnormal scan findings require tissue confirmation of malignancy so that patients are not
precluded from having potentially curative surgery. (CHEST 2003; 123:147S156S)
Key words: CT scan; lung cancer; mediastinum; metastases; noninvasive; positron emission tomography; staging
Abbreviations: CI confidence interval; FDG
18
F-fluoro-deoxy-D-glucose; NSCLC non-small cell lung cancer;
PET positron emission tomography; SPECT single-photon emission CT
A
fter a tissue diagnosis of lung cancer has been
established or in patients in whom the clinical
suspicion is high and surgery is the recommended
next step, consideration must turn to the determina-
tion of the extent of disease, or stage, because this
will impact directly on the management of the
disease and the patients prognosis. The most signif-
icant dividing line is between those patients who are
candidates for surgical resection and those who are
inoperable but will benefit from chemotherapy, ra-
diation therapy, or both. Staging with regard to a
patients potential for surgical resection is most
applicable to non-small cell lung cancer (NSCLC),
whereas for small cell lung cancer a more simplified
staging classification of limited and extensive disease
is employed. Except in rare cases of surgically oper-
able, limited-stage, small cell cancer, the implication
of staging on the management of small cell lung
cancer is between chemotherapy and radiation for
limited-stage disease and chemotherapy alone for
extensive disease.
1
The basis for staging NSCLC is the TNM sys-
tem.
2,3
(See Table 1 for TNM descriptors and Figure
1 for stage grouping.
3a
) From a practical standpoint,
the involvement of disease in the mediastinum,
*From the Medical University of South Carolina (Drs. Silvestri
and Barker), Charleston, SC; Yale University School of Medicine
(Dr. Tanoue), New Haven, CT; University of Pennsylvania (Dr.
Margolis), Philadelphia, PA; and University of North Carolina
(Dr. Detterbeck), Chapel Hill, NC.
Correspondence to: Gerard A. Silvestri MD, FCCP, Associate
Professor of Medicine, Medical University of South Carolina,
Charleston, SC; e-mail: silvestri@musc.edu
which is reflected in the N designator in the system,
most often determines the appropriateness of the
patient for surgical resection.
Patients with stage IA, IB, IIA, and IIB disease
can benefit from surgical resection. Patients with
stage IIIA, IIIB, and IV almost never meet the
criteria for surgery. There is currently intense study
of surgery for selected patients with stage IIIA
disease in conjunction with neoadjuvant chemother-
apy and radiotherapy.
Staging can be used to predict survival and to
guide the patient toward the most appropriate treat-
ment regimen or clinical trial. Even with clinical
stage I, surgically resectable, potentially curable
disease, the 5-year survival postsurgery is only 50%.
In approximately 60% of patients, cancer recurrence
is presumably from extrathoracic micrometastatic
involvement at presentation, which is not currently
detectable with existing diagnostic modalities. Clin-
ical stage II patients (ie, T1N1M0 or T2N1M0) have
a 5-year survival rate after surgery of 30%. At clinical
stage IIIA, the 5-year survival rate is 17%, and at
clinical stage IIIB it is only 5%.
2
These patients are
generally treated with combined chemotherapy and
radiotherapy. The 5-year survivorship for stage IV
disease is virtually nil, and this disease is treated
either with chemotherapy and supportive care or
with supportive care alone. Thus, one can see that it
is critical to stage patients accurately as the treat-
ment modalities and subsequent patient outcomes
vary widely based on stage designation.
Noninvasive Staging of the Mediastinum
Imaging Modalities
Chest Radiograph: The majority of lung cancers
are initially detected by plain chest radiographs.
In some situations, the plain radiograph may be suf-
ficient to detect spread of the disease to the med-
iastinum. For example, the presence of bulky
lymphadenopathy in the superior or contralateral
mediastinal areas may be considered to be adequate
evidence of metastatic disease to preclude further
imaging evaluation of the chest. This may be partic-
ularly true if the patient is too ill or is unwilling to
undergo treatment of any kind. It is recommended,
however, that tissue confirmation be obtained by the
least invasive method possible. A CT scan of the
chest should be performed in nearly all cases of lung
cancer unless the patient is so debilitated that no
further evaluation or treatment is planned. It is
widely accepted that the chest radiograph is in
general an insensitive measure of mediastinal lymph
Table 1TNM Descriptors
Site Descriptor Comment
Primary tumor (T) TX Primary tumor cannot be assessed or tumor proven by the presence of malignant cells in sputum
or bronchial washings but not visualized by imaging or bronchoscopy
T0 No evidence of primary tumor
Tis Carcinoma in situ
T1 Tumor 3 cm in greatest dimension surrounded by lung or visceral pleura without
bronchoscopic evidence of invasion more proximal than the lobar bronchus (ie, not in the
main bronchus)
T2 Tumor with any of the following features of size or extent: 3 cm in greatest dimension;
involves main bronchus; 2 cm distal to the carina; invades the visceral pleura; associated
with atelectasis or obstructive pneumonitis that extends to the hilar region but does not
involve the entire lung
T3 Tumor of any size that directly invades any of the following: chest wall (including superior sulcus
tumors), diaphragm, mediastinal pleura, parietal pericardium, or tumor in the main bronchus
2 cm distal to the carina but without involvement of the carina or associated atelectasis or
obstructive pneumonitis of the entire lung
T4 Tumor of any size that invades any of the following: mediastinum, heart, great vessels, trachea,
esophagus, vertebral body, carina, or tumor with a malignant pleural or pericardial effusion or
with satellite tumor nodules within the primary-tumor lobe of the lung
Regional lymph nodes (N) NX Regional lymph nodes cannot be assessed
N0 No regional lymph node metastasis
N1 Metastasis to ipsilateral peribronchial and/or ipsilateral hilar lymph nodes and intrapulmonary
nodes involved by direct extension of the primary tumor
N2 Metastasis to ipsilateral mediastinal and/or subcarinal lymph nodes
N3 Metastasis to contralateral mediastinal, contralateral, hilar ipsilateral, or contralateral scalene or
supraclavicular lymph nodes
Distant metastasis (M) MX Presence of distant metastasis cannot be assessed
M0 No distant metastasis
M1 Distant metastasis present
node involvement with lung cancer, and thus further
noninvasive and/or invasive assessment is usually
necessary.
CT Scan of the Chest: In the 1990s, numerous
evaluations of CT scanning were performed that
compared clinical staging by CT scan to the gold
standard of mediastinoscopy or surgery. These stud-
ies demonstrated that, regardless of threshold size,
CT scan findings in isolation could not be considered
as conclusive evidence that lymph nodes were ma-
lignant. In other words, in all studies there are
meaningful numbers of false-positive results that are
detected by CT scan.
A CT scan of the chest is the most widely available
and most commonly used noninvasive modality for
evaluation of the mediastinum in patients with lung
cancer. The vast majority of reports evaluating the
accuracy of CT scanning for mediastinal lymph node
staging have employed the administration of IV
contrast material. IV contrast is not absolutely nec-
essary in performing chest CT scans for this indica-
tion but is very useful in helping to distinguish
vascular structures from lymph nodes as well as in
delineating mediastinal invasion by centrally located
tumors.
Various CT criteria have been used to define
malignant involvement of mediastinal lymph nodes.
The most widely used criterion is a short-axis lymph
node diameter of 1 cm on a transverse CT scan.
However, numerous other criteria also have been
used including the following: (1) a long-axis diameter
of 1 cm; (2) a short-axis diameter of 1.5 cm;
(3) a short-axis diameter of 1 cm plus evidence
of central necrosis or disruption of the capsule; and
(4) a short-axis diameter of 2 cm regardless of
Figure 1. TNM staging of lung cancer. Reprinted with permission from Lababede et al.
3a
nodal morphology. The reported sensitivity and
specificity rates for the identification of malignant
involvement will vary depending on which criteria
are used in the assessment of individual nodal sta-
tions.
4,5
The majority of studies evaluating CT scan
accuracy have used a short axis of 1 cm as the
threshold for the definition of abnormal nodes. In
doing so, a conscious effort has been made to err on
the side of higher sensitivity at the expense of lower
specificity in an understandable effort to minimize
the number of false-negative evaluations.
The background article in this supplement by
Toloza et al describes their study assessing the
performance characteristics of noninvasive imaging
procedures for staging the mediastinum, based on
computerized searches of the medical literature.
Twenty-three studies evaluating the accuracy of CT
scanning for staging the mediastinum were identified
on the basis of the following criteria: (1) publication
in a peer-reviewed journal; (2) study size of 50
patients; (3) patient group not included in a subse-
quent update of the study; (4) histologic or cytologic
confirmation of involvement of mediastinal nodes or
extrathoracic sites, as well as of the primary tumor;
and (5) availability of the raw data (see Table 1 in
the article by Toloza et al). The combined studies
yielded 4,793 evaluable patients. The pooled sensi-
tivity of CT scanning for staging the mediastinum
0.68), and the pooled specificity for CT scanning was
0.81 (95% CI, 0.74 to 0.86). The overall positive
predictive value was 53% (range, 26 to 100%), while
the overall negative predictive value was 82% (range,
63 to 85%) [see evidence Table 1 in the article by
Toloza et al].
CT scanning is thus an imperfect tool for staging
the mediastinum. However, since CT scanning usu-
ally guides the choice of nodes for selective node
biopsy by mediastinoscopy or needle aspiration, it
remains an important diagnostic tool in lung cancer.
Accurate noninvasive node staging is essential in that
the choice of individual nodes for sampling by
nonsurgical invasive techniques, including trans-
bronchial, transthoracic, or transesophageal needle
aspiration, will be directed by the findings of the CT
scan. The limitation of CT scan-based mediastinal
lymph node evaluation is evident in the fact that 5 to
15% of patients with clinical stage T1N0 lesions will
be found to have positive lymph node involvement
by surgical lymph node sampling.
6
Perhaps the most important message in evaluating
the accuracy of CT scanning is that approximately
40% of all nodes that are deemed malignant by CT
scan criteria are actually benign, depending on the
patient population. Specificity can be affected by
clinical factors such as the presence of postobstruc-
tive pneumonitis.
7
There is no node size that can
determine reliably the stage and operability of the
tumor. In cases in which the CT scan criteria for
the identification of a metastatic node are met, the
clinician must still prove beyond reasonable doubt
by biopsy or resection that the node is indeed
malignant. Given the limitations of its imperfect
sensitivity and specificity, it is usually inappropriate
to rely solely on the CT scan to determine medias-
tinal lymph node status. Nonetheless, CT scanning
continues to play an important and necessary role
in the evaluation of patients with lung cancer. This
conclusion is supported by the most recent Ameri-
can Thoracic Society/European Respiratory Society
statement
6
on the pretreatment evaluation of
NSCLC, in which CT scanning was recommended
for the evaluation of mediastinal lymph nodes in all
patients with suspected NSCLC.
Recommendations
1. For patients with either a known or suspected
lung cancer who are eligible for treatment, a
CT scan of the chest should be performed.
grade of evidence, B
2. In patients with enlarged mediastinal lymph
nodes seen on CT scans (ie, 1 cm in the
short axis), further evaluation of the mediasti-
num should be performed prior to surgical
resection of the primary tumor. Level of evi-
dence, fair; benefit, substantial; grade of evi-
dence, B
Positron Emission Tomography: Positron emission
tomography (PET) scanning is an imaging modality
based on the biological activity of neoplastic cells.
Lung cancer cells demonstrate increased cellular
uptake of glucose and a higher rate of glycolysis
when compared to normal cells.
8
The radiolabeled
glucose analog
18
F-fluoro-deoxy-D-glucose (FDG)
undergoes the same cellular uptake as glucose but,
after phosphorylation, is not further metabolized and
becomes trapped in cells.
9
Accumulation of the
isotope then can be identified using a PET camera.
PET scanning is thus a metabolic imaging technique
based on the function of a tissue rather than on its
anatomy. The specific criterion for an abnormal PET
scan is either a standard uptake value of 2.5 or
uptake in the lesion that is greater than the back-
ground activity of the mediastinum. It has proved
useful in differentiating neoplastic from normal tis-
sues. However, the technique is not infallible as
certain nonneoplastic processes, including granulo-
matous and other inflammatory diseases as well as
infections, also may demonstrate positive PET imag-
ing findings. Furthermore, size limitations are also
an issue, with the lower limit of resolution of the
study being approximately 1 to 1.2 cm, depending on
the intensity of uptake of the isotope in abnormal
cells.
10
A burgeoning number of studies in the last several
years have reported on the utility of FDG-PET
scanning in the assessment of the mediastinum in
patients with lung cancer. Increasing the availability
of the technology now allows PET scanning to be
used widely as a diagnostic tool. It should be noted
that PET scanning is primarily a metabolic examina-
tion and has limited anatomic resolution. It is possi-
ble by PET scanning to identify lymph node stations
but not individual lymph nodes. CT scanning pro-
vides much more anatomic detail but lacks the
functional information provided by PET scanning.
The background article by Toloza et al evaluates the
performance characteristics of PET scanning for
staging the mediastinum, based on computerized
searches of the medical literature. Nineteen studies
were identified that met the following criteria:
(1) publication in a peer-reviewed journal; (2) study
size of 20 patients; (3) patient group not included
in a subsequent update of the study; (4) histologic or
cytologic confirmation of mediastinal nodes or ex-
trathoracic sites, as well as the primary tumor; and
(5) availability of the raw data. All of these studies
were interpreted in conjunction with the findings of
patients CT scans so that correlation of the findings
on PET scanning was put in context with the ana-
tomic location of the lesion on the CT scan. In all
studies, FDG was the radiopharmaceutical that was
used for imaging. The combined studies yielded
1,111 evaluable patients. The pooled sensitivity was
0.85 (95% CI, 0.79 to 0.89), and the pooled speci-
ficity was 0.88 (95% CI, 0.82 to 0.92). The overall
positive predictive value was 0.78 (range, 0.40 to
1.00), and the negative predictive value was 0.93
(range, 0.75 to 1.00) [see evidence Table 2 in
background article by Toloza et al].
Thus, it appears that PET scanning has both
higher sensitivity and specificity for the evaluation of
mediastinal lymph nodes than does CT scanning.
This imaging technique will almost certainly assume
an increasingly important role in the evaluation of
patients with lung cancer. However, like CT scan-
ning, PET scanning is also imperfect. While a neg-
ative result of a mediastinal PET scan may obviate
the need for mediastinoscopy prior to thoracotomy, a
positive result of a mediastinal PET scan should not
negate further evaluation or the possibility of resec-
tion. In the latter case, mediastinoscopy or lymph
node sampling still should be pursued, as the possi-
bility of a false-positive PET scan result cannot be
ignored.
Recommendations
3. For patients who are candidates for surgery,
where available, a whole-body FDG-PET scan
is recommended to evaluate the mediastinum.
4. In patients with abnormal FDG-PET scan
findings, further evaluation of the mediasti-
num with sampling of the abnormal lymph
node should be performed prior to surgical
resection of the primary tumor. Level of evi-
dence, fair; benefit, substantial; grade of evi-
dence, B
Other Imaging Modalities: In many places, the
clinical utility of FDG-PET scanning is limited be-
cause of the lack of availability of a PET camera or
the radiopharmaceutical FDG. There is a growing
interest in other noninvasive nuclear imaging tech-
niques that also take advantage of altered biological
characteristics of malignant tissues. One such tech-
nique is based on the increased avidity with which
various neoplasms bind somatostatin. Such tissues
also will bind somatostatin peptide analogs such as
depreotide or octreotide, which, when complexed
with radionuclides such as
99m
Tc, can be imaged
by single-photon emission CT (SPECT) scanning.
SPECT technology is widely available and less ex-
pensive than PET scanning. Initial results
11,12
sug-
gest that the sensitivity and specificity of SPECT
scanning with
99m
Tc depreotide (NeoTect; Berlex
Imaging; Montville, NJ) may be comparable to those
of FDG-PET scanning. Preliminary results also sug-
gest that other radiopharmaceutical agents may be of
potential use with SPECT scanning in the identifi-
cation of malignant neoplasms.
1318
However, expe-
rience with SPECT imaging for lung cancer is still
very limited, and thus SPECT scanning should
not be considered as an alternative to FDG-PET
scanning.
MRI: Like CT scanning, MRI is an anatomic
study. Experience evaluating MRI in the detection of
mediastinal lymph nodes in patients with lung cancer
is minimal. Thus, a comparison of the test character-
istics between MRI and CT scanning cannot be
performed. Two reports
19,20
also suggest that the use
of contrast enhancement may improve the accuracy
of MRI in the evaluation of mediastinal lymph
nodes. However, most centers continue to rely on
CT scanning as the noninvasive anatomic study of
choice for evaluating the potential mediastinal
spread of lung cancer. MRI can be useful in evalu-
ating superior sulcus tumors, especially with regard
to possible invasion of the brachial plexus, and for
vertebral invasion.
Recommendation
lung cancer who are eligible for treatment, an
MRI examination of the chest should not be
performed for staging the mediastinum but
should be performed in patients with NSCLC
involving the superior sulcus to evaluate the
brachial plexus or for vertebral body invasion.
The Search for Metastatic Disease
This section should reinforce the concept that the
clinician should only search for metastatic disease if
there is a compelling reason to do so. The purpose of
extrathoracic scanning in patients with NSCLC is
usually to detect metastatic disease at common met-
astatic sites such as the adrenal glands, liver, brain,
and skeletal system, thereby sparing the patient
fruitless surgical intervention.
21
The preferred scans
for staging patients with NSCLC in 2002 are a CT
scan of the chest, a CT scan or MRI with contrast of
the brain, and
99m
Tc nuclear imaging of the skeletal
system. The use of whole-body PET scans for ex-
trathoracic staging is just beginning, and initial stud-
ies
10,2224
suggest that PET scanning can disclose
non-CNS metastatic disease that has not been de-
tected by standard methods in 10 to 20% of cases.
It is well-established that abnormal symptoms,
physical findings, and routine blood test results in
the initial clinical evaluation of patients with NSCLC
are associated with a significant yield (around 50%)
of abnormal scan findings.
21
Moreover, a rough
semiquantitative relationship has been demonstrated
in some studies between the number of abnormal
clinical factors and the frequency of abnormal scan
results.
21,25
On the other hand, in the absence of all
clinical factors, the scan yield is much lower, giving
rise to the recommendation that scans be omitted in
this setting.
2531
Other important variables focus on
the primary lesion, since more scan abnormalities are
associated with advanced thoracic lesions (ie, T and
N factors).
32,33
This is particularly true for stage N2
disease in which asymptomatic metastases have been
documented at a higher rate than would have been
expected.
33
There has been some controversy with
regard to cell type and the incidence of asymptom-
atic metastases. Several studies have documented a
higher incidence of brain metastases with adenocar-
cinomas as opposed to squamous cell cancers,
34,35
but the largest single series of patients with stage I
and II lung cancer found no difference.
30
Several important caveats must be mentioned at
the outset. First is the issue of false-positive scan
findings. Clinical entities that frequently give rise to
false-positive scan findings include adrenal adeno-
mas (present in 2 to 9% of the general population),
hepatic cysts, degenerative joint disease, old frac-
tures, and a variety of nonmetastatic space-taking
brain lesions. When clinically indicated, additional
imaging studies and/or biopsies are performed to
establish the diagnosis, but the complications and
costs resulting from such subsequent investigations
have received insufficient attention.
36
A second
problem is that of false-negative scan findings, that
is, cases in which metastases are present but are not
picked up by current scanning techniques. This was
demonstrated convincingly by Pagani,
37
who found
metastatic NSCLC in 12% of radiologically normal
adrenal glands by percutaneous biopsy. A third
difficulty is that most studies fail to carefully specify
exactly which elements comprise the prescan clinical
evaluation or use differing clinical indicators. Organ-
specific findings such as headache and non-
organ-specific complaints such as weight loss are
both important.
25,38
The current preferred expand-
ed clinical evaluation includes evaluation of organ-
specific and constitutional signs and symptoms, along
with the performance of simple laboratory tests
(Table 2).
21
Furthermore, Guyatt et al
39
have shown
that careful delineation and quantification of histor-
ical features using a 5-point scale of severity can
importantly affect the subsequent scan yield and
ultimately the incidence of metastases after lung
cancer surgery. A fourth issue is an ascertainment
problem, since abnormal scan findings in many
studies were not followed up with definitive biopsy
proof of metastatic disease. This may relate to ana-
tomic factors, the overall debility or refusal of the
patient, or a variety of other cogent clinical concerns.
Fifth, it must be noted that even biopsy proof of
metastatic disease does not dictate a certain clinical
management pathway. Carefully selected patients
with localized lung cancers in the thorax, accessible
solitary metastases to the brain or adrenal glands,
and other favorable clinical features may obtain
long-term survival with an aggressive treatment ap-
proach, including surgical extirpation of both the
primary and metastatic site.
40,41
Finally, the lack of
prospective randomized trials and outcome studies
in the area of extrathoracic scanning is striking. One
retrospective study
42
showed that scanning asymp-
tomatic patients with early-stage NSCLC did not
help to predict recurrences postoperatively or to
improve survival. The only prospective randomized
trial
43
showed no statistical difference in recurrence
rates or survival in a group randomized to bone
scintigraphy and CT scanning of the head, liver, and
adrenal glands compared with a group assigned to
CT scanning of the chest and mediastinoscopy,
followed by thoracotomy when appropriate.
Specific Considerations
Adrenal and Hepatic Imaging: It is relatively
common to encounter adrenal masses on routine CT
scans, but many of these lesions are probably unre-
lated to the malignant process. Until now, partial
liver imaging and inclusion of the adrenal glands on
chest CT scans have been routine. The calculated
summary estimate for the negative predictive value
of the clinical evaluation is 95% (95% CI, 93 to 96%),
if CT imaging of the adrenal glands and liver is
employed to attempt to disclose occult metastases
(see evidence Table 5 in the background article by
Toloza et al). A unilateral adrenal mass in a patient
with NSCLC is more likely to be a metastasis than a
benign lesion according to some studies,
21,44
but not
others.
45,46
In the presence of clinical stage T1N0
NSCLC, adenomas predominate,
47,48
whereas adre-
nal metastases are frequently associated with large
intrathoracic tumors or other extrathoracic metasta-
ses.
21,49
Many studies have suggested that the size of
a unilateral adrenal abnormality as seen on a CT scan
is an important predictor of metastatic spread,
but this is not a universal finding.
50
Lesions that are
3 cm in size are more likely to signify metastases,
but benign disease is still possible. Four possible
approaches to distinguishing between malignant and
benign processes have been proposed, as follows:
evaluation by CT scan or MRI criteria; evaluation
with subsequent imaging; evaluation by percutane-
ous biopsy; and evaluation by adrenalectomy. Well-
defined, low-attenuation (fatty) lesions with a
smooth rim on unenhanced CT scans are more likely
to be benign adenomas,
5153
but the CT scan appear-
ance of many lesions is insufficiently distinctive.
53
Follow-up scanning with repeat CT scans, serial
ultrasounds, or MRI (especially with chemical shift
and dynamic gadolinium-enhanced techniques
54
), or
131-6-betaiodomethylnorcholesterol scanning
55
can
sometimes help with the critical distinction between
metastatic disease and adenoma. Percutaneous adre-
nal biopsy is a relatively safe and effective means of
achieving a definitive diagnosis in doubtful cases and
is especially important when the histology of the
adrenal mass will dictate subsequent manage-
ment.
56,57
However, this procedure may be nondiag-
nostic or unfeasible due to anatomic constraints.
When insufficient material results from a biopsy,
repeat aspiration or even the performance of an
adrenalectomy should be considered.
50,53
Most liver lesions are benign cysts and hemangi-
omas, but contrast CT scanning (or ultrasound) often
is required to establish a likely diagnosis.
6
A percu-
taneous biopsy can be performed when diagnostic
certainty is required. One meta-analysis
36
that spe-
cifically reviewed hepatic studies derived a pooled
yield of 3% for liver metastases in asymptomatic
patients with NSCLC.
Brain Imaging: In most studies, the yield of CT
scanning/MRI of the brain in NSCLC patients with
negative clinical examination results is 0 to 10%,
5864
possibly rendering the test cost-ineffective.
62
The
calculated summary estimate of the negative predic-
tive value in this setting is 95% (range, 91 to 96%)
[see evidence Table 4 in the background article by
Toloza et al].
An association of positive findings between brain
metastases with stage N2 disease in the chest and
the presence of adenocarcinoma has been de-
scribed.
34,59,64
The false-negative rate wherein pa-
tients return with brain metastases within 12 months
of the original scan is reported to be 3%.
59
False-
positive scan results can be a problem in up to 11%
of patients due to brain abscesses, gliomas, and other
lesions.
65
Therefore, biopsy may be essential in
patients in whom disease management is critically
dependent on the histology of the brain lesion.
MRI is more sensitive than CT scanning of the
Table 2Clinical Findings Suggesting Metastatic Disease*
Testing Finding
Symptoms elicited in history Constitutional: weight loss 10 lb
Musculoskeletal: focal skeletal pain
Neurologic: headaches, syncope, seizures, extremity weakness, recent change in mental status
Signs found on physical examination Lymphadenopathy (1 cm)
Hoarseness, superior vena cava syndrome
Bone tenderness
Hepatomegaly (13-cm span)
Focal neurologic signs, papilledema
Soft-tissue mass
Routine laboratory tests Hematocrit, 40% in men and 35% in women
Elevated alkaline phosphatase, GGT, SGOT, and calcium levels
*GGT gamma-glutamyltransferase; SGOT serum glutamic-oxaloacetic transaminase.
brain and picks up more lesions and smaller le-
sions,
66
but in some studies
67
this has not translated
into a clinically meaningful difference in terms of
survival. While studies show that MRI can identify
additional lesions in patients with metastases, there
are no studies showing that MRI is able to identify
more patients with metastases from lung cancer
compared to CT scanning. Therefore, CT scanning is
an acceptable modality for evaluating patients for
metastatic disease. If the primary lesion is more
advanced than stage T1N0M0, MRI with contrast
can identify asymptomatic, verifiable metastases to
the brain in 22% of patients with NSCLC and
surgically resectable thoracic disease.
68
However,
the use of routine MRI in staging NSCLC patients
with negative clinical evaluations has not been stud-
ied adequately to date.
Bone Imaging: The problem of false-positive scan
abnormalities in radionuclide bone scintigraphy is
particularly nettlesome, owing to the frequency of
degenerative and traumatic skeletal damage and the
difficulty in obtaining a definitive diagnosis via fol-
low-up imaging or biopsy. False-positive bone imag-
ing also occurs with MRI, which may be no more
accurate than nuclear bone imaging.
68
The calcu-
lated summary estimate of the negative predictive
value in this setting is 90% (95% CI, 86 to 93%) for
radionuclide bone imaging with a negative clinical
assessment (see evidence Table 6 in the background
article by Toloza et al). The relatively high frequency
of unsuspected positive scan results has led some to
recommend routine bone scanning in all preopera-
tive patients.
69
Unfortunately, not all patients had
biopsy-proven metastatic disease, and thus false-
positive scan results may have accounted for some of
the abnormalities. False-negative scan results also
can be a problem, and in one series
70
6% of patients
with initially negative bone scan findings developed
skeletal metastases within 1 year.
In summary, the noninvasive clinical staging of
lung cancer relies on the clinical evaluation and a
number of readily available staging studies. The
clinician must be wary of abnormal scan results that
may falsely suggest metastatic disease to the medi-
astinum and distant sites. Tissue confirmation by
whatever means necessary is the rule rather than the
exception prior to deciding on the correct stage and
the most appropriate treatment.
Recommendations
lung cancer, a thorough clinical evaluation
similar to that listed in Table 2 should be
performed. Level of evidence, good; benefit,
substantial; grade of evidence, A
7. Patients with abnormal clinical evaluations
should undergo imaging for extrathoracic me-
tastases. Site-specific symptoms warrant di-
rected evaluation of that site with the most
appropriate study (eg, head CT scan, bone scan,
or abdominal CT scan). Level of evidence,
good; benefit, substantial; grade of evidence, A
8. Patients with clinical stage I or II lung cancer
and normal results of a clinical evaluation
require no further imaging for detection of
extrathoracic disease. Level of evidence, good;
benefit, substantial; grade of evidence, A
9. Patients with stage IIIA and IIIB disease
should have routine imaging studies for the
detection of extrathoracic metastases (eg, head
CT scan, bone scan, and abdominal CT scan).
grade of evidence, C
10. Patients with abnormal imaging study results
should not be excluded from potentially cur-
ative surgery without tissue confirmation or
overwhelming clinical and radiographic evi-
dence of metastases. Level of evidence, good;
lung cancer who are eligible for treatment, a
CT scan of the chest should be performed.
2. In patients with enlarged mediastinal lymph
nodes on CT scans (ie, 1 cm on the short
axis), further evaluation of the mediastinum
should be performed prior to surgical resec-
tion of the primary tumor. Level of evidence,
fair; benefit, substantial; grade of evidence, B
3. For patients who are operative candidates,
where available, a whole-body FDG-PET scan
is recommended to evaluate the mediastinum.
4. In patients with abnormal results of FDG-
PET scanning, further evaluation of the me-
diastinum with sampling of the abnormal
lymph node should be performed prior to
surgical resection of the primary tumor. Level
evidence, B
lung cancer who are eligible for treatment, an
MRI of the chest should not be performed for
staging the mediastinum but should be per-
formed in patients with NSCLC involving the
superior sulcus for evaluation of the brachial
plexus or for evaluation of vertebral body
invasion. Level of evidence, fair; benefit, sub-
stantial; grade of evidence, B
lung cancer, a thorough clinical evaluation
similar to that listed in Table 2 should be
performed. Level of evidence, good; benefit,
substantial; grade of evidence, A
7. Patients with abnormal clinical evaluations
should undergo imaging for extrathoracic me-
tastases. Site-specific symptoms warrant di-
rected evaluation of that site with the most
appropriate study (eg, head CT scan, bone
scan, and abdominal CT scan). Level of evi-
evidence, A
8. Patients with clinical stage I or II lung cancer
and a normal clinical evaluation require no
further imaging for extrathoracic disease.
grade of evidence, A
9. Patients with stage IIIA and IIIB disease
should have routine imaging for the detection
of extrathoracic metastases (eg, head CT scan,
bone scan, and abdominal CT scan). Level of
evidence, C
10. Patients with abnormal imaging study results
should not be excluded from potentially cur-
ative surgery without tissue confirmation or
overwhelming clinical and radiographic evi-
dence of metastases. Level of evidence, good;
References
1 Darling GE. Staging of the patient with small cell lung
cancer. Chest Surg Clin N Am 1997; 7:8194
3 Mountain CF. A new international system for staging lung
cancer. Chest 1986; 89(suppl):225s233s
3a.Lababede O, Meziane MA, Rice TW. TNM staging of lung
cancer: a quick reference chart. Chest 1999; 115:233235
4 Bollen ECM, Goei R, Hof-Grootenboer BE, et al. Interob-
server variability and accuracy of computed tomographic
assessment of nodal status in lung cancer. Ann Thorac Surg
1994; 58:158162
5 Ratto GB, Frola C, Cantoni S, et al. Improving clinical
efficacy of computed tomographic scan in the preoperative
assessment of patients with non-small cell lung cancer. Tho-
rac Cardiovasc Surg 1990; 99:416425
6 American Thoracic Society/European Respiratory Society.
Pretreatment evaluation of non-small-cell lung cancer. Am J
7 McLoud TC, Bourgouin PM, Greenberg RW, et al. Broncho-
genic carcinoma: analysis of staging in the mediastinum with
CT by correlative lymph node mapping and sampling. Radi-
ology 1992; 182:319323
8 Nolop KB, Rhodes CG, Brudin LH, et al. Glucose utilization
in vivo by human pulmonary neoplasms. Cancer 1987; 60:
26822689
9 Wahl RI, Hitchins GD, Buchsbaum DJ, et al. 18F-2 deoxy-
2-fluoro-D-glucose uptake into human tumor xenografts:
feasibility studies for cancer imaging with positron-emission
tomography. Cancer 1991; 67:15441550
10 Pieterman RM, van Putten JWG, Meuzelaar JJ, et al. Preop-
erative staging of non-small-cell lung cancer with positron-
emission tomography. N Engl J Med 2000; 343:254261
11 Blum JE, Handmaker H, Rinne NA. The utility of a soma-
tostatin-type receptor binding peptide in the evaluation of
solitary pulmonary nodules. Chest 1999; 115:224232
12 Blum J, Handmaker H, Lister-James J, et al. A multicenter
trial with a somatostating analog
99m
Tc depreotide in the
evaluation of solitary pulmonary nodules. Chest 2000; 117:
12321238
13 Tsutami M, Yutani K, Nishimura T. Evaluation of lung cancer
by
99m
Tc-tetrofosmin SPECT: comparison with (
18
F) FDG-
PET. J Comput Assist Tomogr 2000; 24:574580
14 Spagnolo S. The diagnostic strategy for lung cancer. Chest
2000; 117:12191220
15 Shreve S, Steventon RS, Deters EC, et al. Oncologic diagno-
sis with 2-(fluorine-18) fluoro2-deoxy-D-glucose imaging: du-
al-head coincidence gamma camera vs positron emission
tomographic scanner. Radiology 1998; 207:431437
16 Kubota K. Changing pattern of lung cancer and its imaging:
201
T1 SPECT vs (
18
F)FDG PET. J Nucl Med 2001; 42:1497
1498
17 Higashi K, Ueda Y, Sakuma T, et al. Comparison of (
18
F)
FDG PET, and
201
T1 SPECT in evaluation of pulmonary
nodules. J Nucl Med 2001; 42:14891496
18 Goldsmith SJ, Kostakoglu L. Nuclear medicine imaging of
lung cancer. Radiol Clin North Am 2000; 38:511524
19 Kernstine KH, Stanford W, Mullen BF, et al. PET, CT, and
MRI with Combidex for mediastinal staging in non-small cell
lung carcinoma. Ann Thorac Surg 1999; 68:10221028
20 Crisci R, Di Cesare E, Lupattelli L, et al. MR study of N2
disease in lung cancer: contrast-enhanced method using gado-
linium-DTPA. Eur J Cardiothorac Surg 1997; 11:214217
for detecting metastatic lung cancer: a meta-analysis. Am J
22 Saunders CAB, Dussek JE, ODoherty MJ, et al. Evaluation
of fluorine-18-fluorodeoxyglucose whole body positron emis-
sion tomography imaging in the staging of lung cancer. Ann
Thorac Surg 1999; 67:790797
23 Weder W, Schmid RA, Bruchhaus H, et al. Detection of
extrathoracic metastases by positron emission tomography in
lung cancer. Ann Thorac Surg 1998; 66:886893
24 MacManus MP, Hicks RJ, Matthews JP, et al. High rate of
unsuspected distant metastases by PET in apparent stage III
non-small-cell lung cancer: implications for radical radiation
therapy. Int J Radiat Oncol Biol Phys 2001; 50:287293
25 Hooper RG, Tenholder MF, Underwood, et al. Computed
tomographic scanning in initial staging of bronchogenic car-
cinoma. Chest 1984; 85:774777
26 Turner P, Haggith JW. Preoperative radionuclide scanning in
27 Ramsdell JW, Peters RM, Taylor AT, et al. Multiorgan scans
for staging lung cancer-correlation with clinical evaluation.
J Thorac Cardiovasc Surg 1977; 73:653659
Respir Dis 1978; 118:279286
29 Osada H, Kojima K, Tsukada H, et al. Cost-effectiveness
associated with the diagnosis and staging of non-small-cell
lung cancer. Jpn Thorac Cardiovasc Surg 2001; 49:110
30 Tanaka K, Kubota K, Kodama T, et al. Extrathoracic staging
is not necessary for non-small-cell lung cancer with clinical
stage T1-2 N0. Ann Thorac Surg 1999; 68:10391042
31 British Thoracic Society, Society of Thoracic Surgeons of
Great Britain, Ireland Working Party. Guidelines on the
selection of patients with lung cancer for surgery. Thorax
2001; 56:89106
102:17481751
33 Grant D, Edwards D, Goldstraw P. Computed tomography of
the brain, chest, and abdomen in the preoperative assessment
of non-small cell lung cancer. Thorax 1988; 43:883886
34 Tarver RD, Richmond BD, Klatte EC. Cerebral metastases
from lung carcinoma: neurological and CT correlation work in
progress. Radiology 1984; 153:689692
35 Salvatierra A, Baamonde C, Llamas JM, et al. Extrathoracic
staging of bronchogenic carcinoma. Chest 1990; 97:10521058
36 Hillers TK, Sauve MD, Guyatt GH. Analysis of published
studies on the detection of extrathoracic metastases in pa-
tients presumed to have operable non-small cell lung cancer.
Thorax 1994; 49:1419
37 Pagani JJ. Non-small cell lung carcinoma adrenal metastases:
computed tomography and percutaneous needle biopsy in
their diagnosis. Cancer 1984; 53:10581060
38 Quinn DL, Ostrow LB, Porter DK, et al. Staging of non-small
cell bronchogenic carcinoma: relationship of the clinical
evaluation to organ scans. Chest 1986; 89:270275
39 Guyatt GH, Cook DJ, Griffith LE, et al. Surgeons assessment
of symptoms suggesting extrathoracic metastases in patients
with lung cancer. Ann Thorac Surg 1999; 68:309315
40 Shahidi H, Kvale PA. Long-term survival following surgical
treatment of solitary brain metastasis in non-small cell lung
cancer. Chest 1996; 109:271276
41 Urchel JD, Finley RK, Takita H. Long-term survival after
bilateral adrenalectomy for metastatic lung cancer: a case
report. Chest 1997; 112:848850
42 Ichinose Y, Hara N, Ohta M, et al. Preoperative examination
to detect distant metastases is not advocated for asymptom-
atic patients with stages I and II non-small cell lung cancer:
preoperative examination for lung cancer. Chest 1989; 96:
11041109
43 The Canadian Lung Oncology Group. Investigating extratho-
racic metastatic disease in patients with apparently operable
44 Goerg C, Schwerk WB, Wolf M, et al. Adrenal masses in lung
cancer: sonographic diagnosis and follow-up. Eur J Cancer
1992; 28A:14001403
45 Oliver TW, Bernardino ME, Miller JI, et al. Isolated adrenal
masses in non-small cell bronchogenic carcinoma. Radiology
1984; 153:217218
46 Burt M, Heelan RT, Coit D, et al. Prospective evaluation of
unilateral adrenal masses in patients with operable non-small
cell lung cancer: impact of magnetic resonance imaging.
J Thorac Cardiovasc Surg 1994; 107:584589
47 Pearlberg JL, Sandler MA, Beute GH, et al. T1N0M0
bronchogenic carcinoma: assessment by CT. Radiology 1985;
157:187190
48 Heavey LR, Glazer GM, Gross BH, et al. The role of CT in
staging radiographic T1N0M0 lung cancer. AJR Am J Roent-
genol 1986; 146:285290
49 Eggesbo HB, Hansen G. Clinical impact of adrenal expansive
lesions in bronchial carcinoma. Acta Radiol 1996; 37:343347
50 Ettinghausen SE, Burt ME. Prospective evaluation of unilat-
eral adrenal masses in patients with operable non-small cell
51 Porte HL, Ernst OJ, Delebecq T, et al. Is computed tomog-
raphy guided biopsy still necessary for the diagnosis of
adrenal masses in patients with resectable non-small cell lung
cancer. Eur J Cardiothorac Surg 1999; 15:597601
52 Macari M, Rofsky NM, Naidich DP, et al. Non-small cell lung
carcinoma: usefulness of unenhanced helical CT of the
adrenal glands in an unmonitored environment. Radiology
1998; 209:807881
53 Gillams A, Roberts CM, Shaw P, et al. The value of CT scanning
and percutaneous fine needle aspiration of adrenal masses in
biopsy-proven lung cancer. Clin Radiol 1992; 46:1822
54 Heinz-Peer G, Honigschnabi S, Schneider B, et al. Character-
ization of adrenal masses using MR imaging with histopathologic
correlation. AJR Am J Roentgenol 1999; 173:1522
55 Gross MD, Shapiro B, Bouffard JA, et al. Distinguishing
benign from malignant euadrenal masses. Ann Intern Med
1988; 109:613618
56 Nielson ME, Heaston DK, Dunnick NR, et al. Preoperative
CT evaluation of adrenal glands in non-small cell broncho-
genic carcinoma. AJR Am J Roentgenol 1982; 139:317320
57 Chapman GS, Kumar D, Redmond J, et al. Upper abdominal
computerized tomography scanning in staging non-small cell
lung carcinoma. Cancer 1984; 54:15411543
58 Cole JFH, Thomas JE, Wilcox AB, et al. Cerebral imaging in
the asymptomatic preoperative bronchogenic carcinoma pa-
tient: is it worthwhile? Ann Thorac Surg 1994; 57:838840
59 Kormas P, Bradshaw JR, Jeyasingham K. Preoperative com-
puted tomography of the brain in non-small cell bronchogenic
carcinoma. Thorax 1992; 47:106108
60 Jacobs L, Kinkel WR, Vincent RG. Silent brain metastasis
from lung carcinoma determined by computerized tomogra-
phy. Arch Neurol 1977; 77:690693
61 Butler AR, Leo JS, Lin JP, et al. The value of routine cranial
computed tomography in neurologically intact patients with
primary carcinoma of the lung. Radiology 1979; 131:339401
62 Colice GL, Birkmeyer JD, Black WC, et al. Cost-effectiveness
of head CT in patients with lung cancer without clinical
evidence of metastases. Chest 1995; 108:12641271
63 Mintz BJ, Turhim S, Alexander S, et al. Intracranial metas-
tases in the initial staging of bronchogenic carcinoma. Chest
1984; 86:850853
64 Ferrigno D, Buccheri G. Cranial computed tomography as a
part of the initial staging procedures for patients with non-
small cell lung cancer. Chest 1994; 106:10251029
65 Patchell RA, Tibbs PA, Walsh JW, et al. A randomized trial of
surgery in the treatment of single metastases to the brain.
N Engl J Med 1990; 322:494500
66 Davis PC, Hudgins PA, Peterman SB, et al. Diagnosis of
cerebral metastases: double-dose delayed CT vs contrast-
enhanced MR imaging. AJR Am J Roentgenol 1991; 156:
10391046
67 Yokio K, Kamiya N, Matsuguma H, et al. Detection of brain
metastasis in potentially operable non-small cell lung cancer:
a comparison of CT and MRI. Chest 1999; 115:714719
68 Earnest I, F, Ryu JH, Miller GM, et al. Suspected non-small
cell lung cancer: incidence of occult brain and skeletal
metastases and effectiveness of imaging for detection-pilot
study. Radiology 1999; 211:137145
69 Tornyos K, Garcia O, Karr B, et al. A correlation study of
bone scanning with clinical and laboratory findings in the
staging of non-small cell lung cancer. Clin Nucl Med 1991;
16:107109
70 Michel F, Soler M, Imhof E, et al. Initial staging of non-small
cell lung cancer: value of routine radioisotope bone scanning.
Thorax 1991; 46:469473
2003;123;147-156 Chest
Detterbeck
Gerard A. Silvestri, Lynn T. Tanoue, Mitchell L. Margolis, John Barker and Frank
The Noninvasive Staging of Non-small Cell Lung Cancer: The Guidelines
& Services
Updated Information
S
References
S#BIBL
free at:
Citations
S#otherarticles
Reprints
2003;123;157-166 Chest
Eric M. Toloza, Linda Harpole, Frank Detterbeck and Douglas C. McCrory
Evidence
Invasive Staging of Non-small Cell Lung Cancer: A Review of the Current
ISSN: 0012-3692.
Invasive Staging of Non-small Cell Lung
Cancer*
A Review of the Current Evidence
Eric M. Toloza, MD, PhD; Linda Harpole, MD, MPH;
Frank Detterbeck, MD, FCCP; Douglas C. McCrory, MD, MHS
Study objectives: To determine the test performance characteristics of transbronchial needle
aspiration (TBNA), transthoracic needle aspiration (TTNA), endoscopic ultrasound-guided needle
aspiration (EUS-NA), and mediastinoscopy in staging non-small cell lung cancer (NSCLC).
Design, setting, and participants: Systematic search of MEDLINE, HealthStar, and Cochrane
Library databases to July 2001 and print bibliographies. Included were studies comparing staging
results of TBNA, TTNA, EUS-NA, or mediastinoscopy against either tissue histologic confirmation
or long-term clinical follow-up (> 1 year). Patients included were those with NSCLC or small cell
lung cancer.
Measurement and results: For patients with lung cancer, the pooled sensitivity for TBNA was 0.76,
the pooled specificity was 0.96, and the negative predictive value (NPV) was 0.71. For TTNA, the
pooled sensitivity was 0.91, with an NPV of 0.78. EUS-NA had a pooled sensitivity of 0.88, a pooled
specificity of 0.91, and an NPV of 0.77. For standard cervical mediastinoscopy, the pooled
sensitivity was 0.81, with an NPV of 0.91. The addition of either extended cervical mediastinos-
copy or anterior mediastinotomy to standard cervical mediastinoscopy appeared to improve the
sensitivity of any of the procedures alone.
Conclusions: Invasive clinical staging of NSCLC can be performed effectively by TBNA, TTNA,
EUS-NA, or mediastinoscopy. Selection of the appropriate study is dependent on the degree of
suspicion for metastatic disease, the patients comorbid illnesses, and the availability and
performance characteristics of procedural options. (CHEST 2003; 123:157S166S)
Key words: biopsy needle; false-negative rates; lung neoplasm; lymphatic metastasis; mediastinoscopy; predictive value
of tests; sensitivity and specificity
Abbreviations: CI confidence interval; EUS-NA endoscopic ultrasound-guided needle aspiration;
NPV negative predictive value; NSCLC non-small cell lung cancer; PPV positive predictive value;
SCLC small cell lung cancer; TBNA transbronchial needle aspiration; TTNA transthoracic needle aspiration
T
he primary aim of intrathoracic staging in non-
small cell lung cancer (NSCLC) is the evaluation
of mediastinal lymph node involvement. Accurate
assessment of mediastinal lymph node status affects
a patients prognosis and treatment plan, as the
presence of mediastinal lymph node involvement
indicates the presence of stage IIIA or IIIB lung
cancer. This suggests either inoperability and/or the
need for treatment by chemotherapy and/or radio-
therapy. Recently, induction therapy followed by
surgery has been suggested for stage IIIA
NSCLC.
13
Since mediastinal lymph node involve-
ment is found in 30 to 44% of patients with newly
diagnosed lung cancer, testing is required to rule in
or rule out such disease.
Noninvasive techniques to evaluate mediastinal
nodes rely on either lymph node size (CT, endo-
scopic ultrasound) or metabolism (positron emission
tomography) to detect cancerous involvement. How-
ever, while noninvasive tests can identify nodes
suspicious for cancer, they do not provide definitive
tissue diagnosis and often are not sufficient for
initiation of nonsurgical treatment. Thus, invasive
tests are often required to further evaluate nonre-
sectability.
Invasive techniques utilize needle biopsy or surgi-
cal open biopsy to obtain tissue samples to confirm
*From the Departments of Surgery (Dr. Toloza) and Medicine
(Dr. Harpole), and the Center for Clinical Health Policy Re-
search (Dr. McCrory), Duke University Medical Center,
Durham; the Department of Veterans Affairs Medical Center,
Durham; and Department of Surgery (Dr. Detterbeck), Univer-
sity of North Carolina, Chapel Hill, NC.
Supported by a contract from the American College of Chest
Physicians.
Correspondence to: Eric Toloza, MD, PhD, Duke Thoracic
Oncology, DUMC Box 3048, Durham, NC 27710; e-mail:
toloz001@mc.duke.edu
the diagnosis of metastatic disease. Needle biopsy
techniques include transbronchial needle aspiration
(TBNA), transthoracic needle aspiration (TTNA),
and endoscopic ultrasound-guided needle aspiration
(EUS-NA). Surgical open biopsy can be performed
by standard cervical mediastinoscopy, extended cer-
vical mediastinoscopy, or anterior mediastinotomy.
TBNA is performed with the aid of bronchoscopy.
A needle catheter is passed through the working
channel of the bronchoscope and guided to the area
of the tracheobronchial tree overlying the mediasti-
nal lymph node of interest. The needle catheter,
which comes in varying gauges, is then advanced
through the tracheal or carinal wall into the medias-
tinal lymph node, and an aspiration biopsy obtained.
Larger-gauge needles may be used in an attempt to
obtain a core of tissue for histologic examination.
Several passes may be performed until an adequate
specimen is obtained. Bedside or on-site cytopatho-
logic examination of needle aspiration specimens
may improve the yield of TBNA, as the rapid
evaluation of collected specimens would give the
interventionalist feedback as to the need for fur-
ther aspirations. This technique is limited by being
a blind biopsy and by the difficulty of sampling
more than a few nodal stations. Guidance by
emerging imaging techniques, such as real-time CT-
fluoroscopy, endobronchial ultrasound, and virtual
bronchoscopy (using three-dimensional images re-
constructed from routine helical CT scans), may im-
prove the yield of TBNA and are under investigation.
The main complications of TBNA include those inher-
ent to bronchoscopy, such as laryngospasm, and those
specific to the biopsy, such as endobronchial bleeding.
TTNA is performed with the aid of CT or, less
often, with fluoroscopic guidance. The biopsy needle
is passed percutaneously and guided to the medias-
tinal node of interest from which an aspiration biopsy
is obtained. Similar to TBNA, TTNA may require
several passes of the needle to obtain an adequate
specimen, and bedside cytopathologic examination
of collected specimens may minimize the number of
needle passes necessary to yield a diagnosis. As with
TBNA, larger-gauge needles may also be used in
attempt to obtain a core of tissue for histologic
examination. TTNA is radiologically guided, but like
TBNA, is limited in its ability to sample more than a
few nodal stations. The main complications of TTNA
are pneumothorax and intrathoracic bleeding.
EUS-NA is performed with the aid of esophagos-
copy. A biopsy needle catheter is passed through the
working channel of the endoscope, through the
esophageal wall, and guided ultrasonographically
toward mediastinal nodes of interest. Similar to both
TBNA and TTNA, several passes of the needle may
be needed to obtain adequate specimens, and bed-
side cytopathologic examination of collected speci-
mens may minimize the number of needle passes
necessary to yield a diagnosis. This technique is
guided by ultrasound, but it too is limited in its
ability to sample more than a few nodal stations. One
advantage of endoscopic ultrasound is that the simul-
taneous qualitative assessment of the mediastinal
nodes provided by ultrasound analysis for features
such as necrosis may aid in the selection of nodes to
biopsy that are more likely to be malignant.
Standard cervical mediastinoscopy is a surgical
open-biopsy technique that involves making a small
incision above the suprasternal notch. Dissection is
carried down to the pretracheal fascia, beneath
which a mediastinoscope is passed into the medias-
tinum toward the carina. Blunt dissection paratra-
cheally, bilaterally, and subcarinally exposes the me-
diastinal lymph nodes, which can then be sampled
under direct vision. All four paratracheal lymph node
stations (levels 2R, 2L, 4R, and 4L) and the anterior
subcarinal lymph node station (level 7) can be easily
sampled in this fashion. This procedure does require
general anesthesia and carries risks of bleeding and
left laryngeal nerve injury.
Extended cervical mediastinoscopy is an extension
of the standard cervical mediastinoscopy that allows
biopsy of the aortopulmonary window (level 5) and
preaortic (level 6) lymph node stations. Anterior
mediastinotomy also allows biopsy of the aortopul-
monary window and preaortic lymph node stations.
However, extended cervical mediastinoscopy is per-
formed through the same cervical incision as the
standard cervical mediastinoscopy. The mediastino-
scope is passed between the brachiocephalic artery
and the left carotid artery over the aortic arch to the
aortopulmonary window. In contrast, anterior medi-
astinotomy requires a separate incision parasternally,
usually at the level of the second or third intercostal
space. Both procedures are associated with a risk of
bleeding. Extended cervical mediastinoscopy has an
additional risk of embolic stroke, while anterior
mediastinotomy often results in violation of the
pleura or injury to the internal mammary artery.
Both procedures allow biopsy of the lymph nodes
under direct vision, but extended cervical mediasti-
noscopy has the added advantage of allowing access
to the paratracheal and subcarinal lymph nodes for
biopsy without additional incisions.
To evaluate the usefulness of TBNA, TTNA, EUS-NA, medi-
astinoscopy, and mediastinotomy in determining the presence or
absence of metastatic disease in mediastinal lymph nodes, we
analyzed published studies that reported the ability of the various
invasive methods to correctly establish the pathologic stage of
newly diagnosed or suspected lung cancer. In this review, needle
aspiration is considered a type of biopsy. The terms needle aspiration
and needle biopsy are, thus, considered interchangeable.
Key Questions
Two key questions were identified: (1) What are the sensitiv-
ities and specificities of TBNA, TTNA, EUS-NA, mediastinos-
copy, and mediastinotomy in patients with lung cancer for
detecting malignancy in mediastinal lymph nodes? (2) How
accurate are negative lymph node biopsies obtained by TBNA,
TTNA, EUS-NA, mediastinoscopy, and mediastinotomy in pre-
dicting the absence of nodal metastases in patients with lung
cancer?
Study Identification
Computerized searches of the MEDLINE bibliographic data-
base (January 1991 to July 2001), HealthStar, and the Cochrane
Library were performed. Key words used for the search included
lung neoplasm, bronchial neoplasm, mediastinoscopy, neoplasm
staging, neoplasm metastasis, lymphatic metastasis, biopsy, nee-
dle biopsy, CT, mediastinum radiography, emission-CT, and
sensitivity and specificity. In addition, we searched the reference
lists of included studies, selected textbooks, practice guidelines,
systematic reviews, and meta-analyses in order to ensure that all
relevant studies were identified. Only articles published in
English were considered.
Pertinent studies published between 1980 and 1991 were
identified through a previously published search
4
that included
MEDLINE, a review of the table of contents of 10 medical
journals (Annals of Thoracic Surgery; Cancer; CHEST; Interna-
tional Journal of Radiation Oncology, Biology, and Physics;
Journal of Clinical Oncology; Journal of Thoracic and Cardio-
vascular Surgery; Lung Cancer; Radiology; Seminars in Oncol-
ogy; and Thorax), and a review of the reference lists of other
articles. The 10 journals were chosen because they contain
approximately 75% of lung cancer research articles. Studies from
this period were excluded if they involved patients who were
subsequently included in a larger group described in a more
recent publication.
Selection Criteria
Titles and abstracts, and the full text of all articles passing the
title-and-abstract screen, were evaluated independently by at
least two of the authors for inclusion or exclusion based on five
criteria: (1) publication in a peer-reviewed journal; (2) study size
of 20 patients (except for studies involving mediastinoscopy,
which required a study size of 50 patients); (3) patient group
not included in subsequent update of the study; (4) confirmation
of mediastinal nodal biopsy results by histology at the time of
resection, or by long-term clinical follow-up ( 1 year) if the
patient did not go on to resection; and (5) availability of original
data so that sensitivities, specificities, and positive and negative
predictive values (NPVs) could be independently calculated.
Only articles meeting all five of these criteria were included for
further analysis.
Data Abstraction
Data were abstracted from included studies and tabulated
separately by type of invasive procedure (TBNA, TTNA,
EUS-NA, mediastinoscopy, or mediastinotomy). The data were
abstracted and analyzed three ways. First, abstracted data were
analyzed for all patients undergoing invasive techniques, regard-
less of the indication (eg, suspected lung cancer, lymphoma, etc.).
In this case, a definitive diagnosis of any malignancy was consid-
ered positive, and a definitive benign diagnosis was considered
negative. The limitation of this approach is that sensitivities and
specificities may vary by the diseases in the studied population,
and thus the summary results may not directly apply to a specific
population.
In the second case, the abstraction was performed for patients
suspected of having lung cancer. In this case, a definitive
diagnosis of any lung cancer (eg, NSCLC, SCLC) was considered
positive. Patients suspected of a diagnosis other than lung cancer
were excluded, where possible. This analysis assesses an invasive
technique for its ability to stage lung cancer. Patients with both
NSCLC and SCLC were included in the analyses for several
reasons: (1) separation of NSCLC and SCLC, histologically, was
often inconsistent in the 1980s; (2) some patients suspected of
having NSCLC at presentation are later found to have SCLC;
and (3) if detected early, SCLC patients can be treated surgically
and are therefore offered the same staging algorithm as NSCLC
patients for the purpose of determining treatment options.
Finally, data abstraction was performed for patients with a
confirmed diagnosis of NSCLC. In this case, a definitive diagno-
sis of NSCLC was considered positive, and any other biopsy
result was negative. This approach allows for an assessment of the
performance of an invasive test for staging patients with NSCLC.
The rationale for this analysis was that NSCLC behaves and is
usually treated differently from SCLC. However, as in the case
described above, some SCLC patients may not have been
accurately identified as such and may have been included in the
analyses.
Either tissue histologic confirmation or, if that was unavailable,
long-term clinical outcome (whether the patient survived at least
1 year with no evidence of disease) was utilized as the reference
or gold standard. If neither tissue confirmation nor clinical
outcome was available, then the patient was excluded from
further analysis.
Biopsy specimens that contained atypical, but not decidedly
malignant, cells were considered negative. Biopsy attempts that
were aborted (often due to a complication), those that could not
locate nodal tissue, and those that yielded insufficient material for
cytologic examination were considered to be negative. Patients
for whom an adequate sample was not obtained were included in
the calculations of sensitivity and specificity, as they are part of
the pool of patients on whom the procedure is being performed.
However, we excluded patients for whom a sample was not
obtained when calculating the NPV, so that the reported NPVs
estimate the probability that an adequate negative biopsy result is
truly negative.
Positive invasive clinical staging tests usually obviate the need
for therapeutic surgical exploration and resection. Therefore,
false-positive rates are often not available for these tests. All
positive biopsy data were, thus, considered true-positive results.
Biopsy results that were determined to be suspicious for
malignancy were also considered to be positive.
Statistical Analysis
Sensitivity is the percentage of people with the disease who are
detected by the test. It is calculated as the number of true-
positives divided by the sum of the number of true-positives and
false-negatives. Specificity is the percentage of people without
the disease who were correctly labeled by the test as not diseased.
It is calculated as the number of true-negatives divided by the
sum of true-negatives and false-positives. The positive predictive
value (PPV) is the likelihood that a patient with a positive test
result actually has the disease. It is calculated as true-positives
divided by the sum of true-positives and false-positives. The NPV
is the likelihood that a patient with a negative test result actually
does not have the disease. It is calculated as true-negatives
divided by the sum of true-negatives and false-negatives. Both
the PPV and the NPV vary with the prevalence of disease, which
is the frequency of disease in the population. The prevalence can
be calculated as the number of patients with either true-positive
or false-negative results divided by the total number of patients.
Summary sensitivity and specificity, and their respective con-
fidence intervals (CIs), were calculated using the Meta-Test
(New England Medical Center; Boston, MA) statistical software
for meta-analysis of diagnosis tests.
5
Summary PPV, NPV, and
prevalence were calculated based on the total number of true-
positive, false-negative, false-positive, and true-negative results
summed across studies; however, studies in which all subjects had
mediastinal disease were excluded from these calculations.
Results
As described above, data were abstracted and
analyzed three ways. However, the results obtained
by all three methods were essentially the same;
therefore, we present results for scenario two, which
assesses the ability of an invasive procedure to
adequately stage patients with suspected lung cancer
(either NSCLC or SCLC). In this scenario, patients
enrolled in the study but suspected of having a
diagnosis other than lung cancer were not consid-
ered evaluable and were excluded, where possible.
Moreover, if neither tissue confirmation nor clinical
outcome was available as reference by which the test
result is compared, then the patient was not consid-
ered evaluable and was excluded as well.
Mediastinal Staging by TBNA
Twelve studies using TBNA to stage the medias-
tinum met the inclusion criteria (Table 1).
617
They
included a total of 910 evaluable patients, of whom
906 patients (99.7%) were confirmed to have lung
Table 1Accuracy of TBNA of the Mediastinum in Patients with Lung Cancer
Source Year
No. of
Patients Technique Sensitivity Specificity PPV NPV Prevalence Notes
Garpestad et al
6
2001 22 CT fluoroscopy-
guided flexible
TBNA (2219
gauges)
0.86 * * * * NSCLC SCLC (21%);
all patients had
mediastinal
lymphadenopathy
Harrow et al
7
2000 264 Flexible TBNA
(various gauges),
TBNA (17 gauge)
0.93 1.00 1.00 0.84 0.72 NSCLC SCLC (22%)
101/188 (54%) TBNA-
negative patients
excluded because no
surgical confirmation
Bilaceroglu et al
8
1998 134 Flexible TBNA (21
gauge)
0.75 1.00 1.00 0.36 0.88 NSCLC SCLC (18%)
Rong et al
9
1998 44 CT-guided flexible
TBNA
(unspecified
gauge)
1.00 1.00 1.00 1.00 0.66 NSCLC SCLC (10%)
Wilsher et al
10
1996 21 Transtracheal rigid
TBNA
(unspecified
gauge)
0.90 * * * * NSCLC SCLC (14%)
Rodriguez de Castro
et al
11
gauge)
0.77 1.00 1.00 0.81 0.70 NSCLC SCLC (23%)
Vansteenkiste et al
12
1994 80 Transcarinal rigid
TBNA (17 gauge)
0.79 1.00 1.00 0.55 0.79 NSCLC SCLC (18%)
Schenk et al
13
vs 22 gauge)
0.91 1.00 1.00 0.82 0.86 NSCLC SCLC (27%)
Utz et al
14
1993 61 Transcarinal flexible
TBNA (cytology
vs histology
needle)
0.52 * * * * NSCLC SCLC (33%)
Schenk et al
15
gauge)
0.80 1.00 1.00 0.44 0.86 NSCLC SCLC (28%)
Ratto et al
16
1988 47 Transcarinal flexible
TBNA (21 gauge)
0.14 1.00 1.00 0.73 0.30 NSCLC SCLC (8%)
Schenk et al
17
gauge)
0.50 0.96 0.89 0.75 0.39 NSCLC SCLC (17%)
Summary 910 0.76 0.96 1.00 0.71 0.70
*Not defined because all subjects had mediastinal disease.
cancer. The overall sensitivity of TBNA was 0.76
(95% CI, 0.72 to 0.79), and the specificity was 0.96
(95% CI, 0.91 to 1.00). The average NPV was 71%
(range, 36 to 100%) and was associated with an
average prevalence of 70%. However, the prevalence
among studies varied between 30% and 88%. The
two studies with an NPV 45% each had a preva-
lence of 85%, suggesting that the NPV is under-
estimated when the prevalence is high. One study
included a large number of patients but did not
include sufficient original data to calculate the NPV
and was, therefore, excluded.
18
Another study had a
high proportion of unverified negative aspirates;
these patients were excluded from the analysis.
7
Mediastinal Staging by TTNA
Five studies using TTNA to stage the mediastinum
met the inclusion criteria (Table 2).
1923
These stud-
ies involved a total of 215 evaluable patients, of
whom 207 patients (96%) were confirmed to have
lung cancer. The overall sensitivity of TTNA was
0.91 (95% CI, 0.74 to 0.97). The overall NPV was
78% (range, 42 to 100%). As with TBNA, the NPV of
TTNA was inversely correlated with prevalence,
which was 83% overall and varied from 65 to 91%.
Mediastinal Staging by EUS-NA
Five studies met the inclusion criteria and assessed
the use of EUS-NA in mediastinal staging of 215
evaluable lung cancer patients (Table 3).
2428
When
EUS-NA was used to stage the mediastinum, the
overall sensitivity was 0.88 (95% CI, 0.82 to 0.93), and
the overall specificity was 0.91 (95% CI, 0.77 to 0.97).
The overall PPV was 98% (range, 96 to 100%), and the
overall NPV was 77% (range, 68 to 100%), with an
overall prevalence of disease of 69%(range, 63 to 79%).
There was little variability in performance characteris-
tics among the included studies.
Mediastinal Staging by Mediastinoscopy or
Mediastinotomy
Fourteen included studies, involving 5,687 evalu-
able patients, assessed mediastinal staging by stan-
dard cervical mediastinoscopy (Table 4).
2942
The
overall sensitivity was 0.81 (95% CI, 0.76 to 0.85).
The overall NPV was 91% (range, 58 to 97%), with a
prevalence of 37% (range, 21 to 54%).
In addition, two studies, assessing 206 evaluable
patients with lung cancer, met the inclusion criteria
for assessing extended cervical mediastinoscopy alone
or with standard cervical mediastinoscopy.
43,44
The
Table 2Accuracy of TTNA of the Mediastinum in Patients With Lung Cancer
Source Year
No. of
Patients Technique Sensitivity Specificity PPV NPV Prevalence Notes
Protopapas and Westcott
19
1996 32 CT-guided FNA (20
gauge)
1.00 1.00 1.00 1.00 0.91 NSCLC plus SCLC (16%)
Bo cking et al
20
gauge)
0.80 1.00 1.00 1.00 0.65 Any mediastinal disease
(98% lung cancer)
de Gregorio et al
21
1991 48 Fluoroscopy-guided FNA
(1822 gauge)
0.72 1.00 1.00 0.42 0.90 NSCLC plus SCLC plus
17 not specified
Moinuddin et al
22
gauge)
1.00 1.00 1.00 1.00 0.78 NSCLC plus SCLC (48%)
plus 9 not specified
Westcott
23
1981 72 Tomography-guided FNA
(2022 gauge)
0.94 * * * * NSCLC plus SCLC (?%)
Summary 215 0.91 1.00 1.00 0.78 0.83
Table 3Accuracy of EUS-NA of the Mediastinum in Patients With Lung Cancer
Source Year
No. of
Patients Sensitivity Specificity PPV NPV Prevalence Notes
Wiersema et al
24
2001 33 1.00 0.88 0.96 1.00 0.76 NSCLC plus SCLC (9%)
Wallace et al
25
2001 107 0.87 1.00 1.00 0.68 0.79 NSCLC plus SCLC (7%)
Fritscher-Ravens et al
26
2000 25 0.96 * * * * NSCLC plus SCLC (42%)
Gress et al
27
1997 24 0.93 1.00 1.00 0.90 0.63 NSCLC only
Silvestri et al
28
1996 26 0.88 1.00 1.00 0.82 0.65 NSCLC plus SCLC (19%)
Summary 215 0.88 0.91 0.98 0.77 0.69
addition of extended cervical mediastinoscopy to stan-
dard cervical mediastinoscopy allows the sampling of
lymph node stations 5 and 6. In the two included
studies, the overall sensitivity of standard cervical me-
diastinoscopy alone (range, 33 to 52%) and extended
cervical mediastinoscopy alone (range 45 to 51%) in-
creased (range, 69 to 76%) when the two procedures
were combined (Table 5). The slightly higher sensitivity
of the extended cervical mediastinoscopy alone com-
pared to standard cervical mediastinoscopy alone may
be due to a higher proportion of left upper lobe
primary tumors in these series. The overall NPV of
standard or extended mediastinoscopy alone ranged
from 62 to 84%, but increased to a range of 82 to 89%
when the two procedures were combined. The overall
prevalence in these two studies ranged from 29 to 42%.
Anterior mediastinotomy, or the Chamberlain pro-
cedure, is able to assess only lymph node stations 2R,
4R, and 3 when performed through the right chest,
and only lymph node stations 5 and 6 when per-
formed through the left chest, thus limiting its ability
to evaluate the mediastinum. However, anterior
mediastinotomy is an alternative approach to the
extended cervical mediastinoscopy. Two studies as-
sessed both anterior mediastinotomy alone (84 lung
cancer patients) and anterior mediastinotomy in
combination with standard cervical mediastinoscopy
(71 lung cancer patients).
38,42,45
The overall sensitiv-
ity for detecting mediastinal lymph node involve-
ment of anterior mediastinotomy alone (range, 63 to
86%) and standard cervical mediastinoscopy alone
(range, 68 to 73%) increased to 87% when the two
procedures were combined (Table 6).
Discussion
Clinical staging is based on all information ob-
tained about the primary tumor, its draining lymph
node basins, and potential distant sites of metastasis
prior to treatment initiation. Accurate clinical staging
is paramount not only for estimating a patients life
expectancy and 5-year survival probability, but also
because it directly impacts on the selection of appro-
priate therapy.
Clinical staging has been shown to differ markedly
from pathologic staging at the time of resection. In a
large prospective study, 24% of patients were clini-
cally overstaged and 20% understaged when com-
pared with pathologic staging.
46
Another study found
that when applying pathologic staging, 19% of clini-
cally staged N0 patients were actually N1, and
another 24% were N2.
47
Additionally, a study of
clinical N2 patients found that 44% were inappro-
priately staged, with 38% actually being N0 or N1,
and 6% being N3.
48
While clinical staging of N1 nodes may be difficult,
this is currently of minor importance, as therapy
remains surgical resection. In contrast, metastatic
involvement of mediastinal lymph nodes would up-
stage a lung cancer patient from the primarily surgi-
cal stages I and II to possibly surgical stage IIIA or
the primarily nonsurgical stage IIIB. Moreover,
studies demonstrating the benefits of neoadjuvant
chemotherapy and radiation therapy prior to surgical
resection of stage IIIA patients make it even more
important that patients be staged appropriately at
the time of presentation, with confirmation of the
presence of N2 (surgical stage IIIA) or the absence
of N3 nodal (nonsurgical stage IIIB) involvement.
13
While studies such as the Bimodality Lung Oncology
Trial support induction therapy prior to surgery for
patients with N1 disease, the use of neoadjuvant
chemotherapy with or without radiation therapy for
stage II patients is currently only done under proto-
col settings.
49
When selecting a test for staging a patient with
Table 4Accuracy of Standard Cervical Mediastinoscopic Biopsies in Patients With Lung Cancer
Source Year
No. of
Ebner et al
29
1999 116 0.81 1.00 1.00 0.82 0.50 NSCLC plus SCLC (11%)
Hammoud et al
30
1999 1,369 0.85 1.00 1.00 0.92 0.36 NSCLC plus SCLC (?%)
Gdeedo et al
31
1997 100 0.78 1.00 1.00 0.91 0.32 NSCLC only
De Leyn et al
32
1996 500 0.76 1.00 1.00 0.87 0.39 NSCLC only
Aaby et al
33
1995 57 0.84 1.00 1.00 0.89 0.44 NSCLC only
Dillemans et al
34
1994 331 0.72 1.00 1.00 0.84 0.41 NSCLC only
Jolly et al
35
1991 136 0.92 1.00 1.00 0.91 0.54 NSCLC plus SCLC (7%)
Riordain et al
36
1991 74 0.81 1.00 1.00 0.84 0.50 NSCLC plus SCLC (3%)
Ratto et al
37
1990 123 0.88 1.00 1.00 0.94 0.33 NSCLC only
Page et al
38
1987 345 0.73 1.00 1.00 0.80 0.48 NSCLC plus SCLC (18%)
Luke et al
39
1986 1,000 0.85 1.00 1.00 0.91 0.39 NSCLC plus SCLC (12%)
Brion et al
40
1985 153 0.67 1.00 1.00 0.85 0.35 NSCLC plus SCLC (5%)
Coughlin et al
41
1985 1,259 0.92 1.00 1.00 0.97 0.29 NSCLC plus SCLC (4%)
Deneffe et al
42
1983 124 0.68 1.00 1.00 0.88 0.31 NSCLC only
Summary 5,687 0.81 1.00 1.00 0.91 0.37
known or suspected lung cancer, two issues need to
be considered. First, one needs to select a test that
will assess the mediastinum for involved or unin-
volved nodes. This requires a test with a high
sensitivity and specificity. Second, one needs to be
able to accurately interpret the test results for an
individual patient. For example, what is the likeli-
hood that a negative test result means a patient is
free of mediastinal disease or, conversely, that a
positive test confirms metastatic disease? This issue
is addressed by the PPVs and NPVs of a test.
However, the NPVs and PPVs of a test are affected
by the prevalence of disease. Thus, patients may be
approached differently depending on the initial de-
gree of suspicion regarding mediastinal involvement.
A patient suspected of having metastases would
require a test to confirm the diagnosis with high
sensitivity. In contrast, a patient suspected of having
early stage disease would require a test to rule out
advanced disease with high specificity.
In this analysis, TTNA, EUS-FNA, and mediasti-
noscopy were all shown to have similar sensitivities.
However, the higher sensitivity demonstrated in
studies utilizing TTNA and EUS-FNA to assess the
mediastinum may be biased by the fact that these
studies are usually performed in patients for whom
there is radiographic evidence of mediastinal lymph-
adenopathy accessible by a biopsy needle. This se-
lection bias is likely to result in fewer false-negative
results, yielding a higher sensitivity. Mediastinoscopy
often is performed in the absence of enlarged lymph
nodes on radiologic imaging, yet still yields equiva-
lent sensitivity to TTNA and EUS-FNA. Despite
similar sensitivities, the NPVs are quite different.
Mediastinoscopy has an NPV of 91% vs 71 to 78%
for needle aspiration, and thus provides greater
confidence in a negative biopsy result. Additionally,
mediastinoscopy has the added advantages of being
able to obtain relatively larger specimens from a
greater number of nodal stations under direct vision
than any of the needle biopsy techniques.
There have been suggestions that a fine needle
aspiration, which is generally considered a cytologic
evaluation, and a needle biopsy, which is generally
considered to provide a core of tissue that can be
histologically evaluated, have different diagnostic
yields. Since needle aspiration is, in fact, a type of
biopsy, this review does not distinguish between the
terms needle aspiration and needle biopsy, which are
used inconsistently in the literature. Even the terms
Table 5Accuracy of Extended Cervical Mediastinoscopic Biopsies in Patients With Lung Cancer
Source Year
No. of
Standard cervical mediastinoscopy alone
Freixinet Gilart et al
43
2000 106 0.33 1.00 1.00 0.62 0.48 NSCLC plus SCLC (5%)
Ginsberg et al
44
1987 100 0.52 1.00 1.00 0.84 0.29 Left upper lobe NSCLC
only
Extended cervical mediastinoscopy alone
43
2000 106 0.51 1.00 1.00 0.69 0.48 NSCLC plus SCLC (5%)
Ginsberg et al
44
only
Standard cervical mediastinoscopy plus extended cervical mediastinoscopy
43
2000 106 0.76 1.00 1.00 0.82 0.48 NSCLC plus SCLC (5%)
Ginsberg et al
44
only
Table 6Accuracy of Anterior Mediastinotomy Biopsies in Patients With Lung Cancer
Source Year
No.
Anterior mediastinotomy alone
Best et al
45
1987 39 0.63 1.00 1.00 1.00 0.77 NSCLC plus SCLC ( 21%)
Page et al
38
1987 45 0.86 1.00 1.00 0.89 0.47 NSCLC plus SCLC (18%)
Standard cervical mediastinoscopy alone
Page et al
38
1987 345 0.73 1.00 1.00 0.80 0.48 NSCLC plus SCLC (18%)
Deneffe et al
42
1983 124 0.68 1.00 1.00 0.88 0.31 NSCLC only
Anterior mediastinotomy plus standard cervical mediastinoscopy
Page et al
38
1987 32 0.87 1.00 1.00 0.89 0.47 NSCLC plus SCLC (18%)
Deneffe et al
42
1983 39 0.87 1.00 1.00 0.92 0.38 NSCLC only
cytology needle and histology needle do not neces-
sarily specify needle size, and vice versa. Cytology
or aspiration needles usually refer to 20- to 22-
gauge needles but can be as large as 17-gauge.
Histology or core needles usually refer to 18- to
19-gauge needles but may be as small as 20-gauge or
as large as the 14-gauge Tru-cut needle. To add to
the confusion, slotted aspiration needles are pur-
ported to obtain larger specimens than standard
aspiration needles of the same gauge. The small
numbers of patients in the few studies reviewed in
this meta-analysis did not permit the comparison of
diagnostic yields of cytology needles vs histology
needles.
That said, the use of larger needles has been
reported to improve diagnostic yield but may be
more difficult.
13,15
Schenk and coworkers
13,15
re-
ported that TBNA with an 18- or 19-gauge needle
resulted in an 80 to 86% sensitivity compared to a
53% sensitivity with a 22-gauge needle. The use of
larger needles may be limited by lymph node size
and location. Smaller gauge needles were preferred
for lymph nodes 10 mm in short-axis diameter and
for lymph nodes adjacent to the heart and great
vessels.
7,23
The larger, 14-gauge Tru-cut needles
were only considered for lymph nodes with a 20-mm
minimal diameter.
20
The experience of the practitio-
ner performing the needle biopsy has a significant
impact on the diagnostic yield, even when comparing
equivalently sized lymph nodes using equivalently
sized needles.
7
Some practitioners obtain both cytol-
ogy and histology specimens, while others obtain
histology specimens only if the cytology specimens
are inadequate, which may improve the sensitivity of
the test.
13,15
Diagnostic yield may also be improved
by flushing histology needles to obtain additional
cytology specimens and by collecting any specimens
retained in the hub of cytology needles.
Determination of adequacy of the cytology speci-
men often relies on a cytopathologist being present
on-site to prepare and evaluate the specimen imme-
diately, which was often not the case in the studies
reviewed. On-site determination of cytologic ade-
quacy would allow termination of the procedure to
minimize the risk of potential complications or,
conversely, would dictate the need for additional
needle passes to improve diagnostic yield. There was
inadequate data to determine if the method of
processing cytologic specimens altered the diagnos-
tic yield, but on-site determination would be facili-
tated by immediate smearing of the specimens onto
slides, drying and fixation compared to fixation in
fluid followed by centrifugation onto slides. Schenk
and coworkers
15
have suggested that histology nee-
dles obviate the need for cytopathologic support, as
the specimens are large enough for the practitioner,
who is performing the biopsy, to grossly deter-
mine adequacy prior to formalin-fixation, paraffin-
embedding and sectioning.
The higher sensitivity of EUS-NA and TTNA
relative to TBNA is likely due to their ability to reach
more nodal stations, particularly in the inferior me-
diastinum, than TBNA, and even mediastinoscopy.
In addition, EUS-NA and TTNA both utilize imag-
ing techniques to direct the biopsy needle to the
target lymph node. Another advantage of EUS-NA is
its ability to estimate the probability of metastatic
involvement of a particular lymph node through its
echogenic pattern.
Rong and Cui
9
suggested that TBNA guidance by
standard CT may improve diagnostic yield, but they
used a very low (20%) historical baseline diagnostic
yield of nonCT-guided TBNA for comparison.
Standard CT guidance is also time-consuming and
imparts significant radiation exposure. Real-time
CT-fluoroscopic guidance,
6,50,51
endobronchial ultra-
sound guidance,
52
and virtual bronchoscopic guid-
ance (using three-dimensional CT reconstructions)
53
may improve the yield of TBNA with fewer compli-
cations, especially for small or otherwise inaccessible
lymph nodes, but these emerging imaging tech-
niques are still under investigation.
Extended cervical mediastinoscopy and anterior me-
diastinotomy were shown to have lower sensitivities and
NPVs than the other invasive tests. This may be due to
their ability to reach only aortopulmonary lymph node
stations. However, extended cervical mediastinoscopy
or anterior mediastinotomy may be a useful adjunct to
cervical mediastinoscopy, as it adds lymph node sta-
tions 5 and 6 to the number of stations that would
otherwise be evaluable by mediastinoscopy alone, as
demonstrated by the improved sensitivity and NPV of
the combined tests (Tables 5, 6).
A limitation of this current study is that the relative
performance characteristics of the reviewed invasive
procedures for staging the mediastinum have not
been established through direct comparisons. The
studies included in the analyses were not random-
ized trials, but rather prospective or retrospective
evaluations of the performance of the applied inva-
sive tests. Patient selection bias is inherent in this
study design, as the selection of patients for studies is
based on size and location of the lymph node and the
associated risks of the procedure. Without perform-
ing a randomized trial of invasive staging techniques
for patients with lung cancer, one cannot definitely
say which test is best for all patients.
In summary, the invasive clinical staging of
NSCLC can be performed effectively by several
methods, including TBNA, TTNA, EUS-NA, and
mediastinoscopy. Although at this point mediastinos-
copy appears to have the best performance charac-
teristics, selecting the correct test for a presenting
patient should incorporate an assessment of the
patients comorbid illnesses, the degree of suspicion
for metastatic disease, the location of the suspicious
nodes, the availability of a procedure at the patients
institution, and the experience and results obtained
by the performing physicians at the institution for
the considered procedure.
References
1 Rosell R, Gomez-Codina J, Camps C, et al. A randomized
trial comparing preoperative chemotherapy plus surgery with
surgery alone in patients with non-small-cell lung cancer.
N Engl J Med 1994; 330:153158
2 Roth JA, Atkinson EN, Fossella F, et al. Long-term follow-up
of patients enrolled in a randomized trial comparing periop-
erative chemotherapy and surgery with surgery alone in
resectable stage IIIA non-small-cell lung cancer. Lung Can-
cer 1998; 21:16
3 Roth JA, Fossella F, Komaki R, et al. A randomized trial
comparing perioperative chemotherapy and surgery with
surgery alone in resectable stage IIIA non-small-cell lung
4 Detterbeck FC, Jones DR, Parker LA Jr. Intrathoracic
staging. In: Detterbeck FC, Rivera MP, Socinski MA, et al,
based guide for the practicing clinician. Philadelphia, PA:
W.B. Saunders, 2001; 7393
5 Lau J. Meta-Test, version 0.6. Boston, MA: New England
Medical Center, 1997
6 Garpestad E, Goldberg S, Herth F, et al. CT fluoroscopy
7 Harrow EM, Abi-Saleh W, Blum J, et al. The utility of
transbronchial needle aspiration in the staging of broncho-
genic carcinoma. Am J Respir Crit Care Med 2000; 161:6017
8 Bilaceroglu S, Cagiotariotaciota U, Gunel O, et al. Compari-
son of rigid and flexible transbronchial needle aspiration in
the staging of bronchogenic carcinoma. Respiration 1998;
65:441449
9 Rong F, Cui B. CT scan directed transbronchial needle
aspiration biopsy for mediastinal nodes. Chest 1998; 114:
3639
10 Wilsher ML, Gurley AM. Transtracheal aspiration using rigid
bronchoscopy and a rigid needle for investigating mediastinal
masses. Thorax 1996; 51:197199
11 Rodriguez de Castro F, Rey A, Caminero J, et al. Transbron-
chial fine needle aspiration in clinical practice. Cytopathology
1995; 6:2229
Respir J 1994; 7:265268
13 Schenk DA, Chambers SL, Derdak S, et al. Comparison of
the Wang 19-gauge and 22-gauge needles in the mediastinal
staging of lung cancer. Am Rev Respir Dis 1993; 147:1251
258
1993; 104:10121016
15 Schenk DA, Strollo PJ, Pickard JS, et al. Utility of the Wang
18-gauge transbronchial histology needle in the staging of
16 Ratto GB, Mereu C, Motta G. The prognostic significance of
preoperative assessment of mediastinal lymph nodes in pa-
tients with lung cancer. Chest 1988; 93:807813
17 Schenk DA, Bower JH, Bryan CL, et al. Transbronchial
needle aspiration staging of bronchogenic carcinoma. Am Rev
Respir Dis 1986; 134:146148
18 Gasparini S, Zuccatosta L, Zitti P, et al. Integration of TBNA,
and TCNA in the diagnosis of peripheral lung nodules:
influence on staging. Ann Ital Chir 1999; 70:851855
19 Protopapas Z, Westcott JL. Transthoracic needle biopsy of
mediastinal lymph nodes for staging lung and other cancers.
Radiology 1996; 199:489496
20 Bo cking A, Klose KC, Kyll HJ, et al. Cytologic vs histologic
evaluation of needle biopsy of the lung, hilum and mediasti-
num: sensitivity, specificity and typing accuracy. Acta Cytol
1995; 39:463471
21 de Gregorio Ariza MA, Alfonso Aguiran ER, Villavieja Atance
JL, et al. Transthoracic aspiration biopsy of pulmonary and
mediastinal lesions. Eur J Radiol 1991; 12:98103
22 Moinuddin SM, Lee LH, Montgomery JH. Mediastinal nee-
dle biopsy. AJR Am J Roentgenol 1984; 143:531532
23 Westcott JL. Percutaneous needle aspiration of hilar and
mediastinal masses. Radiology 1981; 141:323329
24 Wiersema MJ, Vazquez-Sequeiros E, Wiersema LM. Evalu-
ation of mediastinal lymphadenopathy with endoscopic US-
guided fine-needle aspiration biopsy. Radiology 2001; 219:
252257
25 Wallace MB, Silvestri GA, Sahai AV, et al. Endoscopic
ultrasound-guided fine needle aspiration for staging patients
with carcinoma of the lung. Ann Thorac Surg 2001; 72:1861
1867
26 Fritscher-Ravens A, Soehendra N, Schirrow L, et al. Role of
transesophageal endosonography-guided fine-needle aspira-
tion in the diagnosis of lung cancer. Chest 2000; 117:33945
staging of lung cancer. Ann Thorac Surg 1996; 61:14411445;
discussion 14451446
29 Ebner H, Marra A, Butturini E, et al. Clinical value of
cervical mediastinoscopy in the staging of bronchial carci-
noma. Ann Ital Chir 1999; 70:873879
30 Hammoud ZT, Anderson RC, Meyers BF, et al. The current
role of mediastinoscopy in the evaluation of thoracic disease.
31 Gdeedo A, Van Schil P, Corthouts B, et al. Prospective
evaluation of computed tomography and mediastinoscopy in
mediastinal lymph node staging. Eur Respir J 1997; 10:
154751
32 De Leyn P, Schoonooghe P, Deneffe G, et al. Surgery for
non-small cell lung cancer with unsuspected metastasis to
ipsilateral mediastinal or subcarinal nodes (N2 disease). Eur
J Cardiothorac Surg 1996; 10:649654; discussion 654655
33 Aaby C, Kristensen S, Nielsen SM. Mediastinal staging of
non-small-cell lung cancer: computed tomography and cervi-
cal mediastinoscopy. ORL J Otorhinolaryngol Relat Spec
1995; 57:279285
34 Dillemans B, Deneffe G, Verschakelen J, et al. Value of
computed tomography and mediastinoscopy in preoperative
evaluation of mediastinal nodes in non-small cell lung cancer:
a study of 569 patients. Eur J Cardiothorac Surg 1994;
8:3742
Surg 1991; 102:266270; discussion 270271
36 Riordain DS, Buckley DJ, Aherne T. Mediastinoscopy as a
predictor of resectability in patients with bronchogenic car-
cinoma. Ir J Med Sci 1991; 160:291292
37 Ratto GB, Frola C, Cantoni S, et al. Improving clinical
efficacy of computed tomographic scan in the preoperative
assessment of patients with non-small cell lung cancer.
38 Page A, Nakhle G, Mercier C, et al. Surgical treatment of
bronchogenic carcinoma: the importance of staging in evalu-
ating late survival. Can J Surg 1987; 30:9699
39 Luke WP, Pearson FG, Todd TR, et al. Prospective evalua-
tion of mediastinoscopy for assessment of carcinoma of the
lung. Thorac Cardiovasc Surg 1986; 91:5356
40 Brion JP, Depauw L, Kuhn G, et al. Role of computed
tomography and mediastinoscopy in preoperative staging of
lung carcinoma. J Comput Assist Tomogr 1985; 9:4804
41 Coughlin M, Deslauriers J, Beaulieu M, et al. Role of
mediastinoscopy in pretreatment staging of patients with
primary lung cancer. Ann Thorac Surg 1985; 40:55660
42 Deneffe G, Lacquet LM, Gyselen A. Cervical mediastinos-
copy and anterior mediastinotomy in patients with lung
cancer and radiologically normal mediastinum. Eur J Respir
Dis 1983; 64:613619
43 Freixinet Gilart J, Garcia PG, de Castro FR, et al. Extended
cervical mediastinoscopy in the staging of bronchogenic
carcinoma. Ann Thorac Surg 2000; 70:16411643
44 Ginsberg RJ, Rice TW, Goldberg M, et al. Extended cervical
mediastinoscopy: a single staging procedure for bronchogenic
carcinoma of the left upper lobe. Thorac Cardiovasc Surg
1987; 94:673678
45 Best LA, Munichor M, Ben-Shakhar M, et al. The contribu-
tion of anterior mediastinotomy in the diagnosis and evalua-
tion of diseases of the mediastinum and lung. Ann Thorac
Surg 1987; 43:7881
46 Bulzebruck H, Bopp R, Drings P, et al. New aspects in the
staging of lung cancer: prospective validation of the Interna-
tional Union Against Cancer TNM classification. Cancer
1992; 70:11021110
47 Fernando HC, Goldstraw P. The accuracy of clinical evalua-
tive intrathoracic staging in lung cancer as assessed by
postsurgical pathologic staging. Cancer 1990; 65:25032506
48 Watanabe Y, Shimizu J, Oda M, et al. Aggressive surgical
intervention in N2 non-small cell cancer of the lung. Ann
Thorac Surg 1991; 51:253261
49 Pisters KM, Ginsberg RJ, Putnam JB Jr, et al. Induction
chemotherapy before surgery for early-stage lung cancer: a
novel approach. Thorac Cardiovasc Surg 2000; 119:429439
50 Goldberg SN, Raptopoulos V, Boiselle PM, et al. Mediastinal
lymphadenopathy: diagnostic yield of transbronchial medias-
tinal lymph node biopsy with CT fluoroscopic guidance;
initial experience. Radiology 2000; 216:764767
51 White CS, Weiner EA, Patel P, et al. Transbronchial needle
aspiration: guidance with CT fluoroscopy. Chest 2000; 118:
16301638
52 Shannon JJ, Bude RO, Orens JB, et al. Endobronchial
ultrasound-guided needle aspiration of mediastinal adenopa-
thy. Am J Respir Crit Care Med 1996; 153(4 pt 1):14241430
53 McAdams HP, Goodman PC, Kussin P. Virtual bronchoscopy
for directing transbronchial needle aspiration of hilar and
mediastinal lymph nodes: a pilot study. AJR Am J Roentgenol
1998; 170:13611364
2003;123;157-166 Chest
Eric M. Toloza, Linda Harpole, Frank Detterbeck and Douglas C. McCrory
Evidence
Invasive Staging of Non-small Cell Lung Cancer: A Review of the Current
& Services
Updated Information
S
References
S#BIBL
free at:
Citations
S#otherarticles
Reprints
2003;123;167-175 Chest
Silvestri
Frank C. Detterbeck, Malcolm M. DeCamp, Jr., Leslie J. Kohman and Gerard A.
Invasive Staging: The Guidelines
ISSN: 0012-3692.
Invasive Staging*
The Guidelines
Frank C. Detterbeck, MD, FCCP; Malcolm M. DeCamp, Jr., MD, FCCP;
Leslie J. Kohman, MD, FCCP; and Gerard A. Silvestri, MD, FCCP
A variety of invasive staging tests are available, including mediastinoscopy, thoracoscopy
(video-assisted thoracoscopic surgery), transbronchial needle aspiration (TBNA), transthoracic
needle aspiration (TTNA), and endoscopic ultrasound with fine needle aspiration (EUS-NA). Each
of these tests requires specific skills, has particular risks, and has technical considerations making
it more or less suitable for masses in particular locations. Therefore, direct comparisons among
the tests are not possible, and the issue is to define which procedure is most useful for a particular
situation. Invasive staging procedures are sometimes used to confirm the stage of a lung cancer,
ie, when radiographic staging is not reliable. However, invasive staging procedures are also often
used to confirm the diagnosis (ie, when the radiographic stage is reliable). The first situation
requires a test with a low false-negative rate; the latter requires a test with high sensitivity.
Clinicians must be clear about the question at hand and how to assess the value of a test when
selecting an invasive staging procedure. When confirmation of the diagnosis is the primary issue,
TBNA (or EUS-NA, if available) are good choices because of high sensitivity and low morbidity.
When the primary issue is to confirm that there is no involvement of mediastinal lymph nodes,
mediastinoscopy appears to be best suited to most situations. When the primary goal is to confirm
malignant involvement of mediastinal nodes, mediastinoscopy also appears to be best in general,
although TBNA, TTNA, and EUS-NA may be reasonable alternatives in certain situations.
However, selection of a test will also depend on the local availability of expertise, and
patient-specific anatomic and physiologic considerations. Selection of the optimal approach is
best achieved through a multidisciplinary discussion so that all aspects can be weighed
appropriately. (CHEST 2003; 123:167S175)
Key words: anterior mediastinotomy; bronchoscopy; clinical staging; esophageal ultrasound; mediastinal lymph nodes;
mediastinoscopy; pathologic staging; staging; transbronchial needle aspiration; transthoracic needle aspiration
Abbreviations: APW aortopulmonary window; EUS-NA endoscopic ultrasound and needle aspiration;
FN false-negative; FP false-positive; LUL left upper lobe; NSCLC non-small cell lung cancer;
PET positron emission tomography; SCLC small cell lung cancer; TBNA transbronchial needle aspiration;
TTNA transthoracic needle aspiration; VATS video-assisted thoracic surgery
T
he workup of patients suspected to have a
malignancy consists of several distinct steps (Ta-
ble 1). The paradigm that a diagnosis must be
obtained first, followed by appropriate staging and
treatment, is simplistic and does not correspond to
actual practice. In reality, the first step is an initial
assessment based on the clinical presentation, risk
factors, and radiographic appearance. This allows a
presumptive diagnosis to be made as well as a
presumptive stage. Sometimes, the presumptive di-
agnosis and stage can be made with sufficient reli-
ability that a plan for treatment can be initiated (such
as thoracotomy for a new spiculated parenchymal
mass in a smoker with no enlarged nodes). However,
usually confirmatory tests are necessary as the sec-
ond step to prove or disprove the presumptive diag-
nosis and stage. It is the presumptive diagnosis and
stage, however, that determines which confirmatory
test is most appropriate. Accurate definition of the
diagnosis and stage then allows the optimal treatment
to be selected.
This section deals with invasive staging proce-
dures. Such procedures may be necessary to confirm
*From the Multidisciplinary Thoracic Oncology Program (Dr.
Detterbeck), Division of Cardiothoracic Surgery, University of
North Carolina, Chapel Hill, NC; Department of Thoracic and
Cardiovascular Surgery (Dr. DeCamp), Cleveland Clinic Foun-
dation, Cleveland, OH; Division of Cardiothoracic Surgery (Dr.
Kohman), SUNY Health Science Center, Syracuse, NY; and
Division of Pulmonary and Critical Care Medicine (Dr. Silvestri),
Medical University of South Carolina, Charleston, SC.
The work for these guidelines was performed at the University of
North Carolina at Chapel Hill.
Correspondence to: Frank C. Detterbeck, MD, Division of Car-
diothoracic Surgery, Department of Surgery, University of North
Carolina at Chapel Hill, CB #7065, 108 Burnett-Womack Build-
ing, Chapel Hill, NC 27599-7065; e-mail: fdetter@med.unc.edu
the presumptive stagefor example, in instances
where the clinical staging based on imaging studies
alone is often inaccurate. Invasive staging proce-
dures may also be done primarily to confirm the
diagnosisfor example, in instances where the clin-
ical stage by radiologic tests is considered to be
accurate (ie, small cell lung cancer [SCLC]), but
tissue confirmation of the cancer is necessary before
treatment can be initiated. Because both of these
questions are clinically important, invasive staging
tests will be reviewed with regard to both of these
questions (confirmation of stage vs confirmation of
diagnosis). However, it is important not to use data
that pertain to one of these questions out of context
and apply it to patients in whom a different issue is
to be addressed.
It is difficult to compare the results of one staging
test to another. First, the clinical question being
addressed by an invasive staging procedure varies.
Second, the patient population for whom one clinical
question is important is different from the group for
whom another question is relevant. Finally, there are
technical issues making certain procedures applica-
ble only in specific situations. These considerations
must be kept in mind when comparing various
invasive staging procedures.
Various parameters can be used to assess the
reliability of a test, including sensitivity, specificity,
false-negative (FN), and false-positive (FP) rates.
Sensitivity and specificity relate to theoretical patient
populations, all of which either have or do not have
the condition in question. Therefore, these parame-
ters are defined after the fact by the true disease
status. While there is some theoretical appeal to this
approach, it makes it more difficult to apply prospec-
tively to patients (in whom the true disease status is
not known). These measures are most useful in
choosing which confirmatory test to perform, be-
cause in theory these measures can be compared for
different tests since they deal with the same theoret-
ical population. However, in reality, comparison of
the sensitivity and specificity of different invasive
tests is sometimes difficult, because the patient
populations selected for one test or another are not
the same. Once a test has been performed, however,
interpretation of the results is achieved by consider-
ing the FP and FN rates. These measures are not
abstract because they are based on real populations
of patients, who either have a positive or a negative
test result. The FN and FP rates are sometimes
expressed in a more obtuse manner as the percent-
age of 1 minus the FN (or FP) rate, known as the
negative predictive value and positive predictive
value.
It is important to be clear about the clinical
context when discussing the stage of a patient. Five
distinct contexts are recognized by the American
Joint Committee on Cancer (Table 2). Ideally, we
would like to have information about the presence or
absence of tumor at a cellular level throughout the
body in order to accurately know the true stage. In
general, the pathologic stage is viewed as the closest
approximation to the true stage. The pathologic stage
is established by a surgical resection, including accu-
rate assessment of potential areas of spread (such as
lymph nodes), and careful examination of the tissues
by a pathologist. While this may be the most accu-
rate, it is not available until surgical treatment has
been carried out. The clinical stage is that deter-
mined using all information available prior to any
treatment, and thus is the most useful in actual
practice. The information available may be limited
(ie, involving only a chest CT) or extensive (ie,
involving invasive procedures). It is important to
note that an invasive staging procedure is still con-
sidered part of clinical staging, even though it may
involve a surgical procedure (ie, mediastinoscopy)
and evaluation by a pathologist.
Techniques of Invasive Mediastinal
Staging
A number of invasive tests are available to stage
the mediastinum (Table 3). These include mediasti-
noscopy, the Chamberlain procedure (also known as
Table 1Steps in the Workup of a Patient With
Suspected Lung Cancer
Step 1 (based on clinical evaluation, risk factors, CT appearance)
Presumptive diagnosis
Presumptive cell type (SCLC vs NSCLC)
Presumptive stage
Step 2 (based on further radiologic imaging and invasive tests)
Confirmation of diagnosis
Confirmation of stage
Step 3 Treatment
Surgery
Radiotherapy
Chemotherapy
Multimodality therapy
Table 2Clinical Staging Contexts
Prefix Clinical Context
p Pathologic
c Clinical
y Restaging after initial therapy
r Staging at time of recurrence
a Autopsy
an anterior mediastinotomy), transthoracic needle
aspiration of mediastinum (TTNA), transbronchial
needle aspiration (TBNA), esophageal endoscopic
ultrasound with needle aspiration (EUS-NA), and
video-assisted thoracic surgery (VATS), which is also
known as thoracoscopy. Each of these is discussed
individually first, followed by a discussion of which
test to select for particular clinical questions.
Mediastinoscopy
Mediastinoscopy has long been the gold stan-
dard among staging tests of mediastinal lymph
nodes. It is performed in the operating room, usually
under general anesthesia, and in most US centers
patients are usually discharged the same day if they
are in stable condition after several hours of obser-
vation.
13
The procedure involves an incision just
above the suprasternal notch, insertion of a medias-
tinoscope alongside the trachea, and biopsy of me-
diastinal nodes. Rates of morbidity and mortality as a
result of this procedure are low (2% and 0.08%,
respectively).
4
Left and right high and low paratra-
cheal nodes (stations 2R, 2L, 4R, 4L), pretracheal
nodes (stations 1, 3), and anterior subcarinal nodes
(station 7) are accessible via this approach. Node
groups that cannot be sampled with this technique
include posterior subcarinal (station 7), inferior me-
diastinal (stations 8, 9), and aortopulmonary window
and anterior mediastinal (stations 5, 6) nodes.
The average sensitivity of mediastinoscopy to de-
tect mediastinal node involvement from cancer is
approximately 80 to 85%, and the average FN rate is
approximately 10% (see the chapter, Invasive Stag-
ing for NSCLC: A Review of the Current Evidence,
hereafter referred to as the background evidence
chapter). Several authors have shown that approxi-
mately half (42 to 57%) of the FN cases were due to
nodes that were not accessible by the mediastino-
scope.
59
The FN rate at mediastinoscopy is probably
also affected by the diligence with which nodes are
dissected and sampled at mediastinoscopy. Ideally,
five nodal stations (stations 2R, 4R, 7, 4L, 2L) should
routinely be examined, with at least one node sam-
pled from each station unless none are present after
actual dissection in the region of a particular node
station. The specificity and the FP rates of medias-
tinoscopy are reported to be 100% and 0, respec-
tively. Strictly speaking, these values cannot really be
assessed because patients with a positive biopsy
result were not subjected to any further procedures
(such as thoracotomy) to confirm the results. Never-
theless, it seems reasonable to assume that the FP
rate is low.
The patients included in these series have had
potentially operable, nonmetastatic lung cancer with
very few exceptions. The reported results provide data
regarding the reliability of mediastinoscopy for staging
of the mediastinal nodes as compared to thoracotomy
in patients with lung cancer. Therefore, these results
can be readily applied to potentially resectable patients
to confirm invasively the stage of lung cancer.
Assessment of Aortopulmonary Window
Lymph Nodes
Cancers in the left upper lobe (LUL) have a
predilection for involvement of the nodes in the
aortopulmonary window (APW) [station 5]. These
nodes are classified as mediastinal nodes, and repre-
sent the most important group of N2 nodes that are
not accessible by standard cervical mediastinoscopy.
It has been suggested that nodes in this region
should not be viewed as mediastinal nodes and that
resection of patients should be performed regardless
of APW node involvement, making assessment of
these nodes superfluous.
10
This was based on a
selected subgroup of 23 completely resected patients
who had APW node involvement as the only site of
N2 disease. However, analysis of all of the data in
this regard shows that survival of patients with only
APW node involvement is not substantially different
than that of patients with involvement of only a
single N2 node station in another location.
11
There-
fore, the issue is more a matter of whether patients
with involvement of a single mediastinal node station
should undergo surgical resection, and not whether
APW nodes should be classified as N2 nodes.
The classic way of invasively assessing this area is
the Chamberlain procedure (also known as an ante-
rior mediastinotomy), which involves an incision in
the second or third intercostal space just to the left of
the sternum. Traditionally, an overnight hospital stay
was necessary, but in many institutions this is no
longer found to be necessary, especially as surgeons
have used visualization between the ribs more fre-
quently as opposed to removal of a costal cartilage.
The reliability of this procedure has not been exten-
sively documented, despite its common use. The
sensitivity of the Chamberlain procedure in addition
to a standard cervical mediastinoscopy in patients
with LUL tumors is approximately 87%, and the FN
rate is approximately 10% (see background evidence
Table 3Techniques of Invasive Mediastinal Staging
Mediastinoscopy
Chamberlain procedure
Transthoracic needle aspiration
Transbronchial needle aspiration
Endoscopic ultrasound and needle aspiration
Video-assisted thoracic surgery staging
chapter). Two additional studies regarding this pro-
cedure have not really addressed the reliability of the
procedure for staging of non-small cell lung cancer
(NSCLC). In one study, no actual biopsies were
performed in most patients, and the procedure was
used to assess resectability (resectable patients in-
cluded those with bulky APWnodal involvement in this
series).
12
The other study used anterior mediastinot-
omy primarily for diagnosis (not staging), and included
pulmonary biopsies and evaluation of patients with
mediastinal masses.
13
In fact, only a minority of patients
included in this study had lung cancer.
Extended cervical mediastinoscopy offers an alter-
native way of invasive assessment of APW nodes, but
is used in only a few institutions (see background
evidence chapter). With this procedure, a mediasti-
noscope is inserted through the suprasternal notch
and directed lateral to the aortic arch.
14
In 100
consecutive patients with LUL cancers, standard
mediastinoscopy and extended mediastinoscopy
were found to have a sensitivity of 69% and an FN
rate of 11% for detection of N2,3 disease (preva-
lence, 29%).
14
Similar results (sensitivity, 81%; FN
rate, 9%) were reported in another series of 93 such
patients, all of whom had enlarged APW nodes.
15
In
approximately 550 patients undergoing extended
cervical mediastinoscopy, two major complications (one
stroke and one aortic injury) have been reported.
1418
Thoracoscopy has been used to assess some node
stations not accessible by mediastinoscopy (stations
5, 6, 7, 8, 9). In 40 patients with enlarged APW or
subazygous nodes who had a negative cervical medi-
astinoscopy results, thoracoscopy provided com-
pletely accurate staging in all, with no FN results.
19
The patients included in these series of extended
cervical mediastinoscopy or thoracoscopy have had
potentially operable lung cancer with very few ex-
ceptions. The reported results provide data regard-
ing the reliability of these tests for staging of medi-
astinal nodes as compared to thoracotomy in patients
with lung cancer. Therefore, these results can be
readily applied to potentially resectable patients to
confirm invasively the stage of lung cancer.
EUS-NA also provides an alternative method of
sampling APW nodes (see subsequent discussion).
Data addressing the reliability of this procedure
specifically for APW nodes in patients with LUL
tumors is not available. In general, however, the
sensitivity of this test is very high, although the FN
rate is high enough to potentially be an issue.
Transthoracic Needle Aspiration
TTNA or biopsy for diagnosis and staging of the
mediastinum is distinct from TTNA of parenchymal
masses to achieve a diagnosis. The ability to carry out
TTNA for diagnosis and staging of the mediastinum
has generally been reported to be high ( 90%),
although approximately 10% of patients require
placement of a catheter for evacuation of a pneumo-
thorax.
20
The sensitivity has generally been reported
to be approximately 90% (see background evidence
chapter). However, the FN rate is 20 to 50%,
indicating that a negative (noncancer) result cannot
be relied on in a patient suspected of having cancer.
Patients selected for this procedure have generally
had mediastinal lymph nodes 1.5 cm. This is also
evidenced by the fact that the prevalence of cancer in
the mediastinal nodes was very high (80%). Further-
more, only approximately 75% of the patients had lung
cancer (despite excluding studies in which only a
minority of patients had lung cancer). Therefore, these
results are most applicable to patients with bulky
mediastinal involvement, in whom the purpose of the
procedure was probably primarily to confirm the diag-
nosis and less likely to confirm the stage. Extrapolation
of these results to patients with lesser amounts of
mediastinal spread may be inappropriate.
Transbronchial Needle Aspiration
TBNA, also known as Wang needle aspiration, can
be performed safely with no significant morbidity. It
can be performed as an outpatient procedure, as
with most bronchoscopic procedures. TBNA is used
most frequently to assess subcarinal nodes. Paratra-
cheal lymph nodes are sometimes more difficult to
access, due to an inability to sufficiently angulate the
bronchoscope and the needle. It is reported that it is
feasible to get adequate specimens via TBNA in
approximately 90% of cases.
2124
The sensitivity is
approximately 75%, whereas the specificity is very
high (see background evidence chapter). The sensi-
tivity has been consistently reported to be high in
series with a very high prevalence of N2,3 involve-
ment, whereas it is low in series with a lower
prevalence. The FN rate is approximately 30%, and
occasional FP results have been reported in series in
which this has been specifically examined with a
confirmatory test.
The patients included in studies of TBNA have
generally had a very high prevalence of N2,3 involve-
ment, and the general implication is that the medi-
astinal nodes have been markedly enlarged, although
few studies have reported specifics about node size.
The results should not be applied to patients without
extensive mediastinal involvement. In fact, the stud-
ies show that the sensitivity in these situations is much
lower. Furthermore, the high FN rate makes this test
less useful for staging of the mediastinum; its primary
role is to provide a diagnosis in patients in whom there
is already little doubt about mediastinal involvement
based on the radiographic appearance alone.
Endoscopic Ultrasound and Needle Aspiration
EUS-NA of mediastinal lymph nodes through the
wall of the esophagus has been performed with a
negligible risk of infection or bleeding. Only one
complication (transient fever) has been reported
among five studies involving 288 patients.
2529
This
technique is particularly useful for inferior pulmo-
nary ligament, subcarinal, and APW nodes. Nodes
that are anterolateral to the trachea (2R, 2L, 4R,
4L) are difficult to sample reliably (but are more
commonly involved with lung cancer). There are no
data regarding the feasibility of EUS-NA. The re-
ported sensitivity is approximately 90% when a nee-
dle aspiration is performed. The FN rate of 23% is a
potential drawback (see background evidence chap-
ter). No data are available regarding specificity and
FP rates from studies that involved an appropriate
confirmatory test.
The patients included in these studies have been
patients with NSCLC without evidence of distant
metastases. However, most of the patients have had
markedly enlarged lymph nodes, primarily in loca-
tions amenable to EUS-NA, although it is clear that
nodes that are smaller than 1 cm can be sampled
using this technique.
26,29
Therefore, these results per-
tain primarily to patients with radiographic evidence of
N2,3 involvement in whom the prevalence of such
involvement is expected to be high. This procedure
requires a skilled endoscopist with the necessary equip-
ment, which is currently available in only a small
number of institutions.
Video-Assisted Thoracic Surgery
Insufficient data are available to draw conclusions
about the feasibility and reliability of VATS in assess-
ing the mediastinum. In general, VATS is limited to
an assessment of only one side of the mediastinum
(unless a bilateral VATS is performed).
Other Staging Procedures
In patients with signs of advanced disease, clinical
scenarios often occur that make other invasive pro-
cedures indicated, such as needle aspiration of a
supraclavicular lymph node, thoracentesis or thora-
coscopy of a pleural effusion, or needle aspiration or
biopsy of a metastatic site such as an enlarged
adrenal or hepatic mass. The indications for such
procedures are covered in more detail in the chap-
ters on diagnosis and noninvasive staging, and spe-
cific recommendation regarding such procedures
can be found in these chapters as well. In brief, if an
enlarged supraclavicular lymph node or a pleural
effusion is present, it is generally prudent to pursue
a diagnosis of these lesions. When the clinical pre-
sentation is entirely consistent with locally advanced
disease (stage IIIB), these procedures are usually
indicated because they represent the easiest way to
confirm the diagnosis of lung cancer. When the
clinical presentation is otherwise not consistent with
locally advanced disease, the etiology of these lesions
must be established in order to accurately define the
stage. However, the procedures used to diagnose an
enlarged supraclavicular node (needle aspiration or
surgical biopsy) are the same regardless of whether
the issue is to confirm the diagnosis or to define the
stage. Similarly, in patients with a clinical presenta-
tion consistent with advanced disease (stage IV), an
invasive procedure may be indicated as the easiest
way to confirm the diagnosis and establish the cell
type of lung cancer. In patients with a solitary site
that is suspicious for a distant metastasis, or in
patients with a clinical presentation that seems in-
consistent with advanced disease, an invasive proce-
dure is indicated to accurately define the stage. The
procedures used to assess possible distant sites are
the same regardless of the clinical presentation, and
are dictated primarily by technical and anatomic
factors specific to the particular patient.
No data are available to assess the sensitivity,
specificity, and FN and FP rates of needle aspiration
of a supraclavicular node. General experience indi-
cates that this procedure is usually successful; in
addition, surgical biopsy of such a node is easily
accomplished if a needle aspiration is not diagnostic.
The reliability of procedures to diagnose a pleural
effusion is covered in the chapter on diagnosis.
Thoracentesis has a sensitivity of approximately 60%;
thoracoscopy has a sensitivity of 95%. Procedures
to diagnose suspected distant metastatic sites are too
varied to discuss in detail; furthermore, no data are
available that expressly assesses the reliability of
these tests in patients with lung cancer.
Clinical Questions
Comparison of tests for mediastinal staging is
difficult because the question being addressed is not
always the same for patients undergoing one proce-
dure vs another. For example, invasive staging may
be performed primarily to confirm the diagnosis in a
patient in which there is little doubt as to the
presence of mediastinal involvement. In other pa-
tients, the issue may be to confirm that the medias-
tinum is not involved (ie, in instances where the FN
rate of radiologic staging alone is unacceptably high),
or to prove that mediastinal involvement is present
(ie, in instances where the FP rate of radiologic
staging alone is unacceptably high).
The population of patients undergoing one type of
invasive procedure is usually different from the
population undergoing another invasive test. For
example, patients undergoing mediastinoscopy often
have radiographically normal-appearing mediastinal
lymph nodes, whereas those who undergo TBNA
almost exclusively have enlarged nodes. Further-
more, a particular test may be selected because the
lymph nodes in question are easily accessible by that
technique and not by others. These issues must be
kept in mind when evaluating invasive staging tests.
A summary of some of the most important features
of invasive tests is shown in Table 4.
When the clinical question is to confirm the
presumptive diagnosis of lung cancer and establish
the cell type (ie, the presence of mediastinal node
involvement is already clear based on radiographic
imaging), the determining factors in selection of a
test will be the sensitivity, the ability to perform the
procedure in that particular patient, and the morbid-
ity of the procedure. Procedures that can be per-
formed on an outpatient basis and that carry a low
risk of complications have a definite advantage.
Mediastinoscopy and the Chamberlain procedure
are less desirable because of a risk (albeit low) of
significant complications. TTNA carries a relatively
high risk of a pneumothorax requiring a tube for
evacuation. These considerations make TBNA or
EUS-NA good choices. The sensitivity of EUS-NA is
particularly appealing. However, in the end, despite
these considerations, the determining factors in se-
lecting a diagnostic test are likely to be the local
availability of expertise with particular procedures, as
well as the location of suspicious nodes and their
accessibility to biopsy by a particular technique.
However, in many instances in which there is
mediastinal node enlargement, it is necessary to
more accurately stage the mediastinum due to a high
FP rate of the CT. In this situation both the sensi-
tivity of the test as well as a low FN rate are
important. Furthermore, it is usually also important
to assess a number of mediastinal node stations in
order to have a detailed assessment of the extent of
mediastinal disease in these situations. Mediastinos-
copy provides high sensitivity, a low FN rate, and the
ability to assess most of the clinically important node
stations. EUS-NA, if available, is a reasonable alter-
native whenever the nodes in question are accessible
to this technique. TBNA is less appealing due to the
lower sensitivity, but may be reasonable in some
situations, particularly if bronchoscopy is indicated in
the patients for other reasons. TTNA is also less
appealing because of the risk of pneumothorax. It is
important to note, however, that because of the high
FN rate of EUS-NA, TBNA, or TTNA, a negative
result of these tests in a patient with radiographically
suspicious nodes should be followed up with further
tests such as mediastinoscopy.
In situations where the major question is to con-
firm that there is no malignant involvement of
mediastinal nodes, a test with a low FN rate is
needed. In addition, a test is needed that allows as
thorough an assessment of mediastinal node stations
as possible. In this case, TTNA, TBNA, and EUS-NA
have little role because of the high FN rates and poor
ability to assess all of the mediastinal node stations.
The gold standard is still mediastinoscopy, which
allows sampling of multiple nodal stations in a
reliable fashion. The exceptions are nodes in the
inferior pulmonary ligament (which are rarely in-
volved, even in the case of lower lobe tumors), APW
nodes (which can be accessed via the Chamberlain
procedure in the case of LUL tumors), and occasion-
ally posterior subcarinal nodes. The same arguments
in favor of mediastinoscopy apply when the issue is
to rule out involvement of contralateral lymph nodes,
for example, if patients with stage IIIA but not stage
IIIB NSCLC are considered eligible for a trimodality
treatment approach.
How positron emission tomography (PET) scan-
ning should be integrated into the diagnosis and
staging of lung cancer is still being defined. This is
discussed further in the chapter on noninvasive
staging. In general, the currently available data
Table 4Comparison of Characteristics of Invasive Tests
Tests Sensitivity, % Specificity, %* FP Rate, %* FN Rate, %
Patient
Population
Mediastinoscopy 81 100 0 9 Clinical N02
Chamberlain procedure 87 100 0 15 Clinical N02
TTNA 91 100 0 22 Clinical N2
EUS-NA 88 91 2 23 Clinical N2
TBNA 76 96 0 29 Clinical N2
*Results for specificity and FP rates are not true results because most patients did not undergo a confirmatory test in the case of a positive result.
In addition to a cervical mediastinoscopy for LUL tumors.
indicates that a PET scan that is positive in the
mediastinum should be confirmed. It would stand to
reason that the confirmatory test of a positive PET
scan be one that provides access to the node station
or stations that were positive on the PET scan.
Furthermore, the confirmatory test should have a
low FN rate. Based on these considerations, medi-
astinoscopy appears to be generally the best invasive
test. It is unclear whether and under what circum-
stances a PET scan that is negative in the mediasti-
num should be confirmed (see discussion in the
noninvasive staging chapter). If confirmation of a
negative mediastinal PET is deemed necessary, me-
diastinoscopy is usually the best choice because of
the low FN rate.
Recommendations
1. In patients who have extensive mediastinal
infiltration with tumor (T4 involvement or in-
volvement to the point of not being able to see
discrete lymph nodes), the primary goal of an
invasive procedure is to provide confirmation of
the diagnosis. (The radiographic staging of me-
diastinal node involvement is compelling.)
For these patients, TTNA and EUS-NA are
procedures of choice based on high sensitivity
(approximately 90%) and low morbidity (outpa-
tient procedure). Level of evidence, fair; ben-
efit, moderate; grade of recommendation, B
For these patients, TBNA is an alternative
with appropriately located mediastinal involve-
ment, but in general has lower sensitivity (ap-
proximately 75%) and occasional FP results.
Level of evidence, fair; benefit, moderate;
For these patients, mediastinoscopy is least
useful because of higher morbidity than EUS-
NA, TTNA, and TBNA (although it is also an
outpatient procedure). Level of evidence, fair;
2. In patients suspected of having NSCLC, who
have no evidence of distant metastases, and
who have enlarged, discrete mediastinal nodes
by CT (because of a high FP rate of CT),
mediastinoscopy is the invasive procedure of
choice to rule in mediastinal node involvement.
This recommendation is based on the ability
of mediastinoscopy to stage most of the com-
monly involved mediastinal node stations with a
presumed low FP rate, a low FN rate (approx-
imately 10%) and low morbidity (2%; outpa-
efit, substantial; grade of recommendation, B
For these patients, TBNA, TTNA, and
EUS-NA are alternatives to mediastinoscopy,
but result in less thorough mediastinal staging
because of difficulty in assessing as many node
stations and because of a higher FN rate. Level
of evidence, fair; benefit, moderate; grade of
recommendation, B
For the subset of these patients with LUL
cancer, the Chamberlain procedure, extended
cervical mediastinoscopy, EUS-NA, or thora-
coscopy to evaluate the APW nodes should also
be performed if other mediastinal node stations
are found to be uninvolved. Level of evidence,
fair; benefit, moderate; grade of recommenda-
tion, B
who have normal mediastinal nodes by CT, but in
whom invasive staging of the mediastinum is
recommended (because of a high FN rate of CT),
choice to rule out mediastinal node involvement.
This recommendation is based on the ability of
mediastinoscopy to stage most of the commonly
involved mediastinal node stations with a low FN
rate (approximately 10%) and low morbidity (2%;
EUS-NAare not recommended because of a high
FN rate. Level of evidence, fair; benefit, none/
negative; grade of recommendation, D
For the subset of these patients who have a
LUL cancer, the Chamberlain procedure, ex-
tended cervical mediastinoscopy, or thoracoscopy
should be additionally performed to evaluate the
APW nodes. Level of evidence, fair; benefit,
4. In patients with a PET scan that is positive in
the mediastinum, confirmation should be ob-
tained by an invasive test that allows sampling
of the PET-positive nodes with a high sensitiv-
ity and a low FN rate. In general, mediastinos-
copy is likely to be the best test. Level of
ommendation, B
5. In patients with a PET scan finding that is
negative in the mediastinum, in whom confir-
mation of the absence of mediastinal involve-
ment is deemed desirable, mediastinoscopy is
generally the best test because of a low FN
rate. Level of evidence, poor; benefit, moder-
ate; grade of recommendation, C
Summary
In clinical practice, the diagnosis and staging of
patients suspected of having lung cancer are closely
linked and often proceed simultaneously. Invasive
tests are sometimes used to confirm the presumed
diagnosis, while at other times they are used to
confirm the presumed stage. It is important to be
clear about what the primary question is when
assessing the value of invasive tests for a particular
clinical scenario.
In patients with extensive mediastinal infiltration,
the primary issue is confirmation of the diagnosis,
because the stage is clearly defined radiographically.
Sputum analysis, bronchoscopy (discussed in more
detail in another chapter), TTNA, EUS-NA, and
possibly TBNA are generally the most useful invasive
tests. In patients without distant metastases but with
discrete, enlarged mediastinal lymph nodes, the pri-
mary goal is to confirm the stage, because of the high
FP rate of this finding on a CT scan. In general
mediastinoscopy is the best invasive procedure, be-
cause it allows a thorough staging to be carried out.
TBNA, TTNA, and EUS-NA may be an alternative,
depending on the location of the lymph nodes and
the availability of expertise with these procedures.
In patients without distant metastases and with
normal mediastinal nodes by CT scan, but who have
a central tumor, an adenocarcinoma or evidence of
hilar node involvement, the issue is also to confirm
the presumed stage because of a high FN rate of the
CT scan for mediastinal node involvement in these
settings. The primary role of an invasive test is to rule
out malignant involvement. Mediastinoscopy is the
procedure of choice, because of the low FN rate of
this procedure.
The most appropriate use of PET scanning for
staging of the mediastinum is still being defined. A
positive PET scan result in the mediastinum should
be confirmed by a test that allows access to the
suspicious nodes. It is controversial whether a neg-
ative PET scan result in the mediastinum is adequate
in settings with a relatively high rate of mediastinal
involvement despite a negative CT (central tumor,
adenocarcinoma, or evidence of hilar involvement).
The FN rate of PET in this specific setting has not
been defined. If confirmation of a negative PET scan
result in the mediastinum is necessary, mediastinos-
copy is the best test.
1. In patients who have extensive mediastinal
infiltration with tumor (T4 involvement or in-
volvement to the point of not being able to see
discrete lymph nodes), the primary goal of an
invasive procedure is to provide confirmation of
the diagnosis. (The radiographic staging of me-
diastinal node involvement is compelling.)
For these patients, TTNA and EUS-NA are
procedures of choice based on high sensitivity
(approximately 90%) and low morbidity (outpa-
For these patients, TBNA is an alternative
with appropriately located mediastinal involve-
ment, but in general has lower sensitivity (ap-
proximately 75%) and occasional FP results.
For these patients, mediastinoscopy is least
useful because of higher morbidity than EUS-
NA, TTNA, and TBNA (although it is also an
who have enlarged, discrete mediastinal nodes
by CT (because of a high FP rate of CT),
choice to rule in mediastinal node involvement.
This recommendation is based on the ability
of mediastinoscopy to stage most of the com-
monly involved mediastinal node stations with a
presumed low FP rate, a low FN rate (approx-
imately 10%), and low morbidity (2%; outpa-
tient procedure). Level of evidence, fair; benefit,
EUS-NA are alternatives to mediastinoscopy, but
result in less thorough mediastinal staging be-
cause of difficulty in assessing as many node
stations and because of a higher FN rate. Level of
evidence, fair; benefit, moderate; grade of recom-
mendation, B
For the subset of these patients with a LUL
cervical mediastinoscopy, EUS-NA, or thoracos-
copy to evaluate the APW nodes should also be
performed if other mediastinal node stations are
found to be uninvolved. Level of evidence, fair;
have no evidence of distant metastases, and who
have normal mediastinal nodes by CT, but in
whom invasive staging of the mediastinum is
recommended (because of a high FN rate of CT),
choice to rule out mediastinal node involvement.
This recommendation is based on the ability of
mediastinoscopy to stage most of the commonly
involved mediastinal node stations with a low FN
rate (approximately 10%) and low morbidity (2%;
EUS-NAare not recommended because of a high
FN rate. Level of evidence, fair; benefit, none/
For the subset of these patients who have LUL
cervical mediastinoscopy, or thoracoscopy should
be additionally performed to evaluate APW
nodes. Level of evidence, fair; benefit, moderate;
4. In patients with a PET scan result that is positive
in the mediastinum, confirmation should be ob-
tained by an invasive test that allows sampling of
the PET-positive nodes with a high sensitivity and
a low FN rate. In general, mediastinoscopy is
likely to be the best test. Level of evidence, fair;
5. In patients with a PET scan result that is
negative in the mediastinum, in whom confir-
mation of the absence of mediastinal involve-
ment is deemed desirable, mediastinoscopy is
generally the best test because of a low FN
rate. Level of evidence, poor; benefit, moder-
ate; grade of recommendation, C
References
1 Cybulsky IJ, Bennett WF. Mediastinoscopy as a routine
outpatient procedure. Ann Thorac Surg 1994; 58:176178
2 Vallie` res EA, Page A, Verdant A. Ambulatory mediastinos-
copy and anterior mediastinotomy. Ann Thorac Surg 1991;
52:11221126
3 Selby JH Jr, Leach CL, Heath BJ, et al. Local anesthesia for
mediastinoscopy: experience with 450 consecutive cases. Am
Surg 1978; 44:679682
4 Kiser AC, Detterbeck FC. General aspects of surgical treat-
ment. In: Detterbeck FC, Rivera MP, Socinski MA et al, eds.
Diagnosis and treatment of lung cancer: an evidence-based
guide for the practicing clinician. Philadelphia, PA: W.B.
Saunders, 2001; 133147
5 Coughlin M, Deslauriers J, Beaulieu M, et al. Role of
mediastinoscopy in pretreatment staging of patients with
primary lung cancer. Ann Thorac Surg 1985; 40:556560
6 Staples CA, Muller NL, Miller RR, et al. Mediastinal nodes in
bronchogenic carcinoma: comparison between CT and medi-
astinoscopy. Radiology 1988; 167:367372
7 Gdeedo A, Van Schil P, Corthouts B, et al. Prospective evalua-
tion of computed tomography and mediastinoscopy in mediasti-
nal lymph node staging. Eur Respir J 1997; 10:15471551
8 Van den Bosch JMM, Gelissen HJ, Wagenaar SS. Exploratory
thoracotomy in bronchial carcinoma. Thorac Cardiovasc Surg
1983; 85:733737
9 Hammoud ZT, Anderson RC, Meyers BF, et al. The current
role of mediastinoscopy in the evaluation of thoracic disease.
10 Patterson GA, Piazza D, Pearson FG, et al. Significance of
metastatic disease in subaortic lymph nodes. Ann Thorac Surg
1987; 43:155159
11 Detterbeck FC, Jones DR. Surgical treatment of stage IIIa
(N2) non-small cell lung cancer. In: Detterbeck FC, Rivera
MP, Socinski MA et al, eds. Diagnosis and treatment of lung
cancer: an evidence-based guide for the practicing clinician.
Philadelphia, PA: W.B. Saunders, 2001; 244256
12 Jiao J, Magistrelli P, Goldstraw P. The value of cervical
mediastinoscopy combined with anterior mediastinotomy in
the preoperative evaluation of bronchogenic carcinoma of the
left upper lobe. Eur J Cardiothorac Surg 1997; 11:450454
13 Best LA, Munichor M, Ben-Shakhar M, et al. The contribu-
tion of anterior mediastinotomy in the diagnosis and evalua-
tion of diseases of the mediastinum and lung. Ann Thorac
Surg 1987; 43:7881
14 Ginsberg RJ, Rice TW, Goldberg M, et al. Extended cervical
mediastinoscopy: a single staging procedure for bronchogenic
carcinoma of the left upper lobe. Thorac Cardiovasc Surg
1987; 94:673678
15 Freixinet Gilart J, Garcia PG, de Castro FR, et al. Extended
cervical mediastinoscopy in the staging of bronchogenic
16 Urschel JD, Vretenar DF, Dickout WJ, et al. Cerebrovascular
accident complicating extended cervical mediastinoscopy.
17 Lopez L, Varela A, Freixinet J, et al. Extended cervical
mediastinoscopy: prospective study of fifty cases. Ann Thorac
Surg 1994; 57:555557
18 Ginsberg RJ. Discussion of: Lopez L, Varela A, Freixinet J, et
al: Extended cervical mediastinoscopy: prospective study of
fifty cases. Ann Thorac Surg 1994; 57:555558
19 Landreneau RJ, Hazelrigg SR, Mack MJ, et al. Thoracoscopic
mediastinal lymph node sampling: useful for mediastinal
lymph node stations inaccessible by cervical mediastinoscopy.
20 Detterbeck FC, Jones DR, Parker LA Jr. Intrathoracic staging.
In: Detterbeck FC, Rivera MP, Socinski MAet al, eds. Diagnosis
and treatment of lung cancer: an evidence-based guide for the
practicing clinician. Philadelphia, PA: W.B. Saunders, 2001;
7393
21 Bilaceroglu S, Cagiotariotaciota U, Gunel O, et al. Comparison of
rigid and flexible transbronchial needle aspiration in the staging of
bronchogenic carcinoma. Respiration 1998; 65:441449
Respir J 1994; 7:265268
23 Rodr guez de Castro F, Rey A, Caminero J, et al. Transbron-
chial fine needle aspiration in clinical practice. Cytopathology
1995; 6:2229
24 Garpestad E, Goldberg S, Herth F, et al. CT fluoroscopy
26 Fritscher-Ravens A, Soehendra N, Schirrow L, et al. Role of
transesophageal endosonography-guided fine-needle aspira-
tion in the diagnosis of lung cancer. Chest 2000; 117:339345
staging of lung cancer. Ann Thorac Surg 1996; 61:14411446
28 Wiersema MJ, Vazquez-Sequeiros E, Wiersema LM. Evaluation
of mediastinal lymphadenopathy with endoscopic US-guided
fine-needle aspiration biopsy. Radiology 2001; 219:252257
29 Wallace MB, Silvestri GA, Sahai AV, et al. Endoscopic ultra-
sound-guided fine needle aspiration for staging patients with
carcinoma of the lung. Ann Thorac Surg 2001; 72:18611867
2003;123;167-175 Chest
Silvestri
Frank C. Detterbeck, Malcolm M. DeCamp, Jr., Leslie J. Kohman and Gerard A.
Invasive Staging: The Guidelines
& Services
Updated Information
S
References
S#BIBL
free at:
Citations
S#otherarticles
Reprints
2003;123;176-180 Chest
Praveen N. Mathur, Eric Edell, Tom Sutedja and Jean-Michel Vergnon
Treatment of Early Stage Non-small Cell Lung Cancer
ISSN: 0012-3692.
Treatment of Early Stage Non-small Cell
Lung Cancer*
Praveen N. Mathur, MBBS, FCCP; Eric Edell, MD, FCCP;
Tom Sutedja, MD, PhD, FCCP; and Jean-Michel Vergnon, MD, PhD
Photodynamic therapy (PDT), brachytherapy, electrocautery, cryotherapy, and Nd-YAG laser
therapy are therapeutic options available for management of endobronchial malignancies. All of
these treatment modalities have been used for both palliation of late obstructing cancers, and
more recently have been used as primary treatment of early radiographically occult cancers. We
reviewed the evidence for the use of these treatment options in the management of early lung
cancer. (CHEST 2003; 123:176S180S)
Key words: carcinoma in situ; cryotherapy; interventional bronchoscopy; lasers; lung cancer; photodynamic therapy
Abbreviation: PDT photodynamic therapy
T
he definition of an early stage cancer for discus-
sion in this article will be as follows: a radio-
graphically occult squamous cell carcinoma that is
2 cm in surface area, appears superficial endo-
scopically, has clearly visible margins, and has no
invasion beyond the bronchial cartilage assessed
either by pathologic assessment or by available im-
agery including high-resolution CT.
Published guidelines on lung cancer diagnosis and manage-
ment were identified by a systematic review of the literature and
evaluated by the Appraisal of Guidelines for Research and
Evaluation method (see Section on Methods and Grading).
Those guidelines including recommendations specific to the
treatment of early stage lung cancer were identified for inclusion
in this section. Supplemental material appropriate to this topic
was obtained by literature search of a computerized database
(MEDLINE) and review of the reference lists of relevant
articles. Recommendations were developed by the section editor
and writing committee, graded by a standardized method (see
Section on Methods and Grading) and then reviewed by the
entire guidelines panel, including the chair and the vice chair.
The majority of the reported studies consisted of small-to-
moderate case series. Clinical outcomes were defined as response
to treatment and included complete, partial, or no response.
Complete response is defined as no evidence of disease visually as
well as on histology and cytology examination. Some studies also
included time to tumor recurrence.
Review of Data
Roentgenographically occult lung cancers can be
detected in high-risk patients with either sputum
cytology or bronchoscopic inspection. Traditionally,
the only treatment available for these cancers was
surgical resection. Even though these cancers are
small, 70% of cases require a lobectomy, and in the
remaining 30%, either a bilobectomy or pneumonec-
tomy is required.
1
There are patients with reduced
cardiopulmonary reserve who are not candidates for
any of these surgical options. Additionally, 1 to 4% of
these patients will have a synchronous lung cancer.
2
Some studies report up to 15% of newly diagnosed
early lung cancer cases have a synchronous lesion.
35
The risk of acquiring a second lung cancer is 1 to
25% per year.
6
Endobronchial therapies that preserve lung func-
tion have been developed and include photodynamic
therapy (PDT),
7
brachytherapy, electrocautery,
8
cryotherapy,
5
and Nd-YAG laser therapy.
9
Most ra-
diographically occult cancers, by chest radiograph
and CT, are histopathologically squamous cell carci-
noma and are located in relatively large central
bronchi.
10
The majority of these occult cancers
invade the bronchial wall but are not metastatic.
11
Early stage squamous cell carcinomas have been
defined as lesions that are roentgenographically oc-
cult with no evidence of invasion beyond the bron-
chial cartilage. Depth of intrabronchial invasion is a
significant challenge, but bronchoscopic evaluation
can provide valuable information regarding depth of
invasion. Both the size of the lesion and its topo-
graphic appearance may determine the depth of
penetration. Lesions 10 mm in greatest dimension
*From the Indiana University Medical Center (Dr. Mathur),
Indianapolis, IN; Mayo Clinic (Dr. Edell), Rochester, MN; Free
University Hospital (Dr. Sutedja), Amsterdam, the Netherlands;
and University Hospitals of St. Etienne Ho pital NORD (Dr.
Vergnon), St. Etienne, France.
Correspondence to: Praveen N. Mathur, MBBS, FCCP, Indiana
University Medical Center, 550 North University Blvd, Suite
5450, Indianapolis, IN 46202-2879
with only superficial thickening of the epithelium
have been reported to invade beyond the bronchial
cartilage in 5% of cases examined. Those with a
nodular or polypoid appearance showed invasion in
18% and 27%, respectively.
12
The time of assessment of complete response
reported in published studies varied from 1 to 3
months. A significant proportion of those who had
complete response acquired recurrent disease on
longer follow-up. Therefore, the actual complete
response rate is lower. Many trials that assessed the
use of PDT in early stage lung cancer included stage
I disease in addition to stage 0. These factors have to
be taken into account when evaluating the results of
published reports. Currently, there has been no good
technique for early stage lung cancer detection,
and thus there has been no consistent method of
treatment.
PDT
PDT is based on the interaction of tumor-selective
photosensitizer and laser light. This interaction
causes selective death of tumor cells. The majority of
clinical data using PDT in early lung cancer has been
for treatment of patients who were deemed nonsur-
gical candidates. The greatest experience has emerged
from Japan in the last 2 decades.
2,7,1321
One hundred
forty-five patients (191 cancers) with early non-small
cell lung cancer have been treated with PDT since
1980. This includes 99 patients with stage 0 and 56
patients with stage IA disease. There were 141 men
and 4 women. The majority of cases (98%) were
squamous cell carcinoma. Complete response was
achieved in 86% of lesions, with a recurrence rate of
13%, thereby resulting in a long-term response of 75%.
When success of treatment was evaluated according to
lesion size, lesions 1.0 cm had a complete response
of 95% and lesions 2 cm had a complete response of
only 46%. Treatment success was also related to
whether the distal margin of the tumor could be clearly
seen bronchoscopically. If the margin was visible, a
complete response rate of 92% was achieved, com-
pared to 67% if the margin was not visible. If the lesion
was 1.0 cm and the margin was visible, complete
response was achieved in 98% of cases.
18,19
Imamura et al
3
studied 29 patients (39 cancers)
and achieved complete response in 64% of lesions.
Recurrence occurred in 36%, giving a long-term
response of 41%. On evaluation of lesion size, 72% of
lesions that were 3 cm
2
achieved a complete
response.
3
Ono et al
21
studied 36 patients (39 can-
cers) and achieved a complete response rate of only
31%, with a recurrence in 33%. Therefore, the
long-term response was only 21%.
21
A number of
smaller studies from Europe and Canada have re-
ported complete response rates of 62 to 91%.
2225
A
multicenter investigator-initiated experience was col-
lated and presented to the US Food and Drug
Administration for approval of porfimer sodium in
the treatment of early superficial squamous cell
carcinoma.
26
A total of 102 patients with radiologi-
cally occult (stage 0, IA, and IB) squamous cell lung
cancer were treated. An overall immediate complete
response rate of 78% was achieved (95% confidence
interval, 7 to 87). Forty-four percent of the patients
had recurrent tumor on follow-up, giving a long-term
response rate of 43%. The median time to tumor
recurrence was 2.8 years (range 0.1 to 10 years).
Analysis of the subgroup of the 24 inoperable pa-
tients revealed a complete response of 92% (95%
confidence interval, 81 to 100%). A similar recur-
rence rate of 46%, a long-term response rate of 50%,
and a median time to tumor recurrence of 2.7 years
was observed.
The Mayo Clinic has reported treatment of 58
nonsurgical patients with early lung cancer.
2733
An
84% complete response rate was achieved after one
treatment. Nineteen patients (39%) recurred and
had a second PDT treatment. The median time to
tumor recurrence after the first treatment was 4.1
years. Following the second treatment, 11 patients
(22%) had recurrences. The long-term complete
response rate was 66%.
PDT as an alternative to surgical resection was
studied in 21 patients with small bronchial cancer.
31
A 71% complete response (15 of 21 patients) was
achieved, with 11 patients (52%) maintaining a com-
plete response 12 months. Patients who did not
respond or had recurrence were offered surgery. Of
the 10 patients who underwent surgery, 3 patients
were found to have N1 disease. Two patients refused
surgery. A total of nine patients (43%) were spared
surgery.
In summary, PDT is affective in managing small
superficial squamous cell carcinoma. The worldwide
data showed that patients with early lung cancer
treated with PDT achieve a complete response in
approximately 75% of cases, with a recurrence rate
of approximately 30%. Complete response rates as
well as recurrence rates are best when lesions are
small ( 1 cm in diameter) and superficial. Experi-
ence remains limited using PDT for patients who are
surgical candidates.
Recommendations
1. For patients with early superficial squamous
cell carcinoma who are not surgical candidates,
PDT should be considered as a treatment
option. Level of evidence, fair; benefit, moder-
ate; grade of recommendation, B
cell carcinoma who are surgical candidates, the
use of PDT appears to be a promising treat-
ment, but more experience is needed to com-
pare PDT to surgical outcomes. Level of evi-
dence, poor; benefit, none/negative; grade of
recommendation, I
Electrocautery
Electrocautery is the less expensive treatment for
endobronchial tumors. Bronchoscopic electrocau-
tery is the use of high-frequency electrical current
that generates heat due to tissue resistance, resulting
in destruction of tissue. A small study in early lung
cancer of 13 patients (15 cancers) showed a complete
response in 80% of lesions with no recurrence at 22
months of follow-up.
8
Recommendation
3. In patients with early superficial squamous cell
carcinoma, electrocautery may be an alterna-
tive to PDT for treatment. Level of evidence,
poor; benefit, moderate; grade of recommen-
dation, C
Cryotherapy
Cryotherapy is a technique where tissue is de-
stroyed by freezing and is the least expensive for
treatment. A recent report included 35 patients (41
cancers) with early stage lung cancer. A complete
response was obtained 91% of the patients with a
recurrence rate of 28% within 4 years. A long-term
response of 63% was achieved, similar to that of PDT.
5
Recommendation
carcinoma, cryotherapy may be an alternative
to PDT for treatment. Level of evidence, poor;
Brachytherapy
Brachytherapy refers to the placement of a radio-
active source within or near an endobronchial ma-
lignancy to deliver local irradiation. This requires the
insertion of an afterloading polyurethane catheter
into the airway adjacent to the tumor during fiber-
optic bronchoscopy. Iridium 192 is generally used. In
two small studies, the use of high-dose brachy-
therapy in three to six sessions reported response
rates similar to PDT.
34,35
Marsiglia et al
34
reported
34 patients with early stage lung cancer with a
complete response of 85% seen over 2 years of
follow-up. Perol et al
35
reported 19 patients with
early stage lung cancer with a complete response rate
of 83%, which fell to 75% at 1-year follow-up.
Recommendation
carcinoma, brachytherapy may be an alterna-
dation, C
Nd-YAG Laser Therapy
Nd-YAG laser therapy is used for direct thermal
ablation of tissue in endobronchial malignancy. It has
been used extensively as a palliative measure to
relieve airway obstruction. The use of laser treat-
ment for early lung cancer has not been widely
studied. A study by Cavaliere et al
9
showed a com-
plete response rate of 100% in 22 patients with small
bronchial cancers. The long-term outcome of these
patients was not reported. Yttrium-aluminum-garnet
laser therapy is not indicated for tumors that are
located in the bronchial wall parallel to the broncho-
scope or for tumors involving smaller bronchial
branches because of the risk of perforation
36
that
would occur due to heat sink effect and absorption of
heat by the tissue.
Recommendation
carcinoma, Nd-YAG laser therapy is not indi-
cated for treatment. The risk of perforation is
very high. Level of evidence, poor; benefit,
none/negative; grade of recommendation, I
Follow-up/Surveillance
Following treatment, patients should be closely
monitored for recurrent disease and development of
metachronous lesions. These patients should undergo
bronchoscopic examination every 3 to 6 months with
both white light and fluorescence bronchoscopy if
available. Please refer to the chapter of Follow-up/
Surveillance in this guideline publication.
Conclusion
PDT is the most extensively studied endobron-
chial treatment for early lung cancer for patients who
are not candidates for surgical resection. Suitable
lesions ( 1 cm) require careful assessment bron-
choscopically and radiographically. The data for use
of PDT for patients who are surgical candidates
appears encouraging. Other endobronchial treat-
ments such as electrocautery, cryotherapy, and
brachytherapy are not as well studied, but appear to
have similar response rates to PDT. The best re-
sponse is seen in highly selected patients with small
lesions and visible margins.
cell carcinoma who are not surgical candidates,
PDT should be considered as a treatment
option. Level of evidence, fair; benefit, moder-
cell carcinoma who are surgical candidates, the
use of PDT appears to be a promising treat-
ment, but more experience is needed to com-
pare PDT to surgical outcomes. Level of evi-
dence, poor; benefit, none/negative; grade of
recommendation, I
carcinoma, electrocautery may be an alterna-
dation, C
carcinoma, cryotherapy may be an alternative
to PDT for treatment. Level of evidence, poor;
carcinoma, brachytherapy may be an alterna-
dation, C
carcinoma, Nd-YAG laser therapy is not indi-
cated for treatment. The risk of perforation is
very high. Level of evidence, poor; benefit,
none/negative; grade of recommendation, I
References
1 Cortese D, Pairolero P, Bergstalh E, et al. Roentgenograph-
ically occult lung cancer: a ten-year experience. J Thorac
2 Okunaka T, Kato H, Konaka C, et al. Photodynamic therapy
for multiple primary bronchogenic carcinoma. Cancer 1991;
68:253258
3 Imamura S, Kusunoki Y, Takifuji N, et al. Photodynamic
therapy and/or external beam radiation therapy for roentgen-
ologically occult lung cancer. Cancer 1994; 73:16081614
4 Fujimura S, Sakurada A, Sagawa M, et al. A therapeutic
approach to roentgenographically occult squamous cell carci-
noma of the lung. Cancer 2000; 89(suppl):24452448
5 Deygas N, Froudarakis M, Ozenne G, et al. Cryotherapy in
early superficial bronchogenic carcinoma. Chest 2001; 120:
2631
6 Johnson B. Second lung cancers in patients after treatment
for an initial lung cancer. J Natl Cancer Inst 1998; 90:1335
1345
7 Lam S. Photodynamic therapy of lung cancer. Semin Oncol
1994; 21:1519
8 Van Boxem T, Venmans B, Schramel F, et al. Radiographi-
cally occult lung cancer treated with fiberoptic bronchoscopic
electrocautery: a pilot study of a simple and inexpensive
technique. Eur Respir J 1998; 11:169172
9 Cavaliere S, Foccoli P, Toninelli C, et al. Nd-YAG laser
therapy in lung cancer: an 11-year experience with 2,253
applications in 1,585 patients, J Bronchology 1994; 1:105111
10 Saito Y, Nagamoto M, Ota S, et al. Results of surgical
treatment for roentgenographically occult bronchogenic
squamous cell carcinoma. J Thorac Cardiovasc Surg 1992;
104:401407
11 Wooner L, Fontana R, Cortese D, et al. Roentgenographi-
cally occult lung cancer: pathologic findings and frequency of
multicentricity during a 10-year period. Mayo Clin Proc 1894;
59:453466
12 Konaka C, Hirano T, Kato H, et al. Comparison of endoscopic
features of early-stage squamous cell lung cancer and histo-
logical findings. Br J Cancer 1999; 80:14351439
13 Hayata Y, Kato H, Konaka C, et al. Photodynamic therapy
(PDT) in early stage lung cancer. Lung Cancer 1993; 9:287
294
14 Hayata Y, Kato H, Konaka C, et al. Hematoporphyrin deriv-
ative and laser photoradiation in the treatment of lung cancer.
Chest 1982; 81:269277
15 Hayata Y, Kato H, Konaka C, et al. Photoradiation therapy
with hematoporphyrin derivative in early and stage I lung
cancer. Chest 1984; 86:169177
16 Hayata Y. The indications of photodynamic therapy (PDT) in
early stage lung cancer [abstract]. Chest 1993; 103:262S
17 Furuse K, Fukuoka M, Kato H, et al. A prospective phase II
study on photodynamic therapy for centrally located early-
stage lung cancer. J Clin Oncol 1993; 11:18521857
18 Kato H, Okunaka T, Shimatani H. Photodynamic therapy for
early stage bronchogenic carcinoma. J Clin Laser Med Surg
1996; 14:235238
19 Kato H. Photodynamic therapy for lung cancer: a review of 19
years experience. J Photochem Photobiol B 1998; 42:9699
20 Kato H, Okunaka T, Tsuchida T, et al. Analysis of the
cost-effectiveness of photodynamic therapy in early stage lung
cancer. Diagn Ther Endosc 1999; 6:916
21 Ono R, Ikeda S, Suemasu K. Hematoporphyrin derivative
photodynamic therapy in roentgenographically occult carci-
noma of the tracheobronchial tree. Cancer 1992; 69:1696
1701
22 Sutedja T, Baas P, Stewart F, et al. A pilot study of
photodynamic therapy in patients with inoperable non-small
cell lung cancer [abstract]. Eur J Cancer 1992; 28:13701373
23 Sutedja T, Lam S, leRiche J, et al. Response and pattern of
failure after photodynamic therapy for intraluminal stage I
lung cancer. J Bronchology 1994; 1:295298
24 Radu A, Grosjean P, Fontolliet C, et al. Photodynamic
therapy for 101 early cancers of the upper aerodigestive tract,
the esophagus, and the bronchi: a single-institution experi-
ence. Diagn Ther Endosc 1999; 5:145154
25 Patelli M, Lazzari A, Poletti V, et al. Photodynamic laser
therapy for the treatment of early-stage bronchogenic carci-
noma. Monaldi Arch Chest Dis 1999; 54:315318
26 Lam S, Haussinger K, Leroy M, et al. Photodynamic therapy
(PDT) with Photofrin, a treatment with curative potential for
early stage superficial lung cancer [abstract]. Presented at:
34th Annual Meeting of the American Society of Clinical
Oncology, Los Angeles, CA, 1998
27 Edell E, Cortese D, McDougall J. Ancillary therapies in the
management of lung cancer: photodynamic therapy, laser
therapy, and endobronchial prosthetic devices. Mayo Clin
Proc 1993; 68:685690
28 Edell E, Cortese D. Bronchoscopic phototherapy with hemato-
porphyrin derivative for treatment of localized bronchogenic
carcinoma: a 5-year experience. Mayo Clin Proc 1987; 62:814
29 Cortese D, Kinsey K. Hematoporphyrin derivative photother-
apy in the treatment of bronchogenic carcinoma. Chest 1984;
86:813
30 Edell E, Cortese D. Bronchoscopic localization and treat-
ment of occult lung cancer. Chest 1989; 96:919921
31 Edell E, Cortese D. Photodynamic therapy: treatment of
early stage lung cancer; results and follow-up. Excerpta
Medica International Congress series No. 894, 1990:205210
32 Edell E, Cortese D. Photodynamic therapy in the manage-
ment of early superficial squamous cell carcinoma as an
alternative to surgical resection. Chest 1992; 102:13191322
33 Cortese D, Edell E, Kinsey J. Photodynamic therapy for early
stage squamous cell carcinoma of the lung. Mayo Clin Proc
1997; 72:595602
34 Marsiglia H, Baldeyrou P, Lartigau E, et al. High-dose-rate
brachytherapy as sole modality for early-stage endobronchial
carcinoma. Int J Radiat Oncol Biol Phys 2000; 47:665672
35 Perol M, Caliandro R, Pommier P, et al. Curtive irradiation of
limited endobronchial carcinomas with high dose brachyther-
apy: results of a pilot study. Chest 1997; 111:14171423
36 Lam S. Bronchoscopic, photodynamic, and laser diagnosis
and therapy of lung neoplasms. Curr Opin Pulm Med 1996;
2:271276
2003;123;176-180 Chest
Praveen N. Mathur, Eric Edell, Tom Sutedja and Jean-Michel Vergnon
Treatment of Early Stage Non-small Cell Lung Cancer
& Services
Updated Information
S
References
S#BIBL
free at:
Reprints
2003;123;181-187 Chest
W. Roy Smythe
Treatment of Stage I Non-small Cell Lung Carcinoma
ISSN: 0012-3692.
Treatment of Stage I Non-small Cell
Lung Carcinoma*
W. Roy Smythe, MD, FCCP
The American Joint Committee on Cancer defines stage I non-small cell lung carcinoma (NSCLC)
as consisting of patients with a T1 or T2 primary tumor designation and no evidence of hilar or
mediastinal nodal disease (N0) or metastatic spread (M0). Medically fit patients in this clinical
stage category based on conventional staging techniques should be considered for aggressive
local therapy, and curative treatment is possible. Surgical resection is the accepted treatment for
patients with this stage grouping, and full lobar or greater (lobectomy, pneumonectomy) rather
than sublobar (wedge resection, segmentectomy) resection is strongly suggested. There is
insufficient data to suggest that one method of resection (open thoracotomy, minimally invasive
techniques) is superior to another. The performance of a systematic sampling or full mediastinal
lymph node dissection may improve pathologic staging but is unproven therapeutically. There
are no data supporting the routine use of chemotherapy in an adjuvant or neoadjuvant setting;
however, recent phase II data suggest that neoadjuvant chemotherapy is feasible and safe, and
larger phase III trials are now evaluating this modality. Primary radiation therapy should be
considered for inoperable patients. The use of neoadjuvant or adjuvant radiation therapy in
patients with stage I NSCLC is of unproven benefit. (CHEST 2003; 123:181S187S)
Key words: chemotherapy; lung cancer; radiation therapy; stage I; surgery
Abbreviation: NSCLC non-small cell lung cancer
S
tage I non-small cell lung carcinoma (NSCLC) is
defined by the American Joint Commission on
Cancer as a T1 or T2 tumor in the parenchyma of the
lung, no more proximal than 2 cm from the carina,
and not invading chest wall or parietal pleura. In
addition, patients in this stage grouping have hilar
(N1) and mediastinal (N2) lymph node stations
negative for tumor, and no metastatic (M1) disease.
Naruke et al
1
and Mountain
2
published two of the
largest series evaluating postsurgical survival in
NSCLC. In these studies, 1,500 patients with
stage I NSCLC (1997 American Joint Commission
on Cancer system designation) were treated surgi-
cally, and survival was retrospectively assessed. Five-
year survival for patients with T1N0M0 in these
series combined was 71.25% and for patients with
T2N0M0 was 57%.
This stage, due to differences in survival statistics,
is further subdivided into stage IA (T1N0M0) and
stage IB (T2N0M0).
3
Unfortunately, early stage dis-
ease is relatively less common at presentation than is
more advanced NSCLC. The National Cancer Insti-
tute Surveillance, Epidemiology and End Results
Program
4
recently published data regarding patients
with NSCLC in this country from 1989 to 1996. In
this analysis, 15% of patients were found to have
localized rather than regional, distant, or un-
staged tumors.
Clearly, one should not promise a cure to patients
with a diagnosis of NSCLC at any stage. However,
there is no doubt that survival following treatment in
this disease is stage related, and that patients with
lower stage disease represent those with the best
chance for curative treatment. With this in mind, the
appropriate treatment of patients with earlier-stage
disease takes on perhaps even greater importance, as
the potential for a lost curative opportunity is great-
est. This section will critically evaluate the role of
surgery as the accepted primary modality of therapy
for this malignancy at this time, as well as the
questions of proper use of radiation therapy and
chemotherapy in both the neoadjuvant and adjuvant
settings. In addition, recommendations for alterna-
tive primary therapy for patients who are not surgical
candidates will be discussed. Other chapters will deal
with the staging evaluation of patients with early
stage NSCLC, and stage I small cell lung cancer will
be reviewed elsewhere in these guidelines as well.
*From the Department of Thoracic and Cardiovascular Surgery,
The University of Texas M.D. Anderson Cancer Center, Hous-
ton, TX.
Correspondence to: W. Roy Smythe, MD, FCCP, Department of
Thoracic and Cardiovascular Surgery, The University of Texas
M.D. Anderson Cancer Center, 1515 Holcombe Blvd, Box 445,
Houston, TX; e-mail: rsmythe@mdanderson.org
Although the majority of guideline recommendations
of the American College of Chest Physicians for
treatment of this stage of NSCLC are likely to be less
controversial than treatment of more advanced
stages, the area is a dynamic one, and readers are
strongly encouraged to keep abreast of develop-
ments over the next several years in treatment of this
particular category of patients. It is virtually certain
that advances in diagnostic radiology, radiation ther-
apy, and tailored systemic treatment in medical
oncology will make these recommendations some-
what obsolete in the foreseeable future.
Surgical Resection for Patients With
Stage I NSCLC
What Is the Evidence Regarding Surgical Resection
as the Preferred Primary Treatment Modality for
Patients With Stage I NSCLC?
To date, there have been no large randomized
trials evaluating surgical resection vs other treatment
modalities for stage I NSCLC. However, a compar-
ison of data from trials evaluating independent and
combined modalities, and an examination of rates of
postsurgical survival strongly suggest that surgical
resection should be viewed as the preferred primary
modality. Chemotherapy as a primary treatment
modality in stage I NSCLC has not been extensively
studied, but combination adjuvant and neoadjuvant
studies published to date do not demonstrate any
significant survival benefit over surgical resection
alone. This being noted, the majority of patients that
succumb to NSCLC following surgical resection fail
with distant metastatic disease, arguing against the
use of surgery alone.
Numerous randomized trials evaluating the use of
more active modern systemic chemotherapy agents
(cisplatin-based) in the adjuvant setting with reason-
able clinical staging criteria have been performed.
Although these trials were not specifically designed
to treat stage I patients, a large number in this
category were randomized. Although relatively well
tolerated, no significant survival benefits were noted
in any of these individual trials.
57
When the use of
adjuvant chemotherapy in NSCLC has been evalu-
ated by meta-analysis, it has been shown that the use
of alkylating agents favors surgery alone, but that the
use of cisplatin-based regimens are found to have a
5% approximate benefit. Although modest, this de-
gree of benefit may prove clinically useful consider-
ing the prevalence of the disease.
8
Interestingly,
there have been two randomized trials evaluating the
use of adjuvant chemotherapy in NSCLC patients
utilizing tegafur alone or in combination with other
chemotherapeutic agents vs surgery alone. Signifi-
cant improvement in survival and disease free sur-
vival were noted in both of these trials.
9,10
There has
been a recent resurgence of interest regarding the
use of neoadjuvant chemotherapy based in part on
results of smaller stage IIIA trial results and the
availability of more active conventional chemother-
apy agents.
11,12
A successful feasibility and toxicity
study of neoadjuvant chemotherapy was recently
completed for patients with stages IB or IIIA, and
randomized trials are in progress.
13
Several molecu-
lar and cellular marker studies in NSCLC suggest
that there may be a large degree of biological
heterogeneity for tumors within the stage I grouping.
These markers may provide for identification of
patients with a higher propensity for recurrence or
metastasis, and possibly benefit from chemotherapy.
14
Studies evaluating the use of adjuvant or neoadju-
vant radiation therapy in early stage lung carcinoma
have not proven to add to the survival benefit of
surgical resection alone, and a meta-analysis has
suggested that adjuvant radiation therapy may have
an adverse effect in stage I disease, with similar
adverse effect noted in at least one randomized
neoadjuvant study
15,16
A comparison of survival rates
of patients treated with radiation therapy alone for
NSCLC due to medical inoperability or refusal of
surgical care to those undergoing surgical resection
indicates an approximate 25 to 35% improvement in
survival in patient undergoing surgical resection.
1719
It
is important, however, to consider several caveats re-
garding the findings of adjuvant and neoadjuvant radi-
ation therapy to date. Many of these studies have
utilized older modalities such as cobalt irradiation. The
use of newer modalities as primary therapy such as
proton and stereotactic techniques appear to be much
more promising, although not studied in a randomized
fashion.
20,21
And finally, the survival of medically unfit
patients may be influenced by the underlying nonma-
lignant disease. The use of adjuvant radiation therapy in
patients with positive resection margins has not been
carefully studied, but there is a suggestion of benefit.
22
Who should routinely treat patients with NSCLC?
Interestingly, multiple recent reports strongly sug-
gest that a higher volume of lung cancer treatment at
a given institution as well as the degree of specific
specialty training have a significant positive influence
on the attitudes of clinicians regarding lung cancer
treatment, the treatments administered, and the
success of treatment, including surgical resection.
2326
Recommendations
1. For patients with clinical stage I (IA and IB)
NSCLC and no medical contraindication to
operative intervention, surgery alone is the
preferred treatment modality. Level of evi-
ommendation, B
operative intervention, a complete surgical re-
section (clear surgical margins) is to be
achieved if possible in all cases. Level of evi-
dence, good; benefit, substantial; grade of rec-
ommendation: A
3. All patients considered surgical candidates
should be evaluated for surgical resection by
surgeons trained and board certified or board
eligible in thoracic surgery. Level of evidence,
good; benefit, substantial; grade of recommen-
dation, A
4. Patients with positive resection margins should
be evaluated for additional local treatment mo-
dalities (surgical re-resection or radiation ther-
apy). Level of evidence, fair; benefit, moderate;
operative intervention, the use of neoadjuvant
chemotherapy has been shown to be feasible,
but is not recommended outside the setting of
a clinical trial. Level of evidence, poor; benefit,
small/weak; grade of recommendation, I
operative intervention, the use of adjuvant che-
motherapy is not recommended outside the
setting of a clinical trial. Level of evidence, fair;
tion, D
operative intervention, the routine use of neo-
adjuvant or adjuvant radiation therapy should
not be performed. Level of evidence, good;
tion: D
What Is the Evidence to Support Full Anatomic
Resection (Lobectomy, Pneumonectomy) Over
Lesser Resections (Wedge Resection,
Segmentectomy) for Stage I NSCLC?
Full anatomic resection is defined as a resection of
either a complete lobe of the lung (lobectomy) or the
entire lung (pneumonectomy). Either of these pro-
cedures require dissection and division of hilar vas-
cular and bronchial structures. Virtually all shared
vascular, lymphatic, and vascular divisions are re-
moved as a unit. Wedge resection refers to the
removal of a nonanatomic portion of the lung,
usually performed as removal of a wedge of paren-
chymal with tumor near the pleural surface. Techni-
cally, a segmentectomy is an anatomic resection of a
bronchopulmonary segment; however, hilar dissec-
tion is not usually required. Both of these lesser
operations require division of the lung parenchyma
across shared lobar vasculature, lymphatics and
bronchi, theoretically increasing the risk of local
recurrence. Multiple single institution nonrandom-
ized studies were performed in the 1980s and 1990s
that suggested that there was a substantial risk of
recurrence (14 to 23%) of early stage NSCLC when
a wedge or segmental resection was performed.
2731
A large retrospective study was published in 1995
evaluating subanatomic resections (n 61) and full
anatomic resection (n 511) in patients with stage I
NSCLC.
32
In this study, patients with lesser resec-
tions demonstrated a 5- and 10-year survival of 59%
and 35%, as compared to 77% and 79% in those
undergoing full anatomic resection.
32
Finally, a pro-
spective, randomized controlled trial was completed
by the Lung Cancer Study Group and reported in
1995.
33
In this study, 250 patients were allocated to
either approach. The authors determined that the
lung cancer recurrence rate was 75% greater in the
limited resection due to a tripling of local tumor
recurrence, and that there was a 50% increase in
death with cancer.
33
In the case of patients with adequate pulmonary
function undergoing thoracotomy but not anatomic
parenchymal resection, sublobar pulmonary resec-
tion may be an alternative in stage I NSCLC.
Operative mortality in this setting has been reported
to be 5%, and overall survival when compared to
anatomic resection is approximately 10% lower.
2731
Recommendations
8. Patients with stage I NSCLC who are medically
fit for conventional surgical resection should
undergo lobar or greater resection (lobectomy,
pneumonectomy) rather than sublobar (wedge
or bronchopulmonary segment) resections. Level
of evidence, good; benefit, substantial; grade of
recommendation, A
9. Patients with stage I (IA and IB) NSCLC who
may tolerate operative intervention but not a
lobar or greater lung resection due to comorbid
disease or compromised pulmonary function
should undergo sublobar (wedge or broncho-
pulmonary segment) resection. Level of evi-
dence, poor; benefit, substantial; grade of rec-
ommendation, C
What Is the Evidence To Support Either
Conventional Thoracotomy or Minimally Invasive
Approaches to Anatomic Resection for Stage I
NSCLC?
The use of thoracoscopic or minimally invasive
sublobar resection for NSCLC would not be antici-
pated to have any better outcome than those dis-
cussed earlier via open thoracotomy. In recent years,
a number of investigators have compared the use of
thoracoscopic or minimally invasive anatomic resec-
tion to that of standard open thoracotomy. Although
a large number of trials have been published, the
majority are retrospective in nature.
3439
There is a
prevalent suggestion of an improvement in postop-
erative pain, but this has not been durable in all
studies. In addition, minimally invasive techniques
have not always been evaluated against compara-
tively less invasive (ie, small incision or muscle-
sparing) conventional surgical procedures. The short
and intermediate survival do not appear to be differ-
ent than with open procedures. However, it is
reasonable to consider that without a large prospec-
tive randomized study, survival cannot be assessed
without significant bias, and that even postoperative
pain assessment can be biased without the benefit of
blinding of investigators and/or patients.
Recommendation
operative intervention, the use of video-
assisted surgical techniques for lobar or
greater NSCLC resection may be associated
with less postoperative pain; however, there
are insufficient data at this time to recom-
mend this type of procedure as an alternative
to conventional techniques. Level of evidence,
poor; benefit, small/weak; grade of recom-
mendation, I
What Is the Evidence To Support Nodal Dissection
in Patients With Stage I NSCLC Rather Than
Other Approaches for Evaluation of Involvement of
Mediastinal Lymph Nodes?
The issues to consider in regard to mediastinal
lymph node evaluation at the time of anatomic
resection are the differential benefit to definitive
staging and survival and the potential associated
morbidity. The options for mediastinal lymph node
evaluation are many, and include no mediastinal
lymph node dissection whatsoever, mediastinal lymph
node sampling, conventional mediastinal lymph node
dissection (systematic sampling) and radical en bloc
resection of mediastinal lymph nodes and surrounding
mediastinal fat and other structures.
The question of morbidity has not been carefully
assessed; however, in a retrospective analysis of
systematic nodal dissection vs nodal sampling, no
differences in operative time or blood loss were
noted.
40
Likewise, no significant differences in mor-
bidity were noted when radical en bloc nodal dissec-
tion was compared to conventional lymph node
dissection.
41
The theoretical survival benefit of me-
diastinal lymph node dissection in patients who by
definition have no positive nodes at pathologic ex-
amination is not intuitive. In regard to the relation-
ship of survival to method of mediastinal nodal
evaluation for all comers with the disease, the results
are conflicting. One report has demonstrated a sur-
vival difference for patients with stage II and III
NSCLC undergoing a conventional dissection rather
than sampling, but no data exist in this regard for
stage I disease, and this was a nonrandomized re-
port.
40
A similar dilemma arises in the stage I
grouping in regard to staging benefit from mediasti-
nal nodal dissection if one considers pathologic stage
only. However, it is pertinent to note that many
patients are upstaged by surgery, and that in a study
evaluating retrospectively the use of sampling vs
systematic dissection that more positive nodal sta-
tions were identified.
40
A number of authors claim
that no nodal dissection whatsoever should be un-
dertaken in patients with small peripheral tumors, or
that a sampling at most should be performed. These
investigators demonstrate similar survival in stage I
NSCLC with these characteristics with or without
full nodal dissection.
42,43
However, this is countered
by numerous studies demonstrating the inability of
clinical evaluation, regardless of stage, to accurately
predict mediastinal nodal involvement.
17,40,44
Ran-
domized trials evaluating these questions in stage I
disease do not exist. Interestingly, a number of
investigators are evaluating the use of sentinel
lymph node sampling of mediastinal nodal stations in
an effort to obviate the need for systematic dissec-
tion, but this is an investigational tool at best at this
time.
45,46
Finally, the advent of more sensitive and
specific nonoperative staging modalities, such as
positron emission tomography and other newer im-
aging approaches, may render mediastinal lymph
node dissection unnecessary as a staging endeavor in
the near future.
Recommendations
11. All patients undergoing resection for stage I
NSCLC (IA and IB) should have intraopera-
tive systematic surgical mediastinal lymph
node evaluation for accurate pathologic stag-
Treatment of Nonoperative Candidates
With Stage I NSCLC
What Is the Evidence To Support the Use of
Radiation Therapy as the Primary Treatment
Modality in Patients With Stage I NSCLC Who
Are Unable Medically or Unwilling To Undergo a
Surgical Procedure?
Although anatomic resection is the preferred local
treatment modality for stage I NSCLC, a subset of
patients either refuse or are medically unfit for
surgical treatment. These patients may also benefit
from some form of local control modalitysuch as
definitive radiation therapy or subanatomic resection
as outlined in the previous section. A number of
studies have evaluated the use of radiation therapy in
this cohort, with survival averaging approximately
30% at 5 years.
1719
A recent meta-analysis was
performed to evaluate this paradigm. Utilizing crite-
ria that required patients to receive 40 Gy in 20
fractions over 4 weeks or its radiobiological equiva-
lent, 1 randomized trial and 35 nonrandomized trials
were identified.
47
The cancer-specific survival was 13
to 39% at 5 years in the studies evaluated, and overall
survival at 2 years with continuous hyperfractionated
accelerated radiotherapy (37%) was superior to 60
Gy over 6 weeks (24%). Although supporting radia-
tion therapy over best supportive care, one criticism
of the literature put forth by the authors is a lack of
randomized trials, and trials comparing palliative vs
immediate curative intent radical irradiation.
Newer advances in radiation therapy, such as
three-dimensional conformal techniques and inten-
sity-modulated radiation therapy stereotactic-guided
proton beam promise to improve these results in the
short term, as well as the use of combination thera-
pies with biological modifiers of radiation cellular
response.
20,21,48
The reader is urged to follow the
literature carefully.
Recommendation
12. Patients with stage I NSCLC deemed medi-
cally unable to tolerate operative intervention
or refusing surgical resection and having no
medical contraindication to radiation therapy
should receive this modality as definitive
treatment. Level of evidence, fair; benefit,
In addition to the articles referred to in the text of
this chapter, several preexisting lung cancer treat-
ment guidelines were utilized as reference material
in the creation of these recommendations.
4548
Summary Statement
In most patients, stage I NSCLC is readily treat-
able, and favorable outcomes are possible. There are
good data to support the treatment recommenda-
tions put forth in this chapter, and following these
recommendations is likely to make the treatment of
patients with this stage grouping of NSCLC more
efficient as well as efficacious.
operative intervention, surgery alone is the
preferred treatment modality. Level of evi-
ommendation, B
operative intervention, a complete surgical
resection (clear surgical margins) is to be
achieved, if possible in all cases. Level of evi-
dence, good; benefit, substantial; grade of rec-
ommendation, A
3. All patients considered surgical candidates
should be evaluated for surgical resection by
surgeons trained and board certified or board
eligible in thoracic surgery. Level of evidence,
good; benefit, substantial; grade of recom-
mendation, A
4. Patients with positive resection margins
should be evaluated for additional local treat-
ment modalities (surgical re-resection or radi-
ation therapy). Level of evidence, fair; benefit,
operative intervention, the use of neoadjuvant
chemotherapy has been shown to be feasible,
but is not recommended outside the setting of
a clinical trial. Level of evidence, poor; bene-
fit, small/weak; grade of recommendation, I
operative intervention, the use of adjuvant
chemotherapy is not recommended outside
the setting of a clinical trial. Level of evidence,
fair; benefit, none/negative; grade of recom-
mendation, D
operative intervention, the routine use of neo-
adjuvant or adjuvant radiation therapy should
not be performed. Level of evidence, good;
tion, D
8. Patients with stage I NSCLC who are medi-
cally fit for conventional surgical resection
should undergo lobar or greater resection
(lobectomy, pneumonectomy) rather than
sublobar (wedge or bronchopulmonary seg-
ment) resections. Level of evidence, good; ben-
9. Patients with stage I (IA and IB) NSCLC who
may tolerate operative intervention but not a
lobar or greater lung resection due to comor-
bid disease or compromised pulmonary func-
tion should undergo sublobar (wedge or bron-
chopulmonary segment) resection. Level of
recommendation, C
operative intervention, the use of video-
assisted surgical techniques for lobar or
greater NSCLC resection may be associated
with less postoperative pain; however, there
are insufficient data at this time to recom-
mend this type of procedure as an alternative
to conventional techniques. Level of evidence,
poor; benefit, small/weak; grade of recom-
mendation, I
11. All patients undergoing resection for stage I
NSCLC (IA and IB) should have intraopera-
tive systematic surgical mediastinal lymph
node evaluation for accurate pathologic stag-
12. Patients with stage I NSCLC deemed medi-
cally unable to tolerate operative intervention
or refusing surgical resection and having no
medical contraindication to radiation therapy
should receive this modality as definitive treat-
ment. Level of evidence, fair; benefit, substan-
References
1 Naruke T, Goya T, Tsuchiya R, et al. Prognosis and survival in
resected lung carcinoma based on the new international
staging system. Thorac Cardiovasc Surg 1988; 96:440447
3 American Joint Committee on Cancer Lung. In: Fleming ID,
Cooper JS, Henson DE, et al, eds. AJCC Cancer Staging
Handbook. Philadelphia, PA: Lippincott-Raven, 1997; 117
127
4 Ries LA, Wingo PA, Miller DS, et al. The annual report to the
nation on the status of cancer, 19731997, with a special
section on colorectal cancer. Cancer 2000; 88:23982424
5 Holmes EC, Gail M. Surgical adjuvant therapy for stage II
and stage III adenocarcinoma and large-cell undifferentiated
carcinoma. Lung Cancer Study Group. J Clin Oncol 1986;
4:710715
6 Niiranen A, Niitamokorhonen S, Kouri M, et al. Adjuvant
chemotherapy after radical surgery for non-small-cell lung
cancer: a randomized study. J Clin Oncol 1992; 10:19271932
7 Feld R, Rubinstein L, Thomas PA, et al. Adjuvant chemo-
therapy with cyclophosphamide, doxorubicin, and cisplatin in
patients with completely resected stage I non-small-cell lung
8 Chemotherapy in non-small cell lung cancer: a meta-analysis
using updated data on individual patients from 52 randomised
clinical trials. Non-small Cell Lung Cancer Collaborative
Group. BMJ 1995; 311:899909
9 Imaizumi M. A randomized trial of postoperative adjuvant
chemotherapy in non-small cell lung cancer (the second
cooperative study). The Study Group of Adjuvant Chemo-
therapy for Lung Cancer. Eur J Surg Oncol 1995; 21:6977
10 Wada H, Hitomi S, Teramatsu T, et al. Adjuvant chemother-
apy after complete resection in non-small cell lung cancer.
J Clin Oncol 1996; 14:10481054
11 Roth JA, Swisher SG, Merritt JA, et al. Gene therapy for
non-small cell lung cancer: a preliminary report of a phase I
trial of adenoviral p53 gene replacement. Semin Oncol 1998;
25(3 Suppl 8):3337
12 Rosell R, Gomez-Codina J, Camps C, et al. Preresectional
chemotherapy in stage IIIA non-small-cell lung cancer: a
7-year assessment of a randomized controlled trial. Lung
Cancer 1999; 26:714
13 Pisters KMW, Ginsberg RJ, Giroux DJ, et al. Induction
chemotherapy before surgery for early-stage lung cancer: a
novel approach. Thorac Cardiovasc Surg 2000; 119:429439
14 DAmico TA, Aloia TA, Moore MB, et al. Predicting the sites
of metastases from lung cancer using molecular biologic
markers. Ann Thorac Surg 2001; 72:11441148
15 Postoperative radiotherapy in non-small-cell lung cancer:
systematic review and meta-analysis of individual patient data
from nine randomised controlled trials. PORT Meta-analysis
Trialists Group. Lancet 1998; 352:257263
16 Shields TW. Preoperative radiation therapy in the treatment
of bronchial carcinoma. Cancer 1972; 30:13881394
17 Graham PH, Gebski VJ, Langlands AO. Radical radiotherapy
for early nonsmall cell lung cancer. Int J Radiat Oncol Biol
Phys 1995; 31:261266
18 Morita K, Fuwa N, Suzuki Y, et al. Radical radiotherapy for
medically inoperable non-small cell lung cancer in clinical
stage I: a retrospective analysis of 149 patients. Radiother
Oncol 1997; 42:3136
19 Kupelian PA, Komaki R, Allen P. Prognostic factors in the
treatment of node-negative nonsmall cell lung carcinoma with
radiotherapy alone. Int J Radiat Oncol Biol Phys 1996;
36:607613
20 Uematsu M, Shioda A, Suda A, et al. Computed tomography-
guided frameless stereotactic radiotherapy for stage I non-
small cell lung cancer: a 5-year experience. Int J Radiat Oncol
Biol Phys 2001; 51:666670
21 Bush DA, Slater JD, Bonnet R, et al. Proton-beam radiother-
apy for early-stage lung cancer. Chest 1999; 116:13131319
22 Lad TE. Postsurgical adjuvant therapy in stages I, II, and IIIA
non-small cell lung cancer. Hematol Oncol Clin North Am
1990; 4:11111119
23 Silvestri GA, Handy J, Lackland D, et al. Specialists achieve
better outcomes than generalists for lung cancer surgery.
Chest 1998; 114:675680
24 Laroche C, Wells F, Coulden R, et al. Improving surgical
resection rate in lung cancer. Thorax 1998; 53:445449
25 Schroen AT, Detterbeck FC, Crawford R, et al. Beliefs
among pulmonologists and thoracic surgeons in the therapeu-
tic approach to non-small cell lung cancer. Chest 2000;
118:129137
26 Bach PB, Cramer LD, Schrag D, et al. The influence of
hospital volume on survival after resection for lung cancer.
N Engl J Med 2001; 345:181188
27 Jensik RJ, Faber LP, Kittle CF. Segmental resection for
bronchogenic carcinoma. Ann Thorac Surg 1979; 28:475483
28 Errett LE, Wilson J, Chiu RC, et al. Wedge resection as an
alternative procedure for peripheral bronchogenic carcinoma
in poor-risk patients. Thorac Cardiovasc Surg 1985; 90:656
661
29 Temeck BK, Schafer PW, Saini N. Wedge resection for
bronchogenic carcinoma in high-risk patients. South Med J
1992; 85:11083
30 Weissberg D, Straehley CJ, Scully NM, et al. Less than lobar
resections for bronchogenic carcinoma. Scand Thorac Car-
diovasc Surg 1993; 27:121126
31 Warren WH, Faber LP. Segmentectomy vs lobectomy in
patients with stage I pulmonary carcinoma. Thorac Cardio-
vasc Surg 1994; 107:10871094
32 Martini N, Bains MS, Burt ME, et al. Incidence of local
recurrence in second primary tumors in resected stage I lung
33 Ginsberg RJ, Rubinstein LV. A randomized comparative trial
of lobectomy vs limited resection for patients with T1NO
non-small cell lung cancer. Lung Cancer 1995; 7:8388
34 Giudicelli R, Thomas P, Lonjon T, et al. Video-assisted
minithoracotomy vs muscle-sparing thoracotomy for perform-
ing lobectomy. Ann Thorac Surg 1994; 58:712717
35 Kirby TJ, Rice TW. Thoracoscopic lobectomy. Ann Thorac
Surg 1993; 56:784786
36 Lewis RJ. The role of video-assisted thoracic surgery for
carcinoma of the lung: wedge resection to lobectomy by
simultaneous individual stapling. Ann Thorac Surg 1993;
56:762768
37 McKenna RJ Jr. Lobectomy by video-assisted thoracic sur-
gery with mediastinal node sampling for lung cancer. Thorac
38 Roviaro G, Varoli F, Vergani C, et al. Video-assisted thora-
coscopic surgery (VATS) major pulmonary resections: the
Italian experience. Semin Thorac Cardiovasc Surg 1998;
10:313320
39 Nomori H, Horio H, Naruke T, et al. What is the advantage
of a thoracoscopic lobectomy over a limited thoracotomy
procedure for lung cancer surgery? Ann Thorac Surg 2001;
72:879884
40 Keller SM, Adak S, Wagner H, et al. Mediastinal lymph node
dissection improves survival in patients with stages II and IIIa
non-small cell lung cancer. Eastern Cooperative Oncology
Group. Ann Thorac Surg 2000; 70:358365
41 Izbicki JR, Thetter O, Habekost M, et al. Radical systematic
mediastinal lymphadenectomy in non-small cell lung cancer:
a randomized controlled trial. Br J Surg 1994; 81:229235
42 Sugi K, Nawata K, Fujita N, et al. Systematic lymph node
dissection for clinically diagnosed peripheral non-small-cell
lung cancer less than 2 cm in diameter. World J Surg 1998;
22:290294
43 Ohta Y, Oda M, Wu J, et al. Can tumor size be a guide for
limited surgical intervention in patients with peripheral non-
small cell lung cancer? Assessment from the point of view of
nodal micrometastasis. Thorac Cardiovasc Surg 2001; 122:
900906
44 Takizawa T, Terashima M, Koike T, et al. Lymph node
metastasis in small peripheral adenocarcinoma of the lung.
45 Liptay MJ, Masters GA, Winchester DJ, et al. Intraoperative
radioisotope sentinel lymph node mapping in non-small cell
46 Little AG, DeHoyos A, Kirgan DM, et al. Intraoperative
lymphatic mapping for non-small cell lung cancer: the senti-
nel node technique. Thorac Cardiovasc Surg 1999; 117:220
234
47 Rowell NP, Williams CJ. Radical radiotherapy for stage I/II
non-small cell lung cancer in patients not sufficiently fit for or
declining surgery (medically inoperable): a systematic review.
Thorax 2001; 56:628638
48 Thirion P, et al. Three-dimensional conformal radiation ther-
apy (3DCRT) permits radiobiological dose escalation for
non-small cell lung cancer (NSCLC): preliminary results of a
phase I/II trial [abstract]. Proc Am Soc Clin Oncol 2001;
20:344a
2003;123;181-187 Chest
W. Roy Smythe
Treatment of Stage I Non-small Cell Lung Carcinoma
& Services
Updated Information
S
References
S#BIBL
free at:
Citations
S#otherarticles
Reprints
2003;123;188-201 Chest
Walter J. Scott, John Howington and Benjamin Movsas
Treatment of Stage II Non-small Cell Lung Cancer
ISSN: 0012-3692.
Treatment of Stage II Non-small Cell
Lung Cancer*
Walter J. Scott, MD, FCCP; John Howington, MD, FCCP; and
Benjamin Movsas, MD
Based on clinical assessment alone, patients with stage II non-small cell lung cancer (NSCLC)
comprise only 5% of all patients with NSCLC. In addition, patients with stage II NSCLC
represent a heterogeneous group, since stage II consists of patients with T1-2N1 or T3N0 tumors.
By definition, patients with tumor invading the chest wall apex, mediastinum, diaphragm, or even
the mainstem bronchus may all have T3 tumors. The extent of the data available regarding
treatment of each of these different groups is therefore limited. The quality of the data is limited
as well, because information often comes from small series of patients. Studies of adjuvant
therapy after complete resection of stage II NSCLC are an important exception to this
generalization, since data from large, randomized studies of adjuvant radiation therapy, chemo-
therapy, or a combination of the two are available for analysis. Superior sulcus tumors are
discussed elsewhere in these guidelines. (CHEST 2003; 123:188S201S)
Key words: carcinoma; bronchogenic; chemotherapy; guidelines; lung; neoplasms; neoplasm; staging; practice
guidelines; radiotherapy; surgery
Abbreviations: CI confidence interval; LCSG Lung Cancer Study Group; NSCLC non-small cell lung cancer
A
ccording to the current International Union
Against Cancer and American Joint Committee
for Cancer Staging system, stage II non-small cell
lung cancer (NSCLC) is defined as a T1 or T2 cancer
with N1 nodal metastasis and no distant metastasis
(T1-2N1M0) or a T3 cancer with no nodal or distant
metastasis (T3N0M0).
1
Definitions
Stage IIA consists of T1N1 cancers. For review,
T1 tumors by definition are 3 cm in size and do
not involve the visceral pleura or a main bronchus.
N1 denotes metastasis to ipsilateral peribronchial
and/or ipsilateral hilar lymph nodes, and intrapulmo-
nary nodes involved by direct extension of the pri-
mary tumor.
2
Stage IIB includes T2N1 and T3N0
cancers. T2 denotes a tumor with any of the follow-
ing features: 3 cm in greatest dimension, involves
a main bronchus 2 cm distal to the carina, invades
the visceral pleura, or is associated with atelectasis or
obstructive pneumonitis that extends to the hilar
region but does not involve the entire lung. T3
denotes a tumor of any size that directly invades any
of the following: chest wall (including superior sulcus
tumors), diaphragm, mediastinal pleura, parietal
pleura, parietal pericardium, or tumor in the main
bronchus 2 cm distal to the carina, but without
involvement of the carina, or associated atelectasis or
obstructive pneumonitis of the entire lung.
2
The Current Staging System
Stage IIA is an uncommon presentation for
NSCLC, representing 1 to 5% of patients treated in
most recent surgical series.
36
In the article by
Mountain
2
on revision of the lung cancer staging
system, only 5% of patients had clinical N1 disease
and only 7% were found to have pathologic N1
disease. Mountain
2
noted the infrequent nature of
clinical stage IIA cancer and the tendency for stage
migration to pathologic stage IIA cancer. Stage IIB
cancers, in contrast, represent approximately 15 to
25% of resected cancers in several large surgical
series.
35
Several authors have published data supporting
the current staging system for NSCLC (Table 1).
36
Inoue and colleagues
3
evaluated the prognoses of
1,310 surgically resected patients with NSCLC. All
patients underwent complete surgical resection and
*From the Department of Surgical Oncology (Dr. Scott), Section
of Thoracic Surgical Oncology; the Department of Radiation
Oncology (Dr. Movsas), Fox Chase Cancer Center, Philadelphia,
PA; and Department of Surgery (Dr. Howington), Division of
Thoracic Surgery, University of Cincinnati Medical Center,
Cincinnati, OH.
Correspondence to: Walter J. Scott, MD, FCCP, Fox Chase
Cancer Center, 7701 Burholme Ave, Philadelphia, PA 19111;
e-mail: WScott@fccc.edu
systematic nodal dissection in a period from 1980
through 1993. Routine chest and brain CT as well as
bone scan was performed prior to surgical resection.
They confirmed a significant difference in survival
between patients with T1N1M0 (stage IIA) cancers
(57%) and patients with T2N1M0 (stage IIB) can-
cers (42%). They found no significant difference
between patients with T2N1M0 and T3N0M0 can-
cers. Patients with pathologic T1N1M0 tumors made
up only 4.4% of the 1,310 resected patients, while
14.3% of patients were pathologically staged IIB
patients.
3
A retrospective review of 2,361 patients treated
with surgical resection of NSCLC in stages I, II, and
IIIA was reported by van Rens and colleagues.
4
None of the patients received prior treatment for
NSCLC. Resection was considered complete in
89.9% of the patients. Again, only 3.6% of patients
had stage IIA cancers. They found a significant
difference in 5-year survival between patients with
T1N1M0 (IIA) cancer (52%) and patients with
T2N1M0 (IIB) cancer (33%). They found no differ-
ence in survival between patients with T2N1M0
cancers and patients with T3N0M0 cancers.
4
Like Mountain
2
and Inoue et al
3
, van Rens and
colleagues
4
found no significant difference in 5-year
survival between patients with stage IB cancers and
patients with IIA cancers. In addition, both van Rens
et al
4
and Inoue et al
3
failed to find a significant
survival difference between patients with T3N0M0
cancers and patients with T3N1M0 cancers.
Adebonojo and colleagues
5
reported on the results
of surgical management of lung cancer at a military
medical center. The authors performed a retrospec-
tive review of 552 patients who underwent surgical
resection with curative intent at Walter Reed Army
Medical Center between January 1984 and Decem-
ber 1996. Thoracic lymphadenectomy was not rou-
tinely performed, but mediastinal sampling was ob-
tained from at least four mediastinal stations for
pathologic staging. They also found a significant
difference in the survival rate between patients with
T1N1M0 (IIA) and patients with T2N1M0 (IIB)
cancers (57% and 48%, respectively). There was
no significant difference between patients with
T2N1M0 and T3N0M0 cancer. They also noted the
uncommon nature of stage IIA cancer, with only 17
of the 552 patients (3%) having T1N1M0 cancers.
5
Jassem and colleagues
6
analyzed the survival data
of 586 patients who underwent complete pulmonary
resection and pathologic staging at one institution.
Only six patients (1%) were identified as having stage
IIA cancer. This prevented evaluation of any signif-
icant difference in survival among patient with stage
IIA and stage IIB cancers; however, there was no
significant difference between survival of patients
with T2N1M0 and T3N0M0 cancers (27% and 30%,
respectively).
6
The studies by Inoue et al,
3
van Rens et al,
4
and
Adebonojo et al
5
confirmed the finding of Mountain
2
of a statistically significant difference in 5-year sur-
vival between patients with stage IIA (T1N1M0)
cancers and patients with stage IIB (T2N1M0 and
T3N0M0) cancers. The study by Jassem et al
6
had
too few patients with stage IIA cancers, along with a
strikingly low 17% 5-year survival (one of six pa-
tients) to evaluate for any significant difference
between stage IIA patients and other stage group-
ings. All four studies have found a significant differ-
ence in survival between patients with T3N0M0
cancer and patients with T1-3N2M0 cancer, sup-
porting the movement of T3N0M0 from stage IIIA
to stage IIB.
36
Several authors found no significant difference in
survival between patients with T3N1M0 (IIIA) can-
cers and patients with stage IIB (T3N0 and T2N1)
cancers (Table 2).
3,4,6
However, others, including
Downey and colleagues,
7
have identified N1 disease
as a significant factor in decreased survival in pa-
tients with T3 NSCLC.
8,9
In one series, the survival
at 5 years was 49% for 100 T3N0 patients, compared
to a 27% 5-year survival among 24 patients with
T3N1 NSCLCs.
7
N1 nodal disease represents patients with cancer
metastasis or direct extension to subsegmental, seg-
mental, and lobar lymph nodes (levels 14, 13, and
12), as well as metastasis or direct extension into
interlobar (level 11) and hilar (level 10) lymph nodes.
The presence of N1 nodal disease denotes a group of
patients with intermediate survival. Several studies
have compared the survival results of patients with
lobar (levels 12 to 14) nodal disease with patients
with extralobar/hilar (levels 10 and 11) nodal disease.
Yano and colleagues
10
retrospectively reviewed
the outcomes of 78 patients with pathologic N1
disease who underwent complete resection with
mediastinal lymph node dissection. The overall
5-year survival was 49.2%. The pathologic T stage,
the type of resection, or the addition of adjuvant
Table 1Five-Year Survival Based on Pathologic
Stage II Grouping*
Source
Stage IIA Survival,
pT1N1M0
Stage IIB Survival
pT2N1M0 pT3N0M0
Mountain
2
76 (55) 288 (39) 87 (38)
Inoue et al
3
57 (57) 141 (42) 46 (34)
van Rens et al
4
84 (52) 541 (33) 137 (33)
Adebonojo et al
5
17 (57) 50 (48) 42 (47)
Jassem et al
6
6 (17) 64 (24) 55 (30)
*Data are presented as No. (%).
therapy did not significantly impact survival in this
group of pathologic N1 patients. Lobar nodes (levels
12 and 13) were involved in 38.5% of patients, and
extralobar/hilar nodes (levels 10 and 11) were in-
volved in 61.5% of patients. The survival of lobar N1
disease was significantly better than that of extralobar/
hilar disease (64.5% vs 39.7% at 5 years, p 0.014). A
multivariate analysis revealed that the level of N1
disease (lobar or extralobar) was the only prognostic
factor for patients with pathologic N1 disease.
van Velzen and colleagues
11
analyzed the out-
comes of 391 patients with pathologic T2N1M0
NSCLC. Lymph node involvement in the 391 pa-
tients was by metastasis in 218 patients (55.8%) and
direct extension in 173 patients (44.2%). The cumu-
lative 5-year survival rate for hospital survivors
(n 369) was 37.8%. The 5-year survival for patients
with lobar (levels 12 and 13) metastases was signifi-
cantly better than that of patients with extralobar/
hilar metastases (57.3% vs 30.3%, p 0.0028). In
addition, they found patients without visceral pleura
invasion had a significantly better prognosis than
patients with pleural invasion (survival at 5 years,
43.9% vs 31.1%; p 0.0093).
In a review of 256 patients with pathologic N1
disease treated with complete surgical resection and
mediastinal lymph node dissection, Riquet and col-
leagues
12
reported an overall 5-year survival rate of
47.5%. Survival was not related to the pathologic T
factor, type of resection, number of N1 stations
involved, nor to type of lymph node involvement
(direct extension or metastases). Five-year survival
was significantly better for patients with lobar (levels
12 and 13) metastases (53.6%) compared to patients
with extralobar/hilar (levels 10 and 11) metastases
(38.5%, p 0.02).
Several retrospective series have noted a de-
creased survival in patients with central T3 lung
cancers involving the diaphragm or mediastinum
compared to patients with cancers involving the
lateral chest wall or the superior sulcus of the chest.
A retrospective review from Memorial Sloan Ketter-
ing reported by Burt et al
13
found a 10% 5-year
survival among patients with pathologic T3N0M0
cancers where the mediastinum was involved by
direct tumor extension. In another retrospective re-
view, Pitz and colleagues
14
reported a greater mean
5-year survival in patients with T3 cancers involving the
main bronchus (40%) compared to mean 5-year sur-
vival among patients with T3 cancers with invasion of
mediastinal structures (25%). This difference failed to
reach statistical significance (p 0.05).
Both Inoue et al
3
and Adebonojo et al
5
noted a
worse prognosis for patients with T3N0M0 cancers
involving the pericardium or diaphragm compared to
the chest wall. The numbers in both studies are too
small for statistical significance, and both authors
recommend further evaluation with a larger cohort
of patients.
3,5
In a brief communication, Weksler et al
15
reported
on the rare occurrence of T3 tumors involving the
diaphragm, and the poor survival in these patients
when N2 nodal disease was identified. The eight
patients in this series represented just 0.17% of the
patients operated on for lung cancer during the
21-year period in which these eight patients pre-
sented for treatment.
T3 cancers with invasion of the mediastinum most
often do poorly if treated by surgical resection alone.
This is explained in part by the high number of
incomplete resections as noted in the retrospective
review by Martini and colleagues,
16
in which 20 of
58 patients (34%) with T3 cancer involvement of the
mediastinum underwent an incomplete resection. It
has been well established that incomplete resection
is a predictor of decreased survival.
Treatment Guidelines Stage II NSCLC
(T1-2N1M0)
Intraoperative Management: Sleeve Resection vs
Pneumonectomy
No randomized trials comparing sleeve lobectomy
with pneumonectomy have been reported in the
literature. The data available consist of retrospective
reviews of the outcomes in patients treated with
sleeve lobectomy and compared with matched or
unmatched control subjects treated with pneumo-
nectomy.
Table 25-Year Survival Based on Pathologic Staging*
Source
Stage IIB Survival Stage IIIA Survival
Pathologic T2N1M0 Pathologic T3N0M0 Pathologic T3N1M0 Pathologic T1-3N2M0
Mountain
2
288 (39) 87 (38) 55 (25) 344 (23)
Inoue et al
3
141 (42) 46 (34) 39 (38) 252 (n/a)
van Rens et al
4
541 (33) 137 (33) 89 (25) 261 (16)
Jassem et al
6
64 (24) 55 (30) 26 (35) 128 (7)
*Data are presented as No. (%). n/a authors did not provide data for all N2 patients.
Gaissert and colleagues
17
at the Massachusetts
General Hospital reviewed their experience with 72
consecutive patients treated with sleeve lobectomy
for NSCLC. They compared the results in this
cohort to an unmatched group of patients undergo-
ing pneumonectomy for lung cancer between 1986
and 1990. The actuarial survival at 5 years was 42%
for the patients undergoing sleeve lobectomy and
44% for patients undergoing pneumonectomy. Local
recurrence occurred in 14% of the patients under-
going sleeve lobectomy, while the local recurrence
was not reported for the unmatched pneumonec-
tomy cohort. The majority of the local recurrences
occurred in the mediastinum. There was a higher
mortality (9% vs 4%) and major complication rate
(16% vs 11%) among patients undergoing pneumo-
nectomy.
In a retrospective review of 29 patients undergo-
ing sleeve lobectomy for lung cancer and compared
with a matched cohort of patients undergoing pneu-
monectomy, Yoshino et al
18
found no difference in
the 3-year disease-free survival between the two
groups. In addition, local recurrence occurred in 1 of
15 patients undergoing sleeve lobectomy with N1
disease and in 2 of 11 patients undergoing pneumo-
nectomy with N1 disease. Operative mortality (6.9%
vs 0%) and complication rates (24.1% vs 13.7%) were
significantly higher, respectively (p 0.05), in the
group undergoing pneumonectomy compared to pa-
tients undergoing sleeve lobectomy.
Okada et al
19
compared the outcomes after sleeve
lobectomy and pneumonectomy for patients with
NSCLC distributed according to their nodal involve-
ment status. Between June 1984 and December
1998, 151 patients underwent sleeve lobectomy
while 60 patients underwent pneumonectomy. A
matched group of 60 patients undergoing sleeve
lobectomy was compared with the patients undergo-
ing pneumonectomy. The operative mortality rate
was 0% in the sleeve lobectomy group and 2% in the
pneumonectomy group. Local recurrence developed
in five patients (8%) after sleeve lobectomy and six
patients (10%) after pneumonectomy. Patients un-
dergoing sleeve lobectomy had a significantly longer
5-year survival (48%) than patients undergoing
pneumonectomy (29%). While these studies are
limited by their retrospective method and small
numbers of patients, the authors agree with the
conclusions of these articles that sleeve lobectomy is
preferred over pneumonectomy whenever a com-
plete pathologic resection can be obtained using
bronchoplastic techniques.
Recommendation
1. For patients with N1 lymph node metastases
in whom a complete resection can be achieved
with either technique, sleeve lobectomy is
recommended over pneumonectomy. Level of
evidence, poor; benefit, moderate; grade of
recommendation, C
Adjuvant Radiotherapy for T1-2N1 NSCLC
As noted above, completely resected patients with
stage II NSCLC have a lower 5-year survival rate
than patients with earlier stage disease. In most of
these patients, especially those who have N1 lymph
node metastases, the final pathologic stage is only
determined after histologic analysis of the resected
specimen (ie, intraoperatively or postoperatively).
Therefore, adjuvant therapy to prevent recurrence
and improve survival rates has been studied exten-
sively in this setting.
Several trials randomized patients with completely
resected NSCLC to postoperative radiotherapy or
surgery alone. Data for patients with N1 or stage II
NSCLC are shown in Table 3. Lung Cancer Study
Group (LCSG) trial 773 randomized 230 patients
after complete resection of stage II or III squamous
cell carcinoma to receive 50 Gy postoperatively or to
surgery alone.
20
Two thirds had stage II disease, and
approximately 75% had N1 tumors. For the entire
study population, no difference in survival was noted
for those receiving adjuvant therapy compared with
Table 3Randomized Clinical Trials of Surgery Alone vs Surgery Plus Adjuvant Radiotherapy in Patients With
Completely Resected Stage II NSCLC*
Source
Patients,
No. Stage
Radiotherapy
Dose, Gy
5-yr Survival, % Local Recurrence, %
Surgery
Surgery Plus
Radiotherapy p Value Surgery
Surgery Plus
Radiotherapy p Value
Weisenberger
20
210 II, III 50 37 38 NS 19 1 0.001
Stephens
21
183 N1 40 25 18 NS 49 40 NS
Dautzenberg et al
22
180 II 60 50 24 0.003
Feng et al
23
191 N1 60 47 56 0.288 31 5 0.05
Mayer et al
24
67 N1 5056 22 42 NS 26 8 NS
*NS not significant.
those receiving surgery only. However, there was a
decrease in local recurrence (defined as a first
recurrence in the ipsilateral lung or mediastinum) in
the group receiving postoperative radiotherapy. Only
1% of first recurrences were local in the radiotherapy
group, compared with 19% in the control group
(p 0.001).
The Medical Research Council trial compared
surgery alone to surgery followed by postoperative
radiotherapy (40 Gy in 15 fractions) in patients with
pathologically staged T1-2N1-2M0 NSCLC.
21
Pa-
tients were stratified according to TNM classifica-
tion, and 183 patients had N1 disease. In a subgroup
analysis, no differences were seen in patients staged
N1 with respect to survival (p 0.26). When sus-
pected and definite local recurrences were ana-
lyzed together, there was no clear evidence that
radiotherapy was beneficial. An analysis of the time
to definite local recurrence, however, did show a
significant advantage in the radiotherapy study arm
(p 0.04). This discrepancy may reflect the diffi-
culty in accurately defining the presence of local
failure, particularly in the setting of evolving fibrotic
changes after radiotherapy.
In a large (n 366) randomized trial from China,
radiotherapy was administered postoperatively to
60 Gy in 30 fractions vs observation.
12
Among the
191 patients with pathologic N1 disease, there was a
significant decrease in local recurrence from 31 to
5% (p 0.05). The 5-year survival rates marginally
(but not statistically) favored radiotherapy: 56% vs
47%. This study, however, has been criticized in that
results were analyzed only by the treatment deliv-
ered, rather than by the intent to treat.
Dautzenberg et al
22
reported the results of a
multicenter (Groupe dEtude et de Traitement des
Cancers Bronchiques), randomized trial of postop-
erative radiotherapy (60 Gy) compared to surgery
alone in 728 patients with resected NSCLC. The
study included 180 patients with stage II NSCLC.
This study reported an overall decrease in survival in
the postoperative radiotherapy group (43% vs 30%,
p 0.002) with no difference in local recurrence
rates. In a subgroup analysis, mortality was also
found to be higher in patients with stage II NSCLC
who underwent radiotherapy (50% vs 24%,
p 0.003), although there was no difference noted
in local control rates. Unlike the other studies listed
in Table 3, this is the only one that showed a
decrease in 5-year survival for the radiotherapy study
arm compared to the control study arm. This has
been directly related to the higher daily radiotherapy
doses used by some of centers that participated in
the study.
A meta-analysis from the Medical Research Coun-
cil combined data from nine randomized trials (2,128
patients) that compared surgery alone with surgery
followed by postoperative radiotherapy.
25
For all
patients, postoperative radiotherapy was associated
with an absolute decrease in survival of 7% at 2 years.
The adverse effect of postoperative radiotherapy on
survival was greatest for patients with stage I/II,
N0-N1 disease. The problems noted above for the
study of Dautzenberg et al
22
are even more applica-
ble to this meta-analysis. Indeed, the trial of Daut-
zenberg et al
22
made up approximately one third of
the patients in the meta-analysis. There was a wide
heterogeneity in the doses and fractionation sched-
ules used in the trials included in the postoperative
radiotherapy meta-analysis. The dose fractionation
schema ranged from a total of 30 Gy in 10 fractions
to 60 Gy in 30 fractions. Recent trials suggest that
the potential lethal effects of radiotherapy cited in
the postoperative radiotherapy meta-analysis are
rarely seen today with modern radiotherapy equip-
ment and techniques.
26,27
Because the risk-benefit ratio of adjuvant therapy
in the setting of pathologic N1 disease is controver-
sial, other factors may help in making treatment
decisions in this situation. For example, Yano et al
10
reported a 5-year survival of 40% for patients with
hilar N1 disease compared to 65% for those with
lobar N1 disease (p 0.014). Since the survival of
patients with lobar N1 disease is similar to that of
patients who are pathologic N0, even the local
control benefit of adjuvant radiotherapy is likely to
be very small in this group of patients.
Recommendations
2. For patients who have undergone complete
resection of stage II NSCLC with N1 lymph
node metastases (stage II [N1] NSCLC), rou-
tine administration of postoperative radiation
therapy with a goal of improving survival is not
recommended. Level of evidence, fair; benefit,
none/negative; grade of recommendation, D
node metastases (stage II [N1] NSCLC),
routine administration of postoperative ra-
diation therapy (ie, adjuvant radiation ther-
apy) decreases local recurrence rates. For
patients who have undergone complete re-
section of stage II NSCLC with N1 lymph
node metastases (stage II [N1] NSCLC),
routine administration of postoperative ra-
diation therapy could be used if the goal is
to decrease local recurrence with the under-
standing that survival will not be improved.
Level of evidence, fair; benefit, small; grade
of recommendation, C
node metastases (stage II [N1] NSCLC), the
evidence indicates that adjuvant radiotherapy
improves local control but does not increase
survival. An overall recommendation cannot
be made regarding routine use of adjuvant
radiotherapy in this setting. Level of evidence,
fair; benefit, none/negative; grade of recom-
mendation, D
Adjuvant Chemotherapy
Since most patients with completely resected early
stage NSCLC develop distant (as opposed to local)
recurrences, the administration of postoperative sys-
temic therapy has been tested in a number of
different trials, including randomized trials.
28,29
In
addition, two meta-analyses have been published.
A LCSG study (LCSG 772) randomized 141 pa-
tients with resected stage II and II adenocarcinoma
and large cell undifferentiated carcinoma to re-
ceive postoperative combined chemotherapy with
cyclophosphamide, doxorubicin, and cisplatin, or
postoperative immunotherapy (intrapleural bacille
Calmette-Gue rin and levamisole). Holmes
28
re-
ported 130 evaluable patients and noted a significant
difference in recurrence-free survival in the chemo-
therapy study arm, although overall survival between
the two groups was not significantly different
(p 0.113).
The Non-small Cell Lung Cancer Collaborative
Group reported a meta-analysis using updated pa-
tient data on individual patients from 52 randomized
clinical trials.
29
Data on adjuvant chemotherapy
compared to surgery alone were available from 14
trials (4,357 patients and 2,574 deaths). Based on
data from five trials, the meta-analysis clearly showed
that the hazard ratio for death was greater in patients
who received alkylating agents postoperatively com-
pared to surgery alone. Analysis of eight more recent
trials of cisplatin-based combination chemotherapy
(either with cyclophosphamide, doxorubicin, and
cisplatin, or cisplatin and vindesine) yielded a hazard
ratio for treated patients of 0.87 equivalent to a 13%
reduction in the risk of death (p 0.08). This sug-
gested that adjuvant cisplatin-based chemotherapy
conferred an absolute survival benefit of 3% at
2 years and 5% at 5 years, with the 95% confidence
intervals (CIs) being consistent with a 1% detriment
to a 10% benefit in 5-year survival. Three other trials
included other drug regimens, either tegafur, or
tegafur plus uracil. Based on limited numbers, the
data yielded a hazard ratio of 0.89 in favor of
chemotherapy (p 0.30). Few patients with stage II
NSCLC were enrolled (8.6% and 11.1%, respec-
tively) in the trials of tegafur plus uracil-based
therapy.
The data from randomized, prospective trials do
not clearly demonstrate a benefit from the use of
postoperative chemotherapy in patients with stage II
NSCLC. While meta-analysis suggested that
platinum-based postoperative regimens might pro-
vide a clinically significant survival benefit, the same
analysis suggested that postoperative chemotherapy
with alkylating agents might actually decrease sur-
vival of patients following complete resection. Prob-
lems with the older data include concerns about
staging methods, the lack of modern agents (such as
antiemetics and drugs to minimize hematologic tox-
icity) that minimize the side effects of chemother-
apy, and the fact that in many studies only one half of
the patients assigned to adjuvant chemotherapy com-
pleted the prescribed course of therapy. In addition,
the individual randomized clinical trials lacked the
statistical power to detect a survival advantage of 5 to
10%, an advantage that many authors have stated is
clinically important. All of these problems combined
may account for the fact that a survival benefit of
postoperative chemotherapy in patients with com-
pletely resected NSCLC has never consistently been
demonstrated. A better answer regarding the clinical
utility of adjuvant chemotherapy must await the
results of ongoing and recently completed random-
ized clinical trials.
Recommendation
resection of stage II NSCLC, administration of
postoperative chemotherapy should not be con-
sidered standard therapy at this time, and its
use should be limited to patients enrolled in
clinical trials. Level of evidence, good; benefit,
Adjuvant Chemotherapy and Radiotherapy
In order to decrease both the local recurrence rate
and the rate of development of distant metastases,
postoperative treatment with both chemotherapy
and radiation therapy has been compared to postop-
erative radiotherapy alone in patients with com-
pletely resected stage II NSCLC. Two recent trials
are important. The Groupe dEtude et de Traite-
ment des Cancers Bronchiques reported by Daut-
zenberg et al
30
randomized 267 patients (8 patients
with stage I, 70 patients with stage II, 189 patients
with stage III) to receive postoperative radiation
alone or to postoperative radiation and three courses
of cyclophosphamide, doxorubicin, cisplatin, vincris-
tine, and lomustine. Radiation therapy consisted of
60 Gy in 6 weeks for both groups. Once again, no
significant difference was seen in overall survival for
the study patients. When a subset analysis of N0/N1
patients was performed, no differences in the rates
of local or distant recurrences were noted, but
overall survival was better in the group receiving
radiation therapy alone (34% vs 17%, p 0.03). The
5-year survival rate of 17% is lower than one would
expect in a group of patients with mostly N1 disease,
however. The authors did not think this was due to
any toxic effects of the chemotherapy.
Keller et al
31
reported the results of Eastern
Cooperative Oncology Group trial 3590, a random-
ized trial of postoperative radiotherapy alone com-
pared to concurrent chemotherapy and radiotherapy
in 488 patients following complete resection of stage
II and III NSCLC. The radiotherapy dose was
50.4 Gy administered in 28 daily fractions. Chemo-
therapy consisted of cisplatin and etoposide admin-
istered concurrently. Survival was not prolonged by
concurrent therapy compared to radiotherapy alone
(median survival, 38 months and 39 months, respec-
tively; p 0.56). The risk of intrathoracic recurrence
was also not decreased by concurrent therapy when
compared to radiotherapy alone.
A meta-analysis noted previously
29
also included a
review of the available data on postoperative radio-
therapy compared to postoperative radiotherapy and
chemotherapy. Seven trials (807 patients and 619
deaths) were included in the analysis. Six of these
trials used a platinum-based chemotherapy regimen.
Total planned doses of radiation ranged from 40 Gy
in 10 fractions to 65 Gy in 33 fractions. The delay
between surgery and first postoperative treatment
was scheduled to be no more than 7 weeks. Two of
the trials included some patients with complete
resections, and two trials limited enrollment only to
patients with incomplete resections. The hazard ratio
for platinum-based trials was 0.94 (p 0.46). The
95% CI ranged from a 3% detriment to an 8%
benefit at 5 years. A benefit from combining post-
operative radiotherapy and chemotherapy could not
be clearly identified.
Recommendation
6. In patients who have undergone complete or
incomplete resection of stage II NSCLC, post-
operative combined chemotherapy and radio-
therapy should not be considered standard
therapy at this time, and their use should be
limited to patients enrolled in clinical trials.
Level of evidence, good; benefit, none/nega-
Neoadjuvant (Induction) Therapy for Stage II (N1)
NSCLC
Administering systemic therapy preoperatively for
patients with resectable NSCLC and who are at high
risk for recurrence has the following potential advan-
tages over postoperative administration of systemic
therapy: (1) patients are more likely to complete the
prescribed course of therapy; (2) chemotherapy will
have a greater effect on the primary tumor while its
blood supply is still intact; (3) occult distant disease
will be treated sooner; and (4) surgical resection may
be easier once the tumor has decreased in size
(downstaging of the primary and any N1 nodal
metastases). Therefore, a phase II trial (the Bimo-
dality Lung Oncology Team Trial) was initiated in
the United States to determine if preoperative che-
motherapy followed by surgery was effective and
safe.
The Bimodality Lung Oncology Team Trial was
conducted jointly at Memorial Sloan Kettering Can-
cer Center and M.D. Anderson Cancer Center.
32
Patients with clinical stage T2N0, T1-2N1, and
T3N0-1 NSCLC (all with negative mediastinoscopy
findings) received two cycles of paclitaxel and carbo-
platin therapy (paclitaxel, 225 mg/m
2
over 3 h, and
carboplatin [area under the curve, 6] every 21 days)
preoperatively. Patients who progressed following
preoperative chemotherapy did not undergo resec-
tion and were treated off-study. Three postoperative
cycles of the same chemotherapy were planned for
patients undergoing complete resection. Patients
with superior sulcus tumors were excluded.
There were 94 patients enrolled in the study. At
the time of enrollment, the pretreatment clinical
stage of the patients enrolled in this study was T2N0
in 42 of 94 patients (45%), T1N1 in 1 of 94 patients
(1%), T2N1 in 27 of 94 patients (29%), T3N0 in 17
of 94 patients (18%), and T3N1 in 7 of 94 patients
(7%). Of the patients (all stages combined) receiving
induction chemotherapy, 56% had a major objective
response and 86% underwent complete resection.
Patients not undergoing operation had disease
progression (n 3), were clinically unresectable
(n 1), died (n 1), and were unavailable for fol-
low-up (n 1). Six pathologic complete responses
(6%) were observed. Ninety patients (96%) received
the planned preoperative chemotherapy vs 45%
receiving postoperative chemotherapy.
The authors reported no unexpected chemotherapy-
related or surgical mortality and morbidity. They noted
10 episodes of grade III respiratory infection and
another 10 episodes of grade III toxicity listed as lung
(other), raising the question of whether some possible
increase in toxicity resulted from combining preopera-
tive chemotherapy and surgery. A retrospective study
by Siegenthaler et al
33
concluded that preoperative
chemotherapy and surgery did not affect overall mor-
bidity and mortality. Roberts and coauthors
34
came to
the opposite conclusion. But generally, the results of
this phase II trial have been interpreted as proving
feasibility and efficacy of the preoperative regimen of
paclitaxel and carboplatin. The phase III North Amer-
ican Intergroup trial (S9900) that randomizes patients
to surgery alone vs three cycles of preoperative chemo-
therapy followed by surgery has been enrolling patients
since late 1999.
The only randomized trial of preoperative che-
motherapy to include significant numbers of stage
II patients was reported by DePierre et al.
35
Approximately one half of the patients had clinical
stage IB or II NSCLC, with the remainder stage
IIIA (clinical stage T1N0 patients were excluded).
Patients were randomized to undergo surgery or to
receive two cycles of chemotherapy (mitomycin, 6
mg/m
2
on day 1; ifosfamide, 1.5 gm/m
2
on days 1
to 3; and cisplatin, 30 mg/m
2
on days 1 to 3) 3
weeks apart before surgery. In both study arms,
patients with T3 or N2 disease received postoper-
ative radiotherapy (up to 60 Gy). A total of 355
patients were evaluable. A beneficial effect of
preoperative chemotherapy in terms of survival
was confined to the group of patients with N0 and
N1 disease, with a relative risk of 0.68 (95% CI,
0.49 to 0.96; p 0.027).
A phase II study and a phase III trial that
included stage IB and II patients suggest that
preoperative chemotherapy and surgery may en-
hance survival in patients with stage IB and II
NSCLC with acceptable toxicity; however, ques-
tions about the toxicity of induction chemotherapy
and surgery remain.
34
Despite promising results,
induction chemotherapy and surgery cannot be
considered standard therapy for patients with
stage IB, II, and T3N1 NSCLC until more expe-
rience has been gained. The phase III North
American Intergroup trial (S9900) comparing sur-
gery alone to induction paclitaxel and carboplatin
followed by surgery began enrolling patients in
1999.
Recommendation
7. For patients with stage T2N0, T1-2N1, and
T3N0 NSCLC, routine use of preoperative
chemotherapy followed by surgery should not
be considered standard therapy at this time,
and its use should be limited to patients en-
rolled in clinical trials. Level of evidence, poor;
tion, I
Treatment Guidelines Stage II NSCLC
(T3N0M0)
CT Assessment of Stage II (T3 [Chest Wall])
NSCLC
The ability of chest CT to predict the presence of
chest wall invasion by a lung tumor adjacent to the
chest wall has been investigated by Ratto et al
36
in a
study involving 112 patients. They determined that
the presence of a chest wall mass protruding through
the ribs on CT was the most sensitive predictor of
chest wall invasion, while a ratio of length of chest
wall contact/tumor diameter of 0.5 was the best
predictor of the absence of chest wall invasion. Other
studies have reported similar findings.
37,38
Radio-
graphic signs on chest CT other than a mass protrud-
ing thorough the ribs of the chest wall or gross rib
destruction are not sufficiently accurate to make a
diagnosis of chest wall invasion by a tumor adjacent
to or abutting the chest wall. MRI has not generally
been shown to provide any advantage over CT in
detecting involvement of the lateral chest wall by an
adjacent lung tumor.
39
Recommendation
8. For patients with lung tumors that abut or are
adjacent to the chest wall based on chest CT
(clinical T3 [chest wall] NSCLC), the presence
or absence of chest wall invasion should not be
assumed based on CT findings alone but should
be confirmed by surgical exploration. Level of
recommendation, C.
Resection of Stage II (T3 [Chest Wall]) NSCLC:
Choice of Operative Procedure
There have been a number of published reports of
the results of chest wall resection for NSCLC invad-
ing the parietal pleura or the chest wall. These were
all retrospective studies. In all of these studies, the
most important factor influencing survival following
resection of T3 (chest wall) tumors was completeness
of resection (Table 4). Downey and coauthors
7
ex-
Table 4Complete vs Incomplete Resection of T3
Chest Wall NSCLC
Source Patients, No.
5-yr Survival, %
Incomplete
Resection
Complete
Resection
Pitz et al
41
125 11 29
Ratto et al
42
112 0 16
Downey et al
7
334 4 32
Magdeleinat et al
43
201 13 24
plored 334 patients. Of those, 175 patients had R0
(margins negative on microscopic examination) re-
sections, 94 patients had R1 (margins positive on
microscopic examination) or R2 (margins grossly
positive) resections, and 65 patients underwent ex-
ploration without resection. Overall survival of pa-
tients undergoing R0 resections was 32%. The 5-year
survival of R0 patients with N0 disease was 49%. The
survival of incompletely resected patients was indis-
tinguishable from that of patients undergoing no
resection at all, with only 4% of incompletely re-
sected patients and 0% of unresected patients alive
after 3 years. The authors concluded that the most
striking finding . . . is that an incomplete resection,
even if only microscopic disease, offers the patient
no curative benefit. The type of resection that
results in a complete (R0) resection for a specific
patient is the one that should be performed.
Downey et al
7
based the choice of operation
(extrapleural resection vs en bloc resection) on care-
ful intraoperative assessment. If an extrapleural re-
section was attempted but the extrapleural plane
could not be developed easily, suggesting tumor
invasion into the chest wall, then en bloc resection
was performed. For 175 patients managed in this
fashion, and for whom a complete resection was
achieved, no survival difference was noted (5-year
survival of 36% for both groups, p 0.57). The
authors concluded that survival did not depend on
the extent of resection (en bloc vs extrapleural) as
long as a complete resection was achieved. Others
have reported that en bloc resection is associated
with better survival than extrapleural resection, even
for those T3 (chest wall) tumors whose depth of
invasion is limited to the parietal pleura only.
44
These data are not generally supported by the results
of larger series.
7,9
Ratto et al
42
reported that 19 patients in their
series underwent discontinuous resection, a situation
where extrapleural dissection was carried out initially
but a portion of chest wall was resected separately
during the same operation, usually out of concern
that the pleural margin was close or positive for
malignant cells. None of the 19 patients survived
longer than 30 months. Pitz et al
41
performed dis-
continuous resection in 7 patients. None of the seven
patients survived beyond 18 months. Based on lim-
ited data, it would seem that a discontinuous resec-
tion does not lead to a complete resection, which is
the most important determinant of survival in the
treatment of patients with T3 (chest wall) tumors.
Many authors (Table 5) have reported that the
morbidity of operations incorporating en bloc resec-
tion of T3 (chest wall) tumors is similar to that of
operations where an extrapleural resection is per-
formed.
7,4345
Therefore, there should be no reluc-
tance to perform this procedure when intraoperative
evaluation of T3 (chest wall) tumors suggests tumor
invasion beyond the parietal pleura.
In summary, completeness of resection is the most
important predictor of survival in patients undergo-
ing resection of T3 (chest wall) tumors. (Lymph
node status is the next most important factor; see the
next chapter on stage IIIA NSCLC.) The appropri-
ate operation is the one that yields a complete
resection (either extrapleural resection or en bloc
resection). Limited data suggest that discontinuous
resections are not equivalent to complete resections.
Data suggest that the mortality rate of operations
incorporating en bloc chest wall resection for T3
(chest wall) tumors is similar to that associated with
extrapleural resection.
Recommendations
9. For patients with T3 (chest wall) NSCLC that,
at the time of operation, may extend beyond
the parietal pleura, en bloc resection of T3
(chest wall) NSCLC must be performed un-
less one is confident that extrapleural invasion
does not exist. Long-term survival after surgi-
cal treatment for T3 (chest wall) NSCLC is
highly dependent on the completeness of
resection. Level of evidence, poor; benefit,
10. For patients with T3 (chest wall) NSCLC that
does not extend beyond the parietal pleura, an
extrapleural resection may be performed. As
long as a complete resection is achieved,
survival following extrapleural resection is
similar to that achieved following en bloc
11. At the time of surgery for T3 (chest wall)
NSCLC suspected of invading beyond the
Table 5Morbidity of Operations
Source
Patients,
No.
Operative Mortality, No. of
Patients/Total Patients (%)
En Bloc
Resection
Extrapleural
Resection
Albertucci et al
44
37 2/21 (9.5) 2/16 (12.5)
Harpole et al
45
47 0/47 (0) NA
Downey et al
7
175 10/175 (6)
Magdeleinat et al
43
201 3/79 (4%) 11/122 (9%)*
*p 0.2 for the difference in mortality between extrapleural and en
bloc resection.
Includes 25 patients with superior sulcus tumors.
NA no patients in this series underwent extrapleural resection.
En bloc, 92 patients; extrapleural, 80 patients; discontinuous, 3
patients; authors noted that en bloc resections and extrapleural
resection were associated with similar mortalities.
parietal pleura, separation of the tumor from
the chest wall followed by resection of that
portion of the chest wall that the tumor was
originally adherent to (ie, discontinuous or
non-en bloc resection) should be avoided.
Limited data suggest that a discontinuous
resection results in very inferior survival com-
pared to en bloc resection. Level of evidence,
poor; benefit, substantial; grade of recommen-
dation, C
Postoperative (Adjuvant) Radiotherapy for T3
(Chest Wall) NSCLC
No randomized trials have been performed that
compare surgery alone to surgery and adjuvant
radiotherapy for resected T3 (chest wall) NSCLC.
To determine the value of postoperative radiother-
apy in patients who have undergone chest wall
resection, one should evaluate well-staged patients
who have no lymph node involvement. The reported
experience is small.
7,46,47
Patterson et al
46
reported 35 patients with T3
(chest wall) tumors. Thirty patients underwent com-
plete resections, and 5 patients had incomplete
resections. Postoperative radiotherapy was adminis-
tered to 22 patients. The authors reported that the
patients who received postoperative radiotherapy
had a better 5-year survival (56% vs 30%, p value not
calculated). However, most recent studies suggest no
effect from radiotherapy in patients with completely
resected T3 (chest wall) NSCLC.
Subsequently, Piehler and colleagues
47
reported
93 patients operated on for lung cancer invading the
chest wall. Sixty-six patients underwent complete en
bloc resections, and 31 of those had T3N0 disease.
Sixteen of those were selected to receive postopera-
tive radiotherapy. The selection criteria were not
given. The actuarial survival at 5 years was the same
whether or not radiotherapy was administered
(53.3% vs 54.4%). Downey and colleagues,
7
in the
largest series to date, reported that the 5-year sur-
vival after complete resection of T3N0 chest wall
NSCLC was not different in 79 patients who under-
went surgery alone compared to 21 who received
postoperative radiotherapy (48% vs 53%, p 0.63).
Of note, these authors could not find any benefit
from postoperative radiotherapy in patients with
completely resected chest wall tumors who were
found to have N1 or N2 metastases at final staging.
Based on limited data, there seems to be no
survival advantage of postoperative radiotherapy for
patients who have undergone complete resection of
T3 (chest wall) NSCLC. The few studies that address
postoperative radiotherapy in patients who have
undergone an incomplete resection of T3 (chest
wall) NSCLC did not identify a survival advantage in
this group either.
42,48
A meaningful analysis cannot
be performed because of the small number of
patients to be analyzed.
Recommendations
resection of T3 (chest wall) NSCLC, routine
postoperative radiotherapy does not provide a
documented survival benefit and should not
be used in these patients. Level of evidence,
poor; benefit, none/negative; grade of recom-
mendation, I
13. In patients who have undergone incomplete
resection (or resection with a negative but
close margin) of T3 (chest wall) NSCLC,
postoperative radiotherapy may provide a sur-
vival benefit and can be used. Level of evi-
dence, poor; benefit, small/weak; grade of
recommendation, C
Treatment of T3 (Mediastinal) NSCLC
Approximately one half of patients with T3 tumors
have tumors that invade the lateral chest wall or apex
of the lung. The other half have tumors that either
invade the mediastinum, have grown to within 2 cm
of the carina, or less commonly invaded the dia-
phragm. As noted above, the prognosis for these
patients seems to be worse than for patients with
more peripheral tumors. Surgical resection provides
limited 5-year survival in most instances, with the
exception of tumors that invade or contact the
mediastinal fat or pericardium over a small area
(these are often discovered only at the time of
surgery). For patients with resected T3 (mediastinal)
NSCLC, the average 5-year survival of reported
series is about 25% (range, 9 to 37%).
3,13,16,41
No
controlled studies of adjuvant radiotherapy following
complete resection of T3 (mediastinal) NSCLC have
been reported. Martini et al
16
reported the results of
a series of patients (n 15) treated with interstitial
brachytherapy following incomplete resection of T3
(mediastinal) NSCLC. They reported a 5-year sur-
vival of 20% in the adjuvant radiotherapy group
compared to 7% in the group undergoing incomplete
resection only and 0% in the group treated with
radiation alone. This suggests that patients may
derive some benefit from adjuvant radiotherapy after
incomplete resection of T3 (mediastinal) NSCLC.
There are no data to support its use when a complete
resection has been performed.
Recommendations
14. For patients with T3 (mediastinal) NSCLC
who have undergone incomplete resection,
postoperative radiotherapy may be of benefit.
Level of evidence, poor; benefit, small; grade
15. Adjuvant radiotherapy for T3 (mediastinal)
NSCLC for completely resected patients
should not be performed. Level of evidence,
mendation, I
Patients with mainstem bronchial involvement are
usually reported in series of sleeve resections, often
mixed in with other stages. The range of 5-year
survival in reported series varies from 12 to
40%,
43,49,50
with two small series reporting 5-year
survival of 80% with T3 mainstem involvement
(see previously mentioned recommendations regard-
ing type of resection under treatment of T1-2N1
tumors).
40,51
The presence of N2 lymph node metastases sig-
nificantly affects survival (Table 6). This seems to be
especially true for patients with central T3 tumors,
based on the experience of groups operating on
patients with tumors classified as T3 because of
airway proximity.
41,49,50
Staples et al
52
found that the
prevalence of N2 metastases was 54% for central
tumors and 27% for peripheral tumors in 151 pa-
tients undergoing mediastinoscopy. Since the 5-year
survival of patients with T3N2 NSCLC is low, and
central tumors are more likely to have occult N2
metastases, mediastinoscopy should be performed in
patients with T3 central tumors prior to resection.
Recommendation
16. In all patients with centrally located clinical
T3 NSCLC, histologic assessment of medias-
tinal lymph nodes should be performed prior
to resection. Preoperative identification of N2
lymph node metastases precludes surgical re-
section as initial therapy in this setting. Level
of evidence, poor; benefit, substantial; grade
Some authors have recommended induction ther-
apy for patients with central T3 NSCLC and N2
metastases. Responders could then undergo resec-
tion. No data are available to evaluate this.
Summary
Stage II NSCLC consists of a number of different
groups of patients. Because of this and because only
5 to 10% of all patients with NSCLC are classified as
stage II, the quality of the evidence for certain
recommendations listed here is limited. Other limi-
tations of these guidelines include the fact that it is
not possible to include all of the varied clinical
situations that a practicing physician is likely to
encounter in a document such as this, and that
ongoing research may render sections of these guide-
lines obsolete not long after they are published.
Despite these limitations, several conclusions can be
drawn regarding the current treatment of patients
with stage II NSCLC and areas for future research.
Adjuvant therapy of any sort (radiotherapy, che-
motherapy, combined chemotherapy and radiother-
apy) has never proven to prolong survival following
complete resection of stage II NSCLC. Although a
number of randomized clinical trials have been
performed, the issue of postoperative radiotherapy is
still controversial because of the techniques used in
those trials. Modern methods for administering ra-
diotherapy may decrease the toxicity associated with
treatment while maintaining effective local control
capabilities. Postoperative radiotherapy has a role in
treating patients with close or positive margins after
resection. Ongoing clinical trials (and some recently
completed) will hopefully determine if currently
available chemotherapy agents in combination are
more effective than previous agents when used in the
adjuvant setting. Therefore, we believe that the use
of most adjuvant therapies should be limited to
patients enrolled in clinical trials.
Table 6Effects of N2 Lymph Node Metastases on Survival
Source Type of T3 Tumor % 5-yr Survival T3N2
Magdeleinat et al
43
Chest wall 21
Elia et al
9
Chest wall 0
Downey et al
7
Chest wall 15
Harpole et al
45
Chest wall (18 mo median survival)
Pitz et al
41
Chest wall 14
Albertucci et al
44
Chest wall 0
Ratto et al
42
Chest wall 0
Pitz et al
14
Mediastinum/main bronchus 0
Dartevelle et al
48
Main bronchus (0 of 8 patients with paratracheal N2)*
Nakahashi et al
40
All types 0
*n 13, subcarinal nodal involvement only, 34% at 3 yr; 3 of 13 patients survived up to 74 mo, 81 mo, and 90 mo, respectively.
Expanding the use of chemotherapy (alone or
combined with radiotherapy) into the preoperative
period (induction therapy or so-called neoadjuvant
therapy) has been a major theme for thoracic oncol-
ogy in the 1990s. Following some success in patients
with locally advanced (stage IIIA) NSCLC, the use
of multimodality approaches has been applied to
stage II, traditionally treated by surgery alone or
sometimes radiotherapy alone. Induction chemo-
therapy and radiotherapy has also been applied to
patients who have stage II (T3) NSCLC that is
centrally located. Evidence regarding the usefulness
of these approaches awaits the analysis of the results
of the ongoing phase III North American Intergroup
trial S9900 and a recently closed North American
Intergroup trial, Radiation Therapy Oncology Group
trial 9309.
Future areas of research include the refinement
of the staging system to better differentiate pa-
tients based on known factors, such as metastases
to N1 hilar vs N1 intrapulmonary lymph nodes, for
example. The importance of lymph node micro-
metastases also needs to be determined. In addi-
tion, the role of new imaging and staging modali-
ties such as combined CT/positron emission
tomography or endobronchial ultrasound for eval-
uation of hilar lymph nodes needs to be evaluated
in order to more accurately identify stage II (N1)
NSCLC preoperatively. The development of a
multicenter registry for patients with less common
types of stage II NSCLC may aid in the collection
of data regarding outcomes and treatment strate-
gies. Regarding treatment, despite effective local
control, many patients develop distant recur-
rences. Newer systemic therapies or combinations
of therapies (such as antiangiogenesis agents and
traditional types of chemotherapy) will be neces-
sary in order to reach the goal of improved survival
for these patients.
1. For patients with N1 lymph node metastases in
whom a complete resection can be achieved
with either technique, sleeve lobectomy is rec-
ommended over pneumonectomy. Level of ev-
idence, poor; benefit, moderate; grade of rec-
ommendation, C
therapy with a goal of improving survival is not
recommended. Level of evidence, fair; benefit,
therapy (ie, adjuvant radiation therapy) de-
creases local recurrence rates. For patients who
have undergone complete resection of stage II
NSCLC with N1 lymph node metastases (stage
II [N1] NSCLC), routine administration of
postoperative radiation therapy can be used if
the goal is to decrease local recurrence with the
understanding that survival will not be im-
proved. Level of evidence, fair; benefit, small;
node metastases (stage II [N1] NSCLC), the
evidence indicates that adjuvant radiotherapy
improves local control but does not increase
survival. An overall recommendation cannot be
made regarding routine use of adjuvant radio-
therapy in this setting. Level of evidence, fair;
tion, D
resection of stage II NSCLC, administration of
postoperative chemotherapy should not be con-
sidered standard therapy at this time, and its
use should be limited to patients enrolled in
clinical trials. Level of evidence, good; benefit,
6. In patients who have undergone complete or
incomplete resection of stage II NSCLC,
postoperative combined chemotherapy and
radiotherapy should not be considered stan-
dard therapy at this time, and their use
should be limited to patients enrolled in
clinical trials. Level of evidence, good; ben-
efit, none/negative; grade of recommenda-
tion, D
7. For patients with stage T2N0, T1-2N1, and
T3N0 NSCLC, routine use of preoperative
chemotherapy followed by surgery should not
be considered standard therapy at this time,
and its use should be limited to patients en-
rolled in clinical trials. Level of evidence, poor;
tion, I
8. For patients with lung tumors that abut or are
adjacent to the chest wall based on chest CT
(clinical T3 [chest wall] NSCLC), the presence
or absence of chest wall invasion should not be
assumed based on CT findings alone but should
be confirmed by surgical exploration. Level of
recommendation, C
9. For patients with T3 (chest wall) NSCLC that,
at the time of operation, may extend beyond
the parietal pleura, en bloc resection of T3
(chest wall) NSCLC must be performed un-
less one is confident that extrapleural invasion
does not exist. Long term survival after surgi-
cal treatment for T3 (chest wall) NSCLC is
highly dependent on the completeness of
10. For patients with T3 (chest wall) NSCLC that
does not extend beyond the parietal pleura, an
extrapleural resection may be performed. As
long as a complete resection is achieved,
survival following extrapleural resection is
similar to that achieved following en bloc
11. At the time of surgery for T3 (chest wall)
NSCLC suspected of invading beyond the
parietal pleura, separation of the tumor from
the chest wall followed by resection of that
portion of the chest wall that the tumor was
originally adherent to (ie, discontinuous or
non-en bloc resection) should be avoided.
Limited data suggest that a discontinuous
resection results in very inferior survival com-
pared to en bloc resection. Level of evidence,
poor; benefit, substantial; grade of recommen-
dation, C
resection of T3 (chest wall) NSCLC, routine
postoperative radiotherapy does not provide a
documented survival benefit and should not
be used in these patients. Level of evidence,
mendation, I
13. In patients who have undergone incomplete
resection (or resection with a negative but
close margin) of T3 (chest wall) NSCLC,
postoperative radiotherapy may provide a sur-
vival benefit and could be used. Level of
evidence, poor; benefit, small; grade of rec-
ommendation, C
14. For patients with T3 (mediastinal) NSCLC
who have undergone incomplete resection,
postoperative radiotherapy may be of benefit.
15. Adjuvant radiotherapy for T3 (mediastinal)
NSCLC for completely resected patients
should not be performed. Level of evidence,
mendation, I
16. In all patients with centrally located clinical
T3 NSCLC, histologic assessment of medias-
tinal lymph nodes should be performed prior
to resection. Preoperative identification of N2
lymph node metastases precludes surgical re-
section as initial therapy in this setting. Level
of evidence, poor; benefit, substantial; grade
References
1 AJCC cancer staging manual, 5th ed. In: Greene FL, Page
DL, Flemming ID, et al, eds. Philadelphia, PA: Lippincott-
Raven Press; 1997
3 Inoue K, Sato M Fujimura S, et al. Prognostic assessment of
1310 patients with non-small cell lung cancer who underwent
complete resection from 19801993. J Thorac Cardiovasc
Surg 1998; 116:407411
4 van Rens MT, de la Riviere AB, Elbers HR, et al. Prognostic
assessment of 2,361 patients who underwent pulmonary
resection for non-small cell lung cancer, stage I, II, and IIIA.
Chest 2000; 117:374379
5 Adebonojo S, Bowser A, Moritz D, et al. Impact of revised
stage classification of lung cancer on survival: a military
experience. Chest 1999; 115:15071513
6 Jassem J, Skokowski J, Dziadziusko R, et al. Results of surgical
treatment of non-small cell lung cancer: validation of the new
postoperative pathologic TNM classification. J Thorac Car-
diovasc Surg 2000; 119:11411146
7 Downey RJ, Martini N, Rusch VW, et al. Extent of chest wall
invasion and survival in patients with lung cancer. Ann Thorac
Surg 1999; 68:188193
8 Vrdoljak E, Mise K, Sapunar D, et al. Survival analysis of
untreated patients with non-small cell lung cancer. Chest
1994; 106:17971800
9 Elia S, Griffo S, Gentile M, et al. Surgical treatment of lung
cancer invading the chest wall: a retrospective analysis of 110
patients. Eur J Cardiothorac Surg. 2001; 20:356360
10 Yano T, Yokoyama H, Inoue T, et al. Surgical results and
prognostic factors of pathologic N1 disease in non-small cell
carcinoma of the lung. J Thorac Cardiovasc Surg 1994;
107:13981402
11 van Velzen E, Snijder RJ, de la Riviere AB, et al. Lymph node
type as a prognostic factor for survival in T2N1M0 non-small
cell lung carcinoma. Ann Thorac Surg 1997; 63:14361440
12 Riquet M, Manach D, Le Pimpec-Barthes F, et al. Prognos-
tic significance of surgical-pathologic N1 disease in non-small
cell carcinoma of the lung. Ann Thorac Surg 1999; 67:1572
1576
13 Burt ME, Pomerantz AH, Bains MS, et al. Results of surgical
treatment of stage III lung cancer invading the mediastinum.
Surg Clin North Am 1987; 67:9871000
14 Pitz CCM, de la Riviere AB, Elbers HRJ, et al. Results of
resection of T3 non-small cell lung cancer invading the
mediastinum or main bronchus. Ann Thorac Surg 1996;
62:10161020
15 Weksler B, Bains M, Burt M, et al. Resection of lung cancer
invading the diaphragm. J Thorac Cardiovasc Surg 1997;
114:500501
16 Martini N, Yellin A, Ginsberg RJ, et al. Management of
non-small cell lung cancer with direct mediastinal involve-
ment. Ann Thorac Surg 1994; 58:14471451
17 Gaissert HA, Mathisen DJ, Moncure AC, et al. Survival and
function after sleeve lobectomy for lung cancer. J Thorac
18 Yoshino I, Yokoyama H, Yano T, et al. Comparison of the
surgical results of lobectomy with bronchoplasty and pneu-
monectomy for lung cancer. J Surg Oncol 1997; 64:3235
19 Okado M, Yamagashi H, Satake S, et al. Survival related to
lymph node involvement in lung cancer after sleeve lobec-
tomy compared with pneumonectomy. J Thorac Cardiovasc
Surg 2000; 119:814819
20 Weisenburger T. Effects of postoperative mediastinal radia-
tion on completely resected stage II and stage III epidermoid
cancer of the lung: N Engl J Med 1986; 315:13771381
21 Stephens RJ. The role of post-operative radiotherapy in
non-small cell lung cancer: a multicentre randomised trial in
patients with pathologically staged T1-2 N1-2 M0 disease.
MRC Lung Cancer Working Party. Br J Cancer 1996;
74:632639
22 Dautzenberg B, Arriagada R, Chammard A, et al. A con-
trolled study of postoperative radiotherapy for patients with
completely resected nonsmall cell lung carcinoma. Cancer
1999; 86:265273
23 Feng Q, Wang M, Wang L, et al. A study of postoperative
radiotherapy in patients with nonsmall cell lung cancer: a
randomized trial. Int J Radiol Oncol Biol Phys 2000; 47:925
929
24 Mayer R, Smolle-Jeuttner F-M, Szolar D, et al. Postoperative
radiotherapy in radically resected non-small cell lung cancer.
Chest 1997; 112:954959
25 Stewart LA. Postoperative radiotherapy in non-small cell lung
cancer: systematic review and meta-analysis of individual
patient data from nine randomized controlled trials. PORT
Meta-analysis Trialists Group. Lancet 1998; 352:257263
26 Movsas B. Role of adjuvant therapy in resected stage II/IIIA
non-small-cell lung cancer: part 1. Oncology 2001; 15:1549
1558
27 Lee J, Machtay M, Shrager J, et al. Postoperative radiother-
apy (XRT) for non-small cell lung carcinoma (NSCLC) does
not lead to an excess rate of intercurrent death [abstract].
Proc Am Soc Clin Oncol 1999; 18:472
28 Holmes EC. Surgical adjuvant therapy for stage I and stage
III adenocarcinoma and large cell undifferentiated carci-
noma. Chest 1994; 106(suppl):293S296S
29 Stewart LA. Chemotherapy in non-small cell lung cancer: a
meta-analysis using updated data on individual patients from
52 randomised clinical trials. Non-Small Cell Lung Cancer
Collaborative Group. BMJ 1995; 311899-909
30 Dautzenberg B, Chastang C, Arriagada R, et al. Adjuvant
radiotherapy vs combined sequential chemotherapy followed
by radiotherapy in the treatment of nonsmall cell lung
carcinoma. Cancer 1995; 76:779786
31 Keller SM, Adak S, Wagner H, et al. A randomized trial of
postoperative adjuvant therapy in patients with completely
resected stage II or IIIA non-small cell lung cancer. N Engl
J Med 2000; 343:12171222
32 Pisters K, Ginsberg R, Giroux D, et al. Induction chemother-
apy before surgery for early-stage lung cancer: a novel
approach. J Thorac Cardiovasc Surg 2000; 119:429439
33 Siegenthaler M, Pisters K, Merriman K, et al. Preoperative
chemotherapy for lung cancer does not increase surgical
morbidity. Ann Thorac Surg 2001; 71:11051112
34 Roberts J, Eustis C, Devore R, et al. Induction chemotherapy
increases perioperative complications in patients undergoing
resection for non-small cell lung cancer. Ann Thorac Surg
2001; 72:885888
35 Depierre A, Milleron B, Moro-Sibilot D, et al. Preoperative
chemotherapy followed by surgery compared with primary
surgery in resectable stage I (except T1N0), II, and IIIa
non-small cell lung cancer. J Clin Oncol 2001; 20:247253
36 Ratto GB, Piacenza G, Frola C, et al. Chest wall involvement
by lung cancer: computed tomographic detection and results
of operation. Ann Thorac Surg 1991; 51:182188
37 Glazer HS, Duncan-Meyer J, Aronberg DJ, et al. Pleural and
chest wall invasion in bronchogenic carcinoma: CT evalua-
tion. Radiology 1985; 157:191194
38 Pennes DR, Glazer Gm, Winbish KJ, et al. Chest wall
invasion by lung cancer: limitations of CT evaluation. Am J
Radiol 1985; 144:507511
39 Webb WR, Gatsonis C, Zerhouni EA, et al. CT and MR
imaging in staging non-small cell bronchogenic carcinoma:
report of the Radiologic Diagnostic Oncology Group. Radi-
ology 1991; 178:705713
40 Nakahashi H, Yasumoto K, Ishida T, et al. Results of surgical
treatment of patients with T3 non-small cell lung cancer. Ann
Thorac Surg 1988; 46:178181
41 Pitz CC, Brutel de la Riviere A, Westermann C, et al. Surgical
treatment of 125 patients with non-small cell lung cancer and
chest wall involvement. Thorax 1996; 51:846850
42 Ratto GB, Piacenza G, Musante F, et al. Chest wall involve-
ment by lung cancer: Computed tomographic detection and
results of operation. Ann Thorac Surg 1991; 51:182188
43 Magdeleinat P, Alifano M, Benbrahem C, et al. Surgical
treatment of lung cancer invading the chest wall: results and
prognostic factors. Ann Thorac Surg 2001; 71:10941099
44 Albertucci M, DeMeester T, Rothberg M, et al. Surgery and
the management of peripheral lung tumors adherent to the
parietal pleura. J Thorac Cardiovasc Surg 1992; 103:812
45 Harpole DH Jr, Healey EA, DeCamp M Jr, et al. Chest wall
invasive non-small cell lung cancer: patterns of failure and
implications for a revised staging system. Ann Surg Oncol
1996; 3:261269
46 Patterson GA, Ilves R, Ginsberg RJ, et al. The value of
adjuvant radiotherapy in pulmonary and chest wall resection
for bronchogenic carcinoma. Ann Thorac Surg 1982; 34:692
697
47 Piehler JM, Pairolero PC, Weiland LH, et al. Bronchogenic
carcinoma with chest wall invasion: factors affecting survival
following en bloc resection. Ann Thorac Surg 1982; 34:684
691
48 Dartevelle P, Khalife J, Chapelier A, et al. Tracheal sleeve
pneumonectomy for bronchogenic carcinoma: report of 55
cases. Ann Thorac Surg 1988; 46:6872
49 Vogt-Moykopf I, Fritz T, Meyer G, et al. Bronchoplastic and
angioplastic operation in bronchial carcinoma: long term
results of a retrospective analysis from 1073 to 1983. Intern
Surg 1986; 71:211220
50 Deslauriers J, Mehran RJ, Guimont C, et al. Staging and
management of lung cancer: sleeve resection. World J Surg
1993; 17:712718
51 Okada M, Tsubota N, Yoshimura M, et al. How should
interlobar invasion be classified? Prognosis of resected T3
non-small cell lung cancer. Ann Thorac Surg 1999; 68:2049
2052
52 Staples CA, Muller NL, Miller R, et al. Mediastinal nodes in
bronchogenic carcinoma: a comparison between CT and
mediastinoscopy. Radiology 1988; 167:367372
2003;123;188-201 Chest
Walter J. Scott, John Howington and Benjamin Movsas
Treatment of Stage II Non-small Cell Lung Cancer
& Services
Updated Information
S
References
S#BIBL
free at:
Reprints
2003;123;202-220 Chest
Lary A. Robinson, Henry Wagner, Jr. and John C. Ruckdeschel
Treatment of Stage IIIA Non-small Cell Lung Cancer
ISSN: 0012-3692.
Treatment of Stage IIIA Non-small Cell
Lung Cancer*
Lary A. Robinson, MD, FCCP; Henry Wagner, Jr. MD; and
John C. Ruckdeschel, MD, FCCP
Stage IIIA non-small cell lung cancer represents a relatively heterogeneous group of patients
with metastatic disease to the ipsilateral mediastinal (N2) lymph nodes and also includes T3N1
patients. Presentations of disease range from apparently resectable tumors with occult micro-
scopic nodal metastases to unresectable, bulky multistation nodal disease. Controversy abounds
as to the optimal treatment of the various stage IIIA subsets, which is fueled by a lack of
meaningful, large randomized trials. Multimodality therapy of some type appears to be prefer-
able in stage IIIA patients. (CHEST 2003; 123:202S220S)
Key words: adjuvant chemotherapy; adjuvant radiotherapy; chemotherapy; guidelines; lung carcinoma; neoadjuvant
therapy; non-small cell lung cancer; pulmonary surgical procedures; radiation therapy
Abbreviations: CALGB Cancer and Leukemia Group B; CAP cyclophosphamide-doxorubicin-cisplatin;
CHART continuous hyperfractionated accelerated radiation therapy; ECOG Eastern Cooperative Oncology
Group; EORTC European Organization for Research and Treatment of Cancer; HART hyperfractionated
accelerated radiation therapy; LCSG Lung Cancer Study Group; MRC Medical Research Council; NSCLC non-
small cell lung cancer; NSCLCCG Non-Small Cell Lung Cancer Collaborative Group; PET positron emission
tomography; PS performance status; SWOG Southwest Oncology Group; UFT uracil-tegafur
T
he evidence-based guidelines that follow are
written primarily to provide a succinct synthesis
of the medical literature and provide specific treat-
ment guidelines that can serve as a useful tool for the
clinician who deals directly with locally advanced
non-small cell lung cancer (NSCLC). Exhaustive
detail about published trials will be avoided in order
to make this a more readable and usable guide. In
order to develop the following guidelines for stage
IIIA disease, the authors reviewed 15 other pub-
lished guidelines, 9 meta-analyses, 12 systematic
reviews, and 80 primary articles on this topic, focus-
ing on the most well-designed, peer-reviewed re-
ports. Selected key references are included in the
bibliography.
Based on the collected series of 5,230 patients
with NSCLC seen in the period from 1975 to 1988 at
the M.D. Anderson Cancer Center reported by
Clifton Mountain in the 1997 revision of lung cancer
staging criteria,
1
30% of all patients have locally
advanced disease at initial presentation. Of those,
one third (10% of the total) have stage IIIA with
ipsilateral N2 lymph node metastases, which in the
United States would then encompass approximately
17,000 new patients yearly. This group forms per-
haps the most therapeutically challenging and con-
troversial subset of patients with lung cancer, with a
published 5-year survival of only 23%.
1
This border-zone subset of stage IIIA patients,
which lies between the generally resectable stage I
and II tumors and unresectable stage IIIB patients,
has been the subject of a wide variety of clinical trials
incorporating various combinations of chemother-
apy, radiotherapy, and surgery. Unfortunately, most
published studies have significant limitations since
they are not randomized, lack rigorous pretreatment
staging, or involve a significant lack of homogeneity
in the study population, making interpretation of the
results difficult. There are a few more rigorous
randomized trials, which will be discussed subse-
quently, that suggest a combined modality approach
may be beneficial in stage IIIA disease. The ap-
proach showing the greatest promise in selected
patients employs initial treatment (induction or neo-
adjuvant therapy) with chemotherapy or chemora-
diotherapy followed by surgery. Nevertheless, more
widespread use of induction therapy followed by
surgery for lung cancer has been used for only 7
years, and as a result there is little reliable data with
larger patient groups. This lack of meaningful, larger,
*From the Thoracic Oncology Program, Department of Interdis-
ciplinary Oncology, H. Lee Moffitt Cancer Center and Research
Institute, University of South Florida, Tampa, FL.
Correspondence to: Lary A. Robinson, MD, FCCP, Division of
Cardiovascular and Thoracic Surgery, Thoracic Oncology Pro-
gram, H. Lee Moffitt Cancer Center and Research Institute,
12902 Magnolia Dr, Tampa, FL 33612-9497; e-mail:
robinson@moffitt.usf.edu
randomized data underscores the importance of
enrolling patients in clinical trials whenever possible.
Since staging and treatment are so very interde-
pendent, intraoperative staging with systematic me-
diastinal node sampling or dissection is critically
important. Unless histologic conformation of medi-
astinal node status is obtained at the time of surgery,
postoperative pathologic staging will be inaccurate,
as will further treatment recommendations and the
discussion of prognosis. Therefore, the standard of
care in modern thoracic surgery dictates that medi-
astinal node sampling or dissection must be per-
formed at the time of every lung resection for lung
cancer.
Under the 1997 revised lung cancer staging sys-
tem,
1
stage IIIA encompasses all tumors with ipsi-
lateral mediastinal lymph node metastases (T13,
N2). Also included in this stage are tumors with
resectable chest wall involvement and hilar node
metastases (T3N1), added primarily because of sim-
ilar survival rates. However, the treatment recom-
mendations and applicable clinical trials for T3N1
are the same as for stage II. Therefore, for the
purposes of these current guidelines, T3N1 tumors
are discussed in the preceding chapter on stage II
tumors. The present chapter will deal only with N2
disease.
Nevertheless, the patients with stage IIIA (N2)
tumors present substantial heterogeneity in clinical
presentation, treatment, and prognosis. Therefore,
for the purposes of generating rational treatment
guidelines, we have chosen to classify N2 tumors into
four subsets (Table 1), which have been published
previously.
2
The subsequent discussion of the liter-
ature and treatment guidelines will be broken down
into these subsets.
Previously Published Guidelines
In drafting these evidence-based guidelines for
stage IIIA NSCLC, the authors not only reviewed
the literature of clinical trials and reviews, but also 15
sets of recently published, major lung cancer guide-
lines were considered.
317
In general, there were few
real differences in the actual recommendations be-
tween guidelines in this area of lung cancer. Some of
the previously published guidelines were consensus
based entirely, while others (such as the current
guidelines) followed the evidence-based format.
Among the latter group of guidelines, there were
only some minor differences, primarily in determin-
ing the strength of the evidence supporting the
specific recommendations.
Treatment of Specific Patient Groups
Incidental N2 Disease (Stage IIIA
12
)
Despite careful preoperative staging including CT
scan, positron emission tomography (PET), and me-
diastinoscopy, some patients will be found to have
metastases to mediastinal N2 lymph nodes at thora-
cotomy. In some, metastatic nodal disease will be
found as a surprise a number of days postoperatively
on the final pathologic examination of the surgical
specimen (stage IIIA
1
). In others, metastases will be
found intraoperatively as an unexpected finding at
thoracotomy with a frozen-section pathologic exam-
ination of mediastinal nodes (stage IIIA
2
). Unex-
pected nodal metastases in this setting are not that
unusual. In the pre-PET scan era, one surgical series
of 102 patients from the Brompton Hospital in
London with no clinical evidence of mediastinal
adenopathy at thoracotomy found 24% of patients
had pathologically positive nodes.
18,19
Surgery: Despite negative preoperative staging
studies including mediastinoscopy, as many as one
fourth of patients will be found at surgery to have
occult N2 metastatic disease.
18,19
If only one nodal
station is unexpectedly found to be involved with
metastatic lung cancer at open thoracotomy, and all
of the involved nodes are technically resectable and
the primary tumor is also technically resectable, then
the surgeon should proceed at that time with the
planned lung resection along with a mediastinal
lymphadenectomy. If a complete resection is not
possible or there is multistation or bulky, unresect-
able extracapsular nodal disease, then the planned
lung resection should be aborted. Although incom-
plete resection rarely results in long-term survival,
collected results of surgery alone in stage IIIA (N2
disease) provides a 14 to 30% 5-year survival, with
the best survival seen in cases with minimal N2
disease and complete resection.
2027
At least 27 to 36% of patients with metastatic
disease to the mediastinal N2 nodes will not have
Table 1Subsets of Stage IIIA (N2)*
Subset Description
IIIA
1
Incidental nodal metastases found on final
pathologic examination of the resection
specimen
IIIA
2
Nodal (single station) metastases recognized
intraoperatively
IIIA
3
Nodal metastases (single or multiple station)
recognized by prethoracotomy staging
(mediastinoscopy, other nodal biopsy, or
PET scan)
IIIA
4
Bulky or fixed multistation N2 disease
*Adapted from Ruckdeschel.
2
involvement of the hilar or lobar lymph nodes.
28,29
If
resection of clinically negative mediastinal lymph
nodes is not performed at the time of lung resection,
it is possible that occult, subclinical metastatic dis-
ease to the N2 nodes will be missed, which will
provide inaccurate pathologic staging and may alter
the clinical course.
The optimal intraoperative approach to deal with
the mediastinal lymph nodes remains unsettled.
There is general agreement that systematic invasive
harvesting of nodes from all possible lymph node
stations is essential for accurate staging, but contro-
versy arises as to whether complete mediastinal
lymph node dissection is of therapeutic benefit in
improving long-term survival rates. Theoretically,
mediastinal lymph node dissection will harvest more
nodes and thereby provide more accurate staging.
Few published randomized studies have addressed
the sampling vs dissection question. In a prospective
randomized trial, Izbicki and associates
30
found no
survival benefit of an en bloc mediastinal lymph node
dissection compared to systematic lymph node sam-
pling in NSCLC. However, data from the recent
North American Intergroup trial comparing adjuvant
postoperative radiotherapy with radiochemotherapy
in N1 and N2 node-positive patients shows a mild
significant benefit for mediastinal dissection, al-
though this analysis was retrospective and the choice
of the approach to nodes in the mediastinum was left
to the surgeon.
31
In a companion analysis of lymph
node harvesting techniques and results from this
North American Intergroup trial, mediastinal lymph
node dissection resulted in a significantly longer
median survival than systematic lymph node sam-
pling, but the survival advantage was limited to
patients with right lung tumors (66.4 months vs 24.5
months, p 0.001).
32
Realistically, the distinction
between what constitutes a mediastinal lymphade-
nectomy as opposed to systematic mediastinal lymph
node sampling is technically somewhat blurred and
is quite surgeon dependent.
However, if prior to thoracotomy, metastatic dis-
ease is found in the N2 nodes at mediastinoscopy, for
example, then further surgery at that time should be
avoided. If appropriate, induction therapy first is
more advantageous, followed later in selected pa-
tients by definitive surgical resection of the primary
lung cancer along with as complete a mediastinal
lymphadenectomy as possible. This topic will be
discussed in a subsequent section.
Recommendations
1. In patients with an occult single-station medi-
astinal node metastasis that is recognized at
thoracotomy and when a complete resection
of the nodes and primary tumor is technically
possible, then proceed with the planned lung
resection and a mediastinal lymphadenec-
tomy. Level of evidence: poor; benefit, small;
grade of recommendation: C
2. In every patient undergoing a lung resection
for lung cancer, systematic mediastinal lymph
node sampling or complete mediastinal lymph
node dissection must be performed. Level of
evidence: good; benefit, substantial; grade of
recommendation: A
Adjuvant Radiotherapy: While it is recognized
that the finding of regional metastatic N2 disease at
surgery is a poor prognostic feature, there is little
consensus as to the appropriate postthoracotomy
management of these patients. Despite the great
frequency of lung cancer, there have been relatively
few patients entered into prospective trials evaluat-
ing the role of adjuvant postoperative radiation
therapy, chemotherapy, or both.
The role of postoperative radiation therapy in
patients with NSCLC has been debated for many
years. The ability of postoperative radiation therapy
in moderate doses, 4,500 to 5,500 cGy, to eradicate
microscopic residual disease and reduce rates of
local recurrence was established in several early
single-institution trials.
3335
What has remained con-
troversial is whether or not the reduction in locore-
gional recurrence also leads to an improvement in
overall survival. While the nonrandomized, single
institution trials suggested that this was the case, data
from the prospective trials have been less supportive.
Two separate issues are likely involved: (1) How
large is the group of patients who have residual
disease locally in the chest without occult distant
metastatic disease, the subgroup for whom adjuvant
mediastinal radiation therapy might be curative?
(2) What is the morbidity and mortality of adjuvant
mediastinal radiotherapy with modern treatment
planning techniques?
The Lung Cancer Study Group (LCSG) con-
ducted a phase III trial in which patients with
resected squamous cell carcinoma of the lung were
randomized between observation and mediastinal
irradiation to 50 Gy in 5 weeks.
36
Entry into the
study was restricted to patients with squamous cell
carcinoma because of the greater tendency of this
tumor to fail locally rather than distantly, compared
with adenocarcinoma and large cell carcinoma. The
majority of patients had N1 disease, but smaller
proportions had N2 or T3N0 disease. The results of
the trial were striking. Local failure as a first site of
relapse was seen in 20% of patients on the observa-
tion arm but was seen in only 1% of those random-
ized to adjuvant nodal irradiation. The LCSG and
other trials of adjuvant postoperative radiation have
been criticized for their small sample size, their
mixture of N1 and N2 patients, and for the reliance
on data on the site of first failure. It should be
remembered that these deficiencies did not prevent
the demonstration of a striking effect of radiotherapy
on local control. What was lacking was the efficacy of
good local control to result in long-term freedom
from disease.
Several other randomized trials have addressed
the same issue in patients with resected NSCLC of
all histologies and have consistently failed to demon-
strate a significant survival advantage (Table 2). In
some trials there have been poorer survivals for
irradiated patients, most likely due to increased
cardiopulmonary toxicity. In both the LCSG and
Medical Research Council (MRC) trials there was a
trend to improved survival for the irradiated N2 but
not N1 patients, but these survival differences did
not reach statistical significance.
The meta-analysis by the Postoperative Radiation
Therapy Meta-analysis Trialist Group of 2,128 pa-
tients treated in nine randomized trials (six previ-
ously published series and three unpublished series)
of postoperative radiation therapy concluded that
this treatment was associated with a highly signifi-
cant increase in the risk of death.
44
Overall, the risk
ratio was 1.21 (p 0.001). The authors concluded
that postoperative radiotherapy as used in these
studies was detrimental and should not be used. It is
important to recognize that there are several signif-
icant differences between the treatment adminis-
tered in a number of the trials included in this
meta-analysis and current practice patterns in the
United States. First, a substantial portion of the
patients included in this study, 562 of 2,128 patients
(26.4%), had stage I disease without demonstrated
nodal metastases. There has never been a strong case
favoring the postoperative irradiation of these stage I
patients, and there is little suggestion from patterns
of their failure after surgery that such treatment
would be beneficial. Thus, one fourth of the patients
in this analysis stood to gain little from treatment.
Second, the details of treatment, including preoper-
ative staging, surgical technique, and radiation dose
and dose delivery, differed substantially from current
practice. Several of the trials required or allowed
very large daily fraction sizes in excess of 2.0 Gy, with
the MRC trial using 2.6 Gy/d and the Slovenian trial
using 3.0 Gy/d. Such larger fraction sizes would be
expected to have an increase in acute and late
complications compared to slower fractionation.
Seven of the nine trials also allowed the use of
60
Co
treatment beams, with their poorer depth-dose char-
acteristics than higher-energy linear accelerator
beams, and only one study included CT scan-based
treatment planning. Therefore, compared with pre-
sent standards of treatment, the likelihood is great
that postoperative radiation therapy would lead to
excess deaths from cardiac and pulmonary damage.
In such a meta-analysis that included patients with
little chance of benefit of treatment, this would likely
result in an overall survival detriment. It is notable
that in this meta-analysis, the increased risk of death
was most marked in those patients with stage I
disease and was not significant for patients with N2
disease. This is consistent with, although it does not
prove, a potential benefit for properly delivered
radiotherapy for resected N2 patients.
At present, postoperative radiation therapy can-
not be recommended on the basis of any proof of
improved survival, but it should be considered in
selected patients to reduce the risk of local recur-
rence, particularly when there is involvement of
multiple nodal stations, extracapsular tumor
spread, or close or microscopically positive resec-
tion margins. While adjuvant mediastinal radio-
therapy has often been viewed as routine, it can be
associated with significant cardiac and pulmonary
toxicity, and care in treatment planning and
delivery is essential.
Table 2Randomized Controlled Trials of Surgery Plus Adjuvant Radiotherapy vs Surgery Alone*
Source Year
Patients,
No.
Radiotherapy
Dose, Gy Stage Survival
Local Recurrence
Surgery Plus
Radiotherapy/Surgery, %
Paterson and Russel
37
1962 202 45 Any ND
Bagma
38
1971 73 46 Any ND
Van Houte et al
39
1980 224 60 I, II ND 4.8/20.7 (p 0.002)
Weisenberger (LCSG 773)
40
1985 210 50 II, IIIA ND 1/19 (p 0.02)
Stephens et al (MRC)
41
1996 308 40 II, IIIA ND 18/29 (p 0.003)
Debevec et al
42
1996 74 30 IIIA ND
Dautzenberg et al
43
1999 728 60 II, IIIA Worse for radiotherapy
group (p 0.002)
*ND no significant difference.
Adjuvant Chemotherapy: Since the predominant
pattern of failure is systemic recurrence of metastatic
disease in patients with fully resected stage IIIA lung
cancer, numerous trials of adjuvant postoperative
chemotherapy have been carried out over the last 3
decades. These trials have been hampered by a
number of problems, including inconsistent staging
especially in the earlier trials, lack of effective che-
motherapeutic agents until recently, and the poor
tolerance of postthoracotomy patients to chemother-
apy due to GI toxicity in an era lacking strong
antiemetic agents.
In the 1970s and 1980s, a number of adjuvant
chemotherapy trials used drug combinations that
predated cisplatin-containing regimens. Most of
these trials used alkalating agents and provided no
survival advantage to patients; in fact, in most there
was a detrimental effect resulting in a relative 15%
increase in death in patients receiving adjuvant
chemotherapy.
45
In the 1990s, a number of controlled, randomized
trials were published using a variety of cisplatin-based
chemotherapy regimens, commonly using cyclophos-
phamide-doxorubicin-cisplatin (CAP). Most of these
trials (Table 3) of adjuvant chemotherapy after lung
resection had a mixture of stages. Common to most
trials was significant GI toxicity (studies predated the
availability of serotonin-receptor antagonist antiemet-
ics), and few patients received the full planned course
of chemotherapy. Almost all trials showed no advantage
in disease-free survival or overall survival with postop-
erative adjuvant chemotherapy. Niiranen and associ-
ates
46
did find a significant increase in survival in
resected T13N0 patients with adjuvant CAP chemo-
therapy. However, the surgery-only control arm had a
high proportion of pneumonectomy cases, and when
the pneumonectomy cases were excluded from analysis
the survival advantage disappeared.
A meta-analysis by the Non-Small Cell Lung
Cancer Collaborative Group (NSCLCCG) in 1995
analyzed the results of five non-cisplatinbased
adjuvant chemotherapy regimens and found no
survival benefit.
45
The NSCLCCG also analyzed
eight cisplatin-based adjuvant chemotherapy trials
and found a 13% decrease in the relative risk of
death with chemotherapy and an absolute survival
benefit of 3% at 3 years and 5% at 5 years, but all of
the differences were not statistically significant. A
later meta-analysis by Le Chevalier
52
in 1998, of all
randomized, controlled adjuvant chemotherapy tri-
als, also suggested a small 5% survival benefit with
cisplatin-based regimens.
A persistent problem with postoperative chemother-
apy has been administering the planned doses and
cycles of chemotherapy. However, with the elimination
of drugs such as doxorubicin and the introduction of
better supportive care drugs such as improved anti-
emetics and cytokine support of hematologic toxicity,
there would theoretically be improved chemotherapy
dose compliance. Unfortunately, the experience of
ongoing trials shows that the problem has not resolved
and only approximately 65% of the planned dose of
chemotherapy is actually received. The recent positive
Japanese experience with low-dose, minimally toxic,
prolonged adjuvant therapy with uracil-tegafur (UFT)
suggests that the standard short-term, dose-intense
adjuvant therapy may not be the best or only approach
to consider.
51
Adjuvant Combination Chemoradiotherapy: With
the lack of any apparent survival advantage in adju-
vant chemotherapy or radiotherapy in resected N2
lung cancer, attention turned to exploration of the
potential benefit of combination chemoradiotherapy
postoperatively. Adjuvant radiotherapy appears to
decrease local recurrence, but failure with distant
metastases is a predominant pattern that theoreti-
cally should be amenable to the addition of adjuvant
systemic chemotherapy.
To date, there have been five published random-
ized controlled trials involving N2 disease patients
(Table 4) with adjuvant chemotherapy and radiother-
apy, beginning with LCSG 791.
53
This trial involved
patients who had incomplete resections (positive
Table 3Randomized Controlled Trials of Surgery Plus Adjuvant Cisplatin-Based Chemotherapy vs Surgery Alone*
Source Year
Patients,
No. Adjuvant Chemotherapy Stage
Disease-Free
Survival
Long-term Survival (5 yr)
Surgery-Chemotherapy/
Surgery, %
Niiranen et al
46
1992 110 CAP IIII (I, 90%) ND 67/56 (p 0.05 for stage I)
Ohta et al
47
1993 181 Cisplatin/vindesine III ND 35/41 (p 0.86)
Feld et al (LCSG)
48
1993 269 CAP III (I, 84%) ND 53/57 (p 0.92)
Figlin and Piantodosi
49
1994 188 CAP IIIII ND ND
SGACLC
50
1995 333 Cisplatin/doxorubicin/UFT IIII (I, 61%) ND 68.7/58.1 (p 0.35)
Wada et al
51
1996 323 Cisplatin/vindesine/UFT vs UFT
alone vs surgery alone
IIII 60.6/64.1/49 (p 0.1)
*See Table 2 and abbreviation list for expansion of abbreviations.
margins or involvement of the most proximal lymph
node in the mediastinum) and compared postopera-
tive split course radiotherapy with the same radio-
therapy plus CAP chemotherapy. There was an
increase in the recurrence-free survival favoring the
chemotherapy arm (p 0.004), but overall survival
was not increased.
Later trials failed to demonstrate any improve-
ment in disease-free survival or overall survival with
the addition of adjuvant chemotherapy to radiother-
apy. The most recently published report is the 488
patient North American Intergroup trial (E3590),
31
which also failed to demonstrate any increase in
median survival or disease-free survival. In a com-
panion laboratory subset analysis, Schiller and
associates
57
found a nonsignificant trend toward
improved median survival in adjuvant chemoradio-
therapy patients who had normal (wild-type) K-ras
expression compared to mutant K-ras patients (me-
dian survival, 42 months vs 25 months; p 0.09).
Nevertheless, evidence is yet to be established sub-
stantiating the benefit of the routine addition of
adjuvant chemotherapy to postoperative radiother-
apy in stage IIIA lung cancer.
Recommendations
3. Adjuvant Radiotherapy: In the patient with
fully resected stage IIIA lung cancer, there is
no definite improvement in survival with ad-
juvant postoperative radiotherapy, but it sig-
nificantly reduces local recurrence and should
be considered in selected patients. Level of
evidence: fair; benefit: small; grade of recom-
mendation: C
4. Adjuvant Chemotherapy: In the patient with
fully resected stage IIIA lung cancer, adjuvant
chemotherapy administered alone might offer
a very modest survival advantage, but this
modality should not be routinely utilized out-
side of a clinical trial. Level of evidence: poor;
benefit: small; grade of recommendation: I
5. Adjuvant Chemoradiotherapy: In the patient
with fully resected stage IIIA lung cancer,
based on randomized clinical trials to date,
there is no survival benefit appreciated by
adding postoperative adjuvant chemotherapy
to adjuvant radiotherapy. Therefore, the rou-
tine use of combined postoperative chemo-
therapy and radiotherapy is not recommended
and should not be employed outside of a
clinical trial. Level of evidence: fair; benefit:
none; grade of recommendation: D
Potentially Resectable N2 Disease (Stage IIIA
3
)
Traditionally, the finding of any metastasis what-
soever to the mediastinal N2 nodes deemed that
patient to have an unresectable lung cancer. With
the development of chemotherapeutic agents with
significant activity against lung cancer, beginning
with cisplatin in the early 1980s, and with the
development of modern radiotherapy techniques,
studies have appeared suggesting that combining
chemotherapy and/or radiotherapy followed by sur-
gery in selected stage IIIA patients may offer ther-
apeutic benefit. The poor survival rates with surgery
alone in N2 disease, even with adjuvant postopera-
tive chemotherapy or radiotherapy, has led to efforts
at giving initial nonsurgical (radiotherapy and/or
chemotherapy) therapy first, often to convert the
unresectable tumor to resectable and, as well, to
improve long-term survival. After a number of initial
phase II trials with various drugs and radiotherapy
doses administered prior to surgery in the neoadju-
vant or induction setting, there were enough positive
Table 4Randomized Controlled Trials of Surgery Plus Adjuvant Chemoradiotherapy vs Surgery Plus
Adjuvant Radiotherapy*
Source Year
Patients,
No. Stage
Chemotherapy
Radiotherapy Regimens
Disease-Free
Survival
Long-term Survival Surgery-
Radiotherapy vs Surgery-
Radiotherapy/Chemotherapy, %
Lad et al (LCSG)
53
1988 164 IIIII CAP 40 Gy (split course) Chemo. favored
(p 0.004)
54/68 (p 0.1) 1 yr
Sawamura et al
54
1988 52 IIIII Tegafur-cisplatin 50 Gy ND ND
Pisters et al
55
1994 72 III Vindesine-cisplatin 40 Gy ND 44/31 (p 0.42) 2 yr
Dautzenberg et al
56
1995 267 IIII Doxorubicin-cyclophosphamide-
lomustine-cisplatin-vincristine
60 Gy
ND 12/13 (p 0.68) 10 yr
Keller et al (North
American
Intergroup
E3590)
31
2000 488 IIIIIA Cisplatin-etoposide 50.4 Gy ND 39 mo/38 mo (p 0.56) median
*See Table 2 for expansion of abbreviation.
results to persuade even the most pessimistic that
this approach may have value.
Induction (Neoadjuvant) Therapy: The majority of
stage IIIA patients have enlarged ( 1.0-cm short-
axis diameter) N2 nodes (our stage IIIA
3
) on chest
CT. Mediastinoscopy should generally be performed
in this setting to document that these nodes actually
contain metastatic tumor, since approximately 40%
of moderately enlarged nodes may be benign, espe-
cially if there is an associated recent pneumonitis.
Adverse prognostic factors associated with positive
mediastinal nodes include extracapsular spread of
tumor, multiple levels of involved lymph nodes, and
bulky enlarged nodes.
58
Of special note is the loca-
tion of the N2 nodes, in that involvement of the
higher, superior mediastinal nodes (nodes found
positive that are generally available for biopsy at
mediastinoscopy) portends a worse prognosis than
patients with a negative mediastinoscopy result yet
who are found to have positive nodes at thoracoto-
my.
59
However, other studies contradict this finding.
Naruke and colleagues
60
found that metastatic dis-
ease to the subcarinal lymph nodes adversely af-
fected prognosis compared to other lymph node
locations. The LCSG retrospectively analyzed 163
patients with stage III disease from their postoper-
ative treatment protocols and found that the survival
rate was worse for patients with subcarinal lymph
node metastases plus nodes from other sites, than for
subgroups of patients with mediastinal nodal metas-
tases in other locations.
61
Miller and associates
62
analyzed their lung-term survival rates in 167 pa-
tients who at thoracotomy were found to have N2
nodal metastases, not suspected preoperatively. The
5-year survival was worse when there was metastatic
disease in the subcarinal or lower lymph nodes
(stations 8 or 9). Also, survival was worse when
multiple lymph node stations were involved.
62
Fi-
nally, Okada and colleagues
63
reviewed their long-
term survival rates in 141 patients with N2 nodal
metastatic disease and found that the survival rate
depended on the location of the lung cancer (upper
or lower lobes) in relationship to the location of the
nodal metastases. For example, upper-lobe lung
cancer patients with metastases limited to upper
mediastinal nodal stations did better than when the
lower mediastinum (subcarinal nodes) was involved
in the upper-lobe cancers.
63
The only conclusion that
can be realistically drawn from the somewhat con-
flicting information from these and other studies is
that multistation nodal disease has a somewhat worse
prognosis that single-station disease, but the location
of metastatic disease to a single nodal station prob-
ably has no significant effect.
There are theoretical advantages of the neoadju-
vant approach including decreasing tumor size to
allow more ready resection, decreased micrometas-
tases, decreased surgical seeding, and increased pa-
tient acceptance. However, neoadjuvant therapy also
has the potential disadvantages of a delay in primary
tumor control and increased surgical morbidity and
mortality. The literature is replete with numerous
phase II nonrandomized clinical trials of neoadjuvant
chemotherapy with or without radiotherapy followed
by lung resection in highly selected patients.
As summarized by Rusch,
58
results of these phase
II trials suggest that the neoadjuvant approach may
offer improved resectability with acceptable surgical
morbidity and mortality and is associated with an
improved survival benefit over single-modality ther-
apy. Martini and colleagues
64
gave induction chemo-
therapy with mitomycin, vindesine or vinblastine,
and cisplatin to stage IIIA patients with bulky medi-
astinal nodal metastases or multilevel nodal disease
and found a 65% complete resection rate, a 15%
treatment-related mortality, and a 28% 3-year sur-
vival, which was far better than historical control
subjects (8% 3-year survival). Other phase II induc-
tion chemotherapy trials have generally confirmed
this trial.
Eight small, randomized phase III trials of neoad-
juvant therapy in stage IIIA patients have been
published over the last decade (Table 5). Many
concerns have been raised about these phase III as
well as the phase II neoadjuvant trials: (1) there was
no consistent surgical (pathologic) staging of the
mediastinal lymph nodes; (2) variable numbers of
patients with a much better prognosis (T3N0 and
T3N1) were included in these trials, which might
have influenced the outcome of the trials; (3) some
patients with a poorer prognosis (stage IIIB) were
mixed in with patients with a better prognosis,
thereby worsening results; and (4) many trials have
small numbers of patients due to poor accrual with
resultant low statistical power. With these caveats in
mind, the results of the largest trials from Barcelona,
Spain
67,68
and the M.D. Anderson Cancer Cen-
ter
69,70
provide promising results. Both of these trials
were closed early to further accrual after the interim
analyses demonstrated significant survival advan-
tages for the induction chemotherapy arm.
Rosell and associates,
67,68
in Barcelona, Spain,
randomized 60 stage IIIA patients to either surgery
alone or three cycles of induction chemotherapy with
mitomycin, ifosfamide, and cisplatin followed by
surgery. All patients received postoperative radiation
therapy. Lymph nodes were pathologically staged
initially by mediastinoscopy in only 73% of patients.
Twenty-seven percent of patients had more favor-
able T3N0 or T3N1 tumors. A significant survival
advantage was seen in the induction chemotherapy-
surgery arm with a 22-month median survival, com-
pared to 10 months in the surgery-only arm
(p 0.005). Two-year and 5-year survivals were 29%
and 17% for the chemotherapy-surgery arm vs 5%
and 0% in the surgery-only arms, respectively. Al-
though encouraging, this study has been criticized
not only for the small number of patients but also for
the significant imbalance of patients with poor prog-
nosis K-ras mutations and aneuploid tumors in the
surgery-only arm, which may have adversely biased
the outcome in this arm. Also, there were no 5-year
survivors in the surgery-only arm, which is surprising
since 27% of the patients had favorable T3N0 or
T3N1 tumors.
Roth and colleagues,
69,70
at the M.D. Anderson
Cancer Center, also randomized 60 stage IIIA pa-
tients to surgical resection alone or three cycles of
induction chemotherapy with cyclophosphamide,
etoposide, and cisplatin followed by surgery and then
three cycles postoperatively. Postoperative radiation
therapy was administered only to incompletely re-
sected patients. Only 83% of patients were invasively
staged prior to treatment. Also, 26% of patients had
more favorable T3N0 or T3N1 tumors. The median
survivals were 21 months for the chemotherapy-
surgery arm vs 14 months for the surgery-only arm
(p 0.048). The 3-year and 5-year survival rates
likewise favored the chemotherapy-surgery arm at
46% and 36% compared to 19% and 15% in the
surgery-only arms, respectively. This study has also
been criticized for its small patient numbers as well
as a significant postoperative stage imbalance, with
40% stage IIIB and IV patients in the surgery-only
arm compared with 11% in the chemotherapy-
surgery arm. However, this imbalance potentially
could have been the result of downstaging in the
chemotherapy-surgery arm due to the induction
therapy. Although encouraging, the results of these
small randomized trials are unclear but strongly
argue for the completion of the current larger trials
in progress.
The most recently reported neoadjuvant trial is
from Depierre and associates
74
with the French
Thoracic Cooperative Group. From 1991 through
1997, they randomized 373 patients with stages IB,
II, and IIIA together into two treatment arms:
Table 5Randomized Controlled Trials of Preoperative Neoadjuvant (Induction) Therapy and Surgery vs Surgery
Alone in Stage IIIA (NSCLS)*
Source Year
Patients,
No. Induction Arms
Median Survival,
mo
Survival,
%
Pass et al (NCI)
65
1992 27 1. Cisplatin, etoposide 29 42 (3 yr)
2. None 16 12 (3 yr)
(p 0.095)
Fleck et al
66
1993 96 1. 5-FU, cisplatin/radiotherapy
30 Gy
NR NR (arm 1 reported better resection
rate and freedom from progression)
2. Cisplatin, vinblastine,
mitomycin
Rosell et al
67,68
1994 60 1. Ifosfamide, mitomycin,
cisplatin
22 29 (2 yr), 17 (5 yr)
2. None 10 5 (2 yr), 0 (5 yr)
(p 0.005)
Roth et al
69,70
1994 60 1. Cisplatin, etoposide,
cyclophosphamide
21 46 (3 yr), 36 (5 yr)
2. None 14 19 (3 yr), 15 (5 yr)
(p 0.048)
Wagner et al (LCSG)
71
1994 57 1. Mitomycin, vinblastine,
cisplatin
12 27 at 4 yr for both arms
2. Radiotherapy, 44 Gy 12
Elias et al (CALGB)
72
1997 57 1. Radiotherapy, 40 Gy 23 NR
2. Cisplatin, etoposide 19 NR
(p 0.64)
Ichinose et al (JCOG)
73
2000 62 1. Cisplatin vindesine 18 8 (5 yr)
2. None 16 25 (5 yr)
Depierre et al
74
2002 167 with IIIA 1. Mitomycin, cisplatin,
ifosfamide
NR 28 (5 yr)
2. None 21 (5 yr)
(estimated) ND
*Adapted from Garland et al.
75
NR not reported; 5-FU 5-fluorouracil; ND not done.
Study closed early due to poor accrual.
Study closed early due to large, significant differences between treatment arms.
Study combined stage IB, II, and IIIA.
(1) primary surgery, and (2) two cycles of preopera-
tive chemotherapy with mitomycin, ifosfamide, and
cisplatin, followed by surgical resection and then two
cycles postoperatively. Patients in both treatment
arms found postoperatively to have pathologic T3 or
N2 disease received postoperative radiotherapy. The
prerandomization stage was determined clinically
based on chest CT, and any lymph node 1 cm in
short-axis diameter was considered positive for pur-
poses of staging. The overall response to preopera-
tive chemotherapy was 64%. The median survival
overall with the combined stages was 37 months in
the chemotherapy-surgery arm and 26 months in the
surgery-only arm (p 0.15). In a subset analysis,
patients with N0 and N1 disease appreciated signif-
icant improvements in disease-free and overall sur-
vival in the chemotherapy-surgery arm compared to
surgery only. For the subset of 167 patients with
stage IIIA disease (92 patients in the chemotherapy-
surgery arm, and 75 patients in the surgery-only
arm), there was no significant difference in survival
in the two treatment arms, with an estimated 5-year
survival of approximately 29% in the chemotherapy-
surgery group compared to 20% in the surgery-only
group (survivals estimated from the published sur-
vival curves). Unfortunately, the subset analysis in
the published report was not complete. This study
may be criticized in a number of aspects, most
notably for the lack of preoperative histologic verifi-
cation of nodal stage prior to randomization, as well
as the combination of diverse stages into the same
study arm, thereby making the subset analysis of
stages a retrospective exercise with potential imbal-
ance of the patient groups. Despite the obvious
deficiencies when evaluating these results for stage
IIIA patients, this study still fails to demonstrate any
significant survival benefit for induction chemother-
apy followed by surgery compared to surgery alone
in locally advanced NSCLC.
Until the larger, multi-institutional randomized
phase III trials (such as the current North American
Intergroup 0139 trial and European Organization for
Research and Treatment of Cancer [EORTC]-
08941) of neoadjuvant therapy designed specifically
for stage IIIA lung cancer are completed, this ap-
proach with preoperative chemotherapy and possibly
radiotherapy for N2 disease, which appears to be
feasible, should not presently be considered the
standard of care in the community. Ideally this
approach should only be administered in the setting
of an investigational protocol. For this reason, we
recommend following the guidelines in this chapter
for the different subsets of stage IIIA patients.
Finally, the older patient or patient with poor per-
formance status should still be approached with
caution when considering these aggressive multimo-
dality protocols.
Recommendations
6. Patients with stage IIIA (N2) lung cancer
identified preoperatively have a relatively poor
prognosis when treated with surgery as a
single modality. Several small trials of induc-
tion chemotherapy have yielded conflicting
results about its effect on survival. The relative
roles of surgery and radiation therapy as the
local treatment modality are also not clearly
defined. Definitive treatment recommenda-
tions are difficult to make in this setting.
Therefore, patients in this subset should be
referred for multidisciplinary evaluation be-
fore embarking on definitive treatment. Level
of evidence: poor; benefit: none; grade of
recommendation: I
7. Whenever possible, induction (neoadjuvant)
therapy followed by surgery for stage IIIA
disease should be carried out in the setting of
a clinical trial. Level of evidence: fair; benefit:
moderate; grade of recommendation: B
8. Bimodality or trimodality therapy is better
than surgery alone for locally advanced stage
IIIA lung cancer. Level of evidence: good;
benefit: substantial; grade of recommenda-
tion: A
Surgical Considerations in Stage IIIA
3
Although the use of neoadjuvant chemotherapy
and/or radiotherapy appears to have potential advan-
tages in the treatment of locally advanced lung
cancer, concern has been raised in numerous publi-
cations about the perceived and real increase in
morbidity and mortality of the subsequent lung
resections. One of the most recent reports by Rob-
erts and associates
76
in 2001 found neoadjuvant
chemotherapy increased the perioperative complica-
tions in their series of 34 patients. However, other
groups, such as Sonett and colleagues
77
in 1999,
reported safe pulmonary resections after chemother-
apy and high-dose thoracic radiation in 19 patients.
Siegenthaler and associates
78
at the M.D. Anderson
Cancer Center in a larger group of patients found no
increased surgical morbidity with preoperative che-
motherapy in lung cancer when compared to their
nonchemotherapy lung resection patients.
There is no doubt that patients with locally ad-
vanced lung cancer who undergo neoadjuvant ther-
apy present more intraoperative, technical chal-
lenges to the thoracic surgeon and require more
careful postoperative care. But with certain extra
precautions, safe lung resections are indeed possible,
especially if the surgeon is experienced with this
patient population and is performing a high volume
of lung resections. As early as 1992, Romano and
Mark
79
reported that hospitals performing a high
volume of lung resections experienced significantly
better outcomes compared to lower volume hospi-
tals. Using the Surveillance, Epidemiology, and End
Results Cancer Registries that are linked to data on
Medicare hospitalizations database, Bach and col-
leagues
80
in 2001 reviewed 2,118 patients from 76
hospitals sampled from 22 states. They found that
patients who undergo lung cancer resections at
hospitals that perform large numbers of the proce-
dures are more likely to survive longer than patients
who undergo such surgery at hospitals performing a
small number of lung resections.
80
Finally, Silvestri
and associates
81
reviewed the South Carolina state-
wide results of lung cancer resections in all nonfed-
eral acute care hospitals from 1991 to 1995. They
found that the mortality for lung cancer resection
was lower when the surgery was performed by a
thoracic surgeon compared to a general surgeon.
81
The definition of what is meant by resectable,
marginally resectable, and unresectable is not
clear in most published studies. The problem is that
this determination is subjective and highly depen-
dent on the experience and expertise of the thoracic
surgeon. For the best possible evaluation of an
induction therapy candidate, the surgeon who ulti-
mately may operate on the stage IIIA patient needs
to be experienced in the handling of these more
complex and technically challenging patients. Also, it
is critically important that the surgeon also is in-
volved initially in the beginning of the evaluation
such that an informed estimate of the surgical re-
sectability of the tumor can be made up front, so that
appropriate candidates for induction therapy are
chosen.
The decision to proceed with surgery after induc-
tion therapy should not be automatic. While there is
evidence that 60 to 75% of patients will respond to
induction regimens, nonresponders should not un-
dergo surgery. In the phase II Southwestern Oncol-
ogy Group trial of induction chemoradiotherapy
followed by surgery in stage IIIA and IIIB disease,
there was complete pathologic clearance of tumor in
22% of resection specimens with an overall 27%
3-year survival.
82
But of particular interest, the pa-
tients with a complete pathologic clearing of residual
disease had a 30-month median survival compared to
10 months for those with residual tumor in the
lymph nodes (p 0.0005). A more recent study by
Bueno and associates
83
emphasized the importance
of residual nodal disease after induction therapy in
stage IIIA tumors. In their study, the long-term
survival stratified by nodal status after induction
therapy and lung resection found that 28% of pa-
tients downstaged to pathologic N0 had a 35.8%
5-year survival, whereas the remainder of patients
with residual nodal disease at surgery had only a 9%
5-year survival. These and other studies suggest that
surgical resection should be avoided in patients after
induction therapy who have definite, biopsy-proven
residual tumor in the mediastinal nodes.
Clinical restaging with standard chest CT scans is
not accurate enough to predict pathologic response
in the lymph nodes, as recently reported by Marga-
ritora and colleagues.
84
The use of PET after induc-
tion therapy to determine response to therapy looks
promising, with 100% accuracy in one small prelim-
inary trial.
85,86
In a retrospective review of the
accuracy of PET scans after induction chemother-
apy, radiotherapy, or both in 56 patients who under-
went subsequent surgery, Akhurst and colleagues
87
found that PET had a 98% positive predictive value
for detecting residual viable disease in the primary
tumor. However, PET overstaged the nodal status in
33%, understaged it in 15%, and was correct in only
52%. Therefore, until further studies are available, it
is premature to routinely employ postinduction ther-
apy PET scans for restaging in order to make
decisions about surgical resectability and nodal in-
volvement. Finally, careful re-evaluation for surgery
after induction therapy is necessary since incomplete
resection or thoracotomy with no resection results in
a poor survival in the stage IIIA patient. And if after
a thoracotomy and resection in the stage IIIA patient
there is known residual nodal disease or there was an
incomplete resection of the primary tumor, postop-
erative radiotherapy should be considered to aid in
local control.
Recommendations
9. Surgical Consideration: Patients with incom-
plete resections have poor survivals, and de-
bulking procedures should be avoided. Level
of evidence: fair; benefit: negative; grade of
recommendation: D
10. Surgical Consideration: Patients with incom-
plete resections and those with residual nodal
disease found at surgery should be considered
for postoperative radiotherapy. Level of evi-
dence: poor; benefit: moderate; grade of rec-
ommendation: B
Unresectable, Bulky N2 Disease (Stage IIIA
4
)
Many patients with stage IIIA lung cancer have
less favorable presentations of their disease because
they have bulky nodal involvement and/or unresect-
able primary tumors. Evaluation of various trials in
this subset of patients is complicated by a lack of
definition of what constitutes bulky nodal disease
as well as what is unresectable. It is generally
agreed that mediastinal lymph nodes 1 cm in
short-axis diameter are suspicious. We then would
define bulky nodal disease as those involving lymph
nodes 2 cm in short-axis diameter measured by
CT, especially with extranodal involvement, multi-
station nodal disease, and/or groupings of multiple,
positive smaller lymph nodes. Nevertheless, this
determination is somewhat subjective, much like the
definition of resectability, which falls back to the
experience and judgment of the thoracic surgeon.
However, aside from the relatively few question-
able presentations, most experienced lung cancer
clinicians can agree on what constitutes unresect-
able, bulky N2 stage IIIA disease that warrants only
nonsurgical therapy. Traditionally, these patients
with locally advanced disease were treated with
conventional radiotherapy alone with relatively poor
long-term survivals, but in the last decade combina-
tion chemoradiotherapy appears to offer improved
results, as discussed in the following sections.
Radiotherapy Alone: Early attempts to use non-
surgical treatment modalities for unresectable locally
advanced disease (our stage IIIA
4
) involved single
modality chest radiotherapy, yielding poor survivals
at 5 years of 5 to 10% with traditional dose and
fractionation schedules (1.8 to 2.0 Gy per fraction
per day to 60 to 70 Gy in 6 to 7 weeks). Patterns of
failure for patients treated with radiotherapy alone
included both locoregional and distant failures. At-
tempts to improve on locoregional control tested
alternative radiotherapy doses and schedules, apply-
ing radiotherapy at escalating doses at shortened
intervals (hyperfractionation) that, in theory, would
maximize cell killing in lung cancers with relatively
short doubling times. A hyperfractionated, higher-
dose radiotherapy trial utilized from 60.0 to 79.2 Gy,
delivered in smaller-than-standard fractions admin-
istered in two fractions per day rather than one.
Hyperfractionation of radiotherapy yielded an im-
proved but still poor 2-year survival of 20%, with an
apparent benefit for patients treated at 69.6 Gy.
There appeared to be acceptable acute or late
toxicity using the hyperfractionated schedule.
88
Further alterations of standard dose and fraction-
ation led to testing accelerated hyperfractionation. In
the United Kingdom, three radiotherapy fractions
were delivered per day in a continuous schedule (7
days rather than 5 days per week) over 12 days to a
total dose of 50.4 Gy or 54 Gy. This continuous
hyperfractionated accelerated radiation therapy
(CHART) regimen yielded good radiographic re-
sponses in tumors with an acceptable early and late
toxicity profile. In a randomized trial comparing
CHART with a standard dose and fractionation
radiotherapy regimen in locally advanced NSCLC,
there was a survival advantage for CHART.
89
Amer-
ican groups have utilized versions of CHART that
eliminate the weekend doses and deliver multiple
daily fractions within an 8-h time period, referred to
as hyperfractionated accelerated radiation therapy
(HART). A recent Eastern Cooperative Oncology
Group (ECOG) pilot study (ECOG 4593) utilized
this schedule and obtained a preliminary median
survival of 13 months with acceptable toxicities,
primarily esophagitis, at the completion of radiother-
apy.
90
In a subsequent companion quality-of-life
assessment of patients undergoing the accelerated
HART regimen in EGOG 4593, Auchter and asso-
ciates
91
found that the decrement in physical and
functional quality of life during treatment returned
to baseline within 4 weeks of completing treatment.
However, the emotional well-being of patients im-
proved at all time points.
91
Recently, the ECOG conducted a multicenter trial
for unresectable, locally advanced NSCLC in which
patients were randomized after induction chemo-
therapy to standard fractionation radiotherapy to a
total dose of 64 Gy or HART to a total dose of 57.6
Gy in a randomized design. One hundred thirty
patients were entered into the trial that is now
closed, and the analysis is pending. An important
trial end point was the local failure rate, as it is hoped
that HART will yield improved local control of
disease and thereby impact favorably on survival.
Combined Chemotherapy With Radiotherapy: Al-
though patients have gained symptomatic benefit
with radiotherapy for unresectable, bulky locally
advanced stage IIIA disease, their outcome has
generally been poor, usually as a result of systemic,
not local, failure. With the development of more
effective platinum-based chemotherapy, attempts to
improve outcome of treatment by decreasing relapse
from distant disease have prompted the addition of
systemic chemotherapy to definitive radiotherapy.
Chemotherapy has been combined with radiother-
apy in different fashions (chemotherapy followed by
radiotherapy, chemotherapy with closely sequenced
radiotherapy, chemotherapy concurrent with radio-
therapy, or induction chemotherapy followed by
concurrent chemotherapy/radiotherapy) in multiple
phase II trials involving heterogeneous and often
poorly staged groups of patients with locally ad-
vanced disease.
In general, trials using platinum-containing che-
motherapy regimens in combination with radiother-
apy have shown good tumor response rates and have
suggested an improvement in survival. One promis-
ing pilot trial showed significantly improved median
and 2-year survivals of 16 months and 30%, respec-
tively, using four cycles of etoposide and cisplatin
with concurrent radiotherapy to 60 Gy.
92
Looking at
collective data from multiple phase II trials, acute
and late toxicities associated with combined chemo-
therapy and radiotherapy have included mild-to-
severe esophagitis, pneumonitis, and also treatment-
related deaths. Overall, however, these trials showed
the feasibility of combined modality therapy and
suggested that chemotherapy plus radiotherapy
would yield improved outcomes compared to radio-
therapy alone.
Multiple phase III trials using platinum chemo-
therapy plus radiotherapy have confirmed improved
survivals for chemotherapy plus radiotherapy com-
pared to radiotherapy alone. Selected key trials are
outlined in Table 6, with some trials discussed below.
Of note, the earliest trial results were negative,
showing no survival benefit with chemotherapy but
the regimens used had either low-dose cisplatin or
non-platinum-based chemotherapy, which might be
expected to be ineffective. Later trials using more
appropriate dose chemotherapy all had positive re-
sults.
A pivotal Cancer and Leukemia Group B
(CALGB) randomized trial initially presented in
1990 showed the benefit of adding chemotherapy in
a sequential fashion to radiotherapy in the setting of
locally advanced disease.
102
The CALGB study com-
pared two cycles of cisplatin and vinblastine added to
standard fractionation radiotherapy to 60 Gy with
radiotherapy alone in patients with favorable prog-
nostic characteristics (good performance status and
minimal weight loss). Objective tumor response rate
was improved for the chemotherapy-plus-radiother-
apy group compared to radiotherapy alone (56% vs
43%, p 0.012), and survival at 2 years and 5 years
was also improved (26% and 13% vs 13% and 6%,
respectively).
The superiority of combined-modality chemother-
apy plus radiotherapy in a sequential fashion com-
pared to radiotherapy alone has been shown in
several other large randomized trials including a
Radiation Therapy Oncology Group trial showing an
improved 1-year and median survival with chemo-
therapy plus conventional radiotherapy compared to
both conventional radiotherapy and hyperfraction-
ated radiotherapy alone.
101
A multicenter French
study reported by Le Chevalier and associates
105
also
confirmed improved survival for the chemotherapy-
plus-radiotherapy arm compared to radiotherapy
alone (11% vs 5% 3-year survival, respectively) with
an improved distant failure rate for chemotherapy
plus radiotherapy (22% vs 46% at 1 year, respective-
ly). Unfortunately, both treatment groups showed
similarly high locoregional failure with 1-year local
control rates of only 15% and 17%, illustrating the
vexing problem of obtaining good locoregional con-
trol of disease in the locally advanced setting. Three
meta-analyses reviewing 50 trials have confirmed
the survival benefit of combined platinum-based
chemotherapy with radiotherapy over radiotherapy
alone in locally advanced, unresectable lung
cancer.
45,106,107
The most recently published study was a large
British trial randomizing 446 patients with localized,
unresectable disease to two arms: (1) chemotherapy
(mitomycin, ifosfamide, and cisplatin) followed by rad-
ical radiotherapy (median, 50 Gy; range, 40 to 60 Gy),
or (2) radical radiotherapy alone (median, 50 Gy; range,
40 to 64 Gy).
104
This trial allowed performance status
(PS)-2 patients, and 15% of the chemoradiotherapy
arm and 11% of the radiotherapy-only arm were PS-2
patients. The median survival and 2-year survival were
not significantly different (p 0.14) between the two
arms: 11.7 months and 20% in the chemoradiotherapy
arm, and 9.7 months and 16% in the radiotherapy arm.
Inclusion of patients with poorer PS in this trial, unlike
most other trials, is believed to have possibly influenced
the results, particularly in the chemoradiotherapy arm.
Concurrent Chemotherapy and Radiotherapy:
Concurrent chemotherapy with radiotherapy has
been studied in the locally advanced setting through
randomized trials that have attempted to capitalize
on the radiosensitizing properties of chemotherapy.
An EORTC three-arm trial published in 1992 com-
pared radiotherapy (split course) concurrent with
daily or weekly concurrent cisplatin to radiotherapy
alone.
98
There were improved 2-year and 3-year
survivals for daily chemotherapy concurrent with
radiotherapy compared with radiotherapy alone
(26% and 16% vs 13% and 2%, respectively). There
was no significant advantage for the weekly chemo-
therapy-plus-radiotherapy arm, with an intermediate
survival compared to the other arms.
Whether concurrent chemotherapy plus radio-
therapy yields an improvement in survival over se-
quential chemotherapy plus radiotherapy has been
addressed by a few trials, including a large Japanese
randomized trial of 320 patients that compared
chemotherapy (mitomycin, vindesine, and cisplatin
for two cycles) concurrent with split-course daily
radiotherapy to 56 Gy compared to chemotherapy
followed by continuous daily radiotherapy to 56
Gy.
108
Esophagitis rates were low with concurrent
therapy. At 5-year median follow-up, 2-year and
5-year survival was improved for concurrent chemo-
therapy over sequential chemotherapy with radio-
therapy (34.6% and 15.8% vs 27.4% and 8.8%,
respectively). Myelosuppression was greater among
T
a
b
l
e
6
R
a
n
d
o
m
i
z
e
d
C
o
n
t
r
o
l
l
e
d
T
r
i
a
l
s
o
f
S
e
q
u
e
n
t
i
a
l
o
r
C
o
n
c
u
r
r
e
n
t
C
h
e
m
o
r
a
d
i
o
t
h
e
r
a
p
y
v
s
R
a
d
i
o
t
h
e
r
a
p
y
A
l
o
n
e
f
o
r
U
n
r
e
s
e
c
t
a
b
l
e
S
t
a
g
e
I
I
I
N
S
C
L
C
*
S
o
u
r
c
e
Y
e
a
r
P
a
t
i
e
n
t
s
,
N
o
.
T
i
m
i
n
g
C
h
e
m
o
t
h
e
r
a
p
y
/
R
a
d
i
o
t
h
e
r
a
p
y
C
h
e
m
o
t
h
e
r
a
p
y
/
R
a
d
i
o
t
h
e
r
a
p
y
R
e
g
i
m
e
n
s
S
t
u
d
y
R
e
s
u
l
t
A
c
u
t
e
T
o
x
i
c
i
t
y
C
h
e
m
o
t
h
e
r
a
p
y
P
l
u
s
R
a
d
i
o
t
h
e
r
a
p
y
/
R
a
d
i
o
t
h
e
r
a
p
y
,
%
2
-
y
e
a
r
S
u
r
v
i
v
a
l
C
h
e
m
o
t
h
e
r
a
p
y
P
l
u
s
R
a
d
i
o
t
h
e
r
a
p
y
/
R
a
d
i
o
t
h
e
r
a
p
y
,
%
S
o
r
e
s
i
e
t
a
l
9
3
1
9
8
8
9
5
C
o
n
c
u
r
r
e
n
t
C
i
s
p
l
a
t
i
n
/
5
0
G
y
N
e
g
a
t
i
v
e
4
0
/
2
5
M
a
t
t
s
o
n
e
t
a
l
9
4
1
9
8
8
2
3
8
S
e
q
u
e
n
t
i
a
l
p
l
u
s
c
o
n
c
u
r
r
e
n
t
C
y
c
l
o
p
h
o
s
p
h
a
m
i
d
e
d
o
x
o
r
u
b
i
c
i
n
c
i
s
p
l
a
t
i
n
/
5
5
G
y
N
e
g
a
t
i
v
e
1
9
/
1
7
A
n
s
a
r
i
e
t
a
l
9
5
1
9
9
1
1
8
3
C
o
n
c
u
r
r
e
n
t
C
i
s
p
l
a
t
i
n
/
6
0
G
y
N
e
g
a
t
i
v
e
1
5
/
9
M
o
r
t
o
n
e
t
a
l
9
6
1
9
9
1
1
1
4
S
e
q
u
e
n
t
i
a
l
M
A
C
C
/
6
0
G
y
N
e
g
a
t
i
v
e
2
1
/
9
2
1
/
1
6
T
r
o
v
o
e
t
a
l
9
7
1
9
9
2
1
7
3
C
o
n
c
u
r
r
e
n
t
C
i
s
p
l
a
t
i
n
/
4
5
G
y
N
e
g
a
t
i
v
e
1
5
/
7
1
3
/
1
3
S
c
h
a
a
k
e
-
K
o
n
i
n
g
e
t
a
l
9
8
1
9
9
2
3
0
8
C
o
n
c
u
r
r
e
n
t
C
i
s
p
l
a
t
i
n
/
5
5
G
y
(
s
p
l
i
t
c
o
u
r
s
e
)
P
o
s
i
t
i
v
e
4
1
/
1
1
2
6
/
1
3
W
o
l
f
e
t
a
l
9
9
1
9
9
4
8
5
S
e
q
u
e
n
t
i
a
l
p
l
u
s
c
o
n
c
u
r
r
e
n
t
V
i
n
d
e
s
i
n
e
-
i
f
o
s
p
h
a
m
i
d
e
-
c
i
s
p
l
a
t
i
n
/
5
0
G
y
P
o
s
i
t
i
v
e
8
.
2
/
1
1
2
4
/
1
2
L
e
C
h
e
v
a
l
i
e
r
e
t
a
l
1
0
0
1
9
9
4
3
5
3
S
e
q
u
e
n
t
i
a
l
V
i
n
d
e
s
i
n
e
-
l
o
u
m
u
s
t
i
n
e
-
c
i
s
p
l
a
t
i
n
-
c
y
c
l
o
p
h
o
s
p
h
a
m
i
d
e
/
6
5
G
y
P
o
s
i
t
i
v
e
2
1
/
1
4
S
a
u
s
e
e
t
a
l
1
0
1
1
9
9
5
4
5
2
S
e
q
u
e
n
t
i
a
l
C
i
s
p
l
a
t
i
n
-
v
i
n
b
l
a
s
t
i
n
e
/
6
0
G
y
P
o
s
i
t
i
v
e
2
6
/
2
2
3
0
/
1
9
D
i
l
l
m
a
n
e
t
a
l
1
0
2
1
9
9
6
1
5
5
S
e
q
u
e
n
t
i
a
l
C
i
s
p
l
a
t
i
n
-
v
i
n
b
l
a
s
t
i
n
e
/
6
0
G
y
P
o
s
i
t
i
v
e
1
4
/
6
2
6
/
1
3
J
e
r
e
m
i
c
e
t
a
l
1
0
3
1
9
9
6
1
3
1
C
o
n
c
u
r
r
e
n
t
C
a
r
b
o
p
l
a
t
i
n
-
e
t
o
p
o
s
i
d
e
/
6
9
.
9
G
y
b
i
d
P
o
s
i
t
i
v
e
5
2
/
3
8
C
u
l
l
e
n
e
t
a
l
1
0
4
1
9
9
9
4
4
6
S
e
q
u
e
n
t
i
a
l
M
i
t
o
m
y
c
i
n
-
i
f
o
s
p
h
a
m
i
d
e
-
c
i
s
p
l
a
t
i
n
/
m
e
d
i
a
n
5
0
G
y
N
e
g
a
t
i
v
e
2
0
/
1
6
(
p
0
.
1
4
)
*
M
A
C
C
m
e
t
h
o
t
r
e
x
a
t
e
-
d
o
x
o
r
u
b
i
c
i
n
-
c
y
c
l
o
p
h
o
s
p
h
a
m
i
d
e
-
C
C
N
U
.
patients in the concurrent arm, but the mortality rate
was low ( 1%) and not significantly different in
both groups. Further evaluation through randomized
trials addressing this issue are required, with an
emphasis on the toxicity profiles of concurrent strat-
egies vs sequential strategies.
Newer-generation chemotherapeutic agents,
alone or in combination with the platinum agents,
are being incorporated into combined modality che-
motherapy plus radiotherapy for locally advanced
disease. As an example, a recent phase II trial in
locally advanced disease used induction paclitaxel
with carboplatin followed by weekly doses concur-
rent with radiotherapy. This treatment yielded a
good response rate (55%) in 38 evaluable patients,
with a 1-year survival of 72% and a tolerable toxicity
profile.
109
Other phase I and II trials have reported the feasi-
bility of combining docetaxel, gemcitabine, and irino-
tecan in concurrent design with radiotherapy but also
do report a range of toxicity profiles. Phase III trials are
needed that incorporate these newer, active agents in
various dosing schedules with radiotherapy in standard
and altered fractionation schedules to define the opti-
mal role of these agents in treatment strategies for
unresectable IIIA (N2) disease.
Recommendations
11. In patients with good PS, radiotherapy should
not be used alone in treating unresectable
stage IIIA lung cancer. Level of evidence:
good; benefit: negative; grade of recommen-
dation: D
12. In the patient with unresectable locally ad-
vanced lung cancer, platinum-based chemo-
therapy plus radiotherapy provides improved
survival rates over radiotherapy alone and
should be used for primary treatment. Level
of evidence: good; benefit: substantial; grade
of recommendation: A
13. Because in patients with stage IIIA lung can-
cer the optimal technique of combining che-
motherapy and radiotherapy has not be deter-
mined, factors such as patient PS and age
should be used to guide treatment planning.
Level of evidence: poor; benefit: small; grade
of recommendation: I
Ongoing Clinical Trials
Perhaps the greatest challenge to the clinician in
the optimal management of stage IIIA disease is the
lack of meaningful, definitive data from large ran-
domized trials on which to base treatment decisions.
A large number of phase I and II trials are accruing
involving locally advanced disease with newer che-
motherapy agents, newer radiotherapy delivery tech-
niques and fractionation schedules, and novel inter-
ventions such as vaccines and gene-based therapy.
Fortunately, a number of large, multicenter phase
III randomized trials are also ongoing, and on com-
pletion should provide results that serve as the basis
for rationale treatment recommendations in the var-
ious clinical presentations of stage IIIA disease.
A. Adjuvant Therapy Phase III Randomized Trials
For fully resected minimal nodal disease patients
(stage IIIA
12
), multiple randomized trials of post-
operative chemotherapy are currently accruing, al-
though recruitment is hampered by the reluctance of
postsurgical patients to undergo chemotherapy and,
particularly, to complete all planned cycles. Equally
difficult is the reluctance of patients to be potentially
randomized in phase III trials to the observation (no
treatment) arm.
1. National Cancer Institute of Canada Clinical
Trials Group/ECOG (JBR-10): This study compares
adjuvant vinorelbine and cisplatin to no chemother-
apy (completed accrual and now closed).
2. FRE-IALT/EU-96010: A European-based
worldwide study comparing adjuvant combination
chemotherapy with cisplatin plus vindesine, vinblas-
tine, vinorelbine, or etoposide to no chemotherapy in
resected stage I, II, and IIIA disease.
3. CNR-NICO-01/EU-97010: A European study
of adjuvant cisplatin and etoposide after completely
resected stage I, II, and IIIA disease.
4. LLCG-BLT/EU-98003/MRC-BLT: A study
comparing adjuvant cisplatin-based chemotherapy to
no chemotherapy.
B. Neoadjvuant (Induction) Therapy Phase III
Randomized Trials in Resectable Stage IIIA
1. North American Intergroup 0139: This trial
compares concurrent combination chemotherapy
with cisplatin and etoposide plus radiotherapy fol-
lowed by surgery or radiotherapy in stage IIIA (N2)
disease (completed accrual and now closed).
2. EORTC 08941: A European study comparing
platinum-based chemotherapy of choice followed (in
responders only) by surgery or radiotherapy.
C. Combination Chemotherapy and Radiotherapy
in Unresectable Stage IIIA: Phase III Randomized
Trials
1. National Cancer Institute T99-0046: This study
compares platinum-based combination chemother-
apy and radiotherapy with or without shark cartilage
extract (AE-941).
2. Southwest Oncology Group S0023: A study of
cisplatin, etoposide, docetaxel, and radiotherapy with
or without oral ZD-1839 in patients with unresect-
able stage III NSCLC.
3. CHNT-PC/MIC; EU-99046: A study of pacli-
taxel and carboplatin vs standard platinum (cisplatin
plus mitomycin/ifosfamide or vinblastine/mitocycin)
therapy in patients with inoperable stage IIIA, IIIB,
or IV NSCLC.
4. ECOG-3598: A study of carboplatin, paclitaxel,
and radiotherapy with or without thalidomide in
patients with unresectable stage IIIA or IIIB
NSCLC.
5. Radiation Therapy Oncology Group-9801: A
study of amifostine mucosal protection in patients
with favorable performance inoperable stage II,
IIIA, or IIIB NSCLC receiving sequential induction
and concurrent hyperfractioned radiotherapy with
paclitaxel and carboplatin.
Despite many earlier studies, the optimal treat-
ment recommendations in the various clinical pre-
sentations of stage IIIA (N2) disease are unclear.
Hopefully, as the current phase III trials accrue and
mature and the much-needed, subsequent random-
ized trials with newer chemotherapy agents and
radiotherapy schemata are started and completed,
more definitive treatment guidelines will emerge.
Until that time, it is critically important that when-
ever possible the clinician who manages locally
advanced NSCLC enroll their patients in every
available clinical trial.
A. Incidental (Occult) N2 Disease Found at
Thoracotomy
1. Surgical Consideration: In patients with an
occult single-station mediastinal node metas-
tasis that is recognized at thoracotomy and
when a complete resection of the nodes and
primary tumor is technically possible, then
proceed with the planned lung resection and a
mediastinal lymphadenectomy. Level of evi-
dence: poor; benefit: small; grade of recom-
mendation: C
2. Surgical Consideration: In every patient un-
dergoing a lung resection for lung cancer,
systematic mediastinal lymph node sampling
or complete mediastinal lymph node dissec-
tion must be performed. Level of evidence:
good; benefit: substantial; grade of recom-
mendation: A
3. Adjuvant Radiotherapy: In the patient with
fully resected stage IIIA lung cancer, there is
no definite improvement in survival with ad-
juvant postoperative radiotherapy, but it sig-
nificantly reduces local recurrence and should
be considered in selected patients. Level of
evidence: fair; benefit: small; grade of recom-
mendation: C
4. Adjuvant Chemotherapy: In the patient with
fully resected stage IIIA lung cancer, adjuvant
chemotherapy administered alone might offer
a very modest survival advantage, but this
modality should not be routinely utilized out-
side of a clinical trial. Level of evidence: poor;
benefit: small; grade of recommendation: I
5. Adjuvant Chemoradiotherapy: In the patient
with fully resected stage IIIA lung cancer,
based on randomized clinical trials to date,
there is no survival benefit appreciated by
adding postoperative adjuvant chemotherapy
to adjuvant radiotherapy. Therefore, the rou-
tine use of combined postoperative chemo-
therapy and radiotherapy is not recom-
mended, and should not be employed outside
of a clinical trial. Level of evidence: fair;
benefit: none; grade of recommendation: D
B. Potentially Resectable N2 Disease
6. Induction (Neoadjuvant) Therapy: Patients
with stage IIIA (N2) lung cancer identified
preoperatively have a relatively poor prognosis
when treated with surgery as a single modal-
ity. Several small trials of induction chemo-
therapy have yielded conflicting results about
its effect on survival. The relative roles of
surgery and radiation therapy as the local
treatment modality are also not clearly de-
fined. Definitive treatment recommendations
are difficult to make in this setting. Therefore,
patients in this subset should be referred for
multidisciplinary evaluation before embarking
on definitive treatment. Level of evidence:
poor; benefit: none; grade of recommenda-
tion: I
7. Induction (Neoadjuvant) Therapy: Whenever
possible, induction (neoadjuvant) therapy fol-
lowed by surgery for stage IIIA disease should
be carried out in the setting of a clinical trial.
Level of evidence: fair; benefit: moderate;
grade of recommendation: B
8. Induction (Neoadjuvant) Therapy: Bimodality
or trimodality therapy is better than surgery
alone for locally advanced stage IIIA lung
cancer. Level of evidence: good; benefit: sub-
stantial; grade of recommendation: A
9. Surgical Consideration: Incompletely re-
sected patients have poor survival, and de-
bulking procedures should be avoided. Level
of evidence: fair; benefit: negative; grade of
recommendation: D
10. Surgical Consideration: Incompletely re-
sected patients and those with residual nodal
disease found at surgery should be considered
for postoperative radiotherapy. Level of evi-
dence: poor; benefit: moderate; grade of rec-
ommendation: B.
C. Unresectable, Bulky N2 Disease
11. Combination Chemotherapy and Radiother-
apy: In patients with good PS, radiotherapy
should not be administered alone in treating
unresectable stage IIIA lung cancer. Level of
evidence: good; benefit: negative; grade of
recommendation: D
apy: In patients with unresectable locally ad-
vanced lung cancer, platinum-based chemo-
therapy plus radiotherapy provides improved
survival rates over radiotherapy alone and
should be used for primary treatment. Level
of recommendation: A
apy: Because in patients with stage IIIA lung
cancer the optimal technique of combining
chemotherapy and radiotherapy has not be
determined, then factors such as patient per-
formance status and age should be used to
guide treatment planning. Level of evidence:
poor; benefit: small; grade of recommenda-
tion: I
References
2 Ruckdeschel JC. Combined modality therapy of non-small
cell lung cancer. Semin Oncol 1997; 24:429439
3 Ettinger DS, Kris MG. NCCN Non-Small Cell Lung Cancer
Practice Guidelines Panel. NCCN: non-small cell lung
cancer. Cancer Control 2001; 8:2231
4 ESMO. ESMO minimum clinical recommendations for
diagnosis, treatment and follow-up of non-small cell lung
cancer. Ann Oncol 2001; 12:10491050
5 The Royal College of Radiologists Clinical Oncology Net-
work. Guidelines on the non-surgical management of lung
cancer. Clin Oncol (R Coll Radiol) 1999; 11:S1S53
6 British Thoracic Society Guidelines. Guidelines on the
selection of patients with lung cancer for surgery. Thorax
2001; 56:89108
7 Clinical practice guidelines for the treatment of unresect-
able non-small-cell lung cancer. American Society of Clini-
cal Oncology. J Clin Oncol 1997; 15:29963018
8 Ginsberg R, Roth, J, Fergusson, M. Lung cancer surgical
practice guidelines. Society of Surgical Oncology prac-
tice guidelines. Oncology (Huntington) 1997; 11:889892
9 Goss G, Paszat, L, Newman, TE, et al. Use of preoperative
chemotherapy with or without postoperative radiotherapy in
technically resectable stage IIIA non-small cell lung cancer.
Provincial Lung Cancer Disease Site Group. Cancer Prev
Control 1998; 2:3239
10 Logan DM, Lochrin CA, Darling G, et al. Adjuvant radio-
therapy and chemotherapy for stage II or IIIA non-small-
cell lung cancer after complete resection. Provincial Lung
Cancer Disease Site Group. Cancer Prev Control 1997;
1:366378
11 Association of Community Cancer Centers. ACCC Oncol-
ogy Patient Management Guidelines. version 3.0. Available
at www.accc-cancer.org/publications/newguide.asp. Accessed
December 20, 2002
12 National Cancer Institute. Non-small cell lung cancer
(PDQ): treatment. Health professionals. Available at: www.
cancernet.gov/cancerinfo/pdq/treatment/non-small-cell-
lung/healthprofessional/. Accessed December 20, 2002
13 Scottish Intercollegiate Guidelines Network. Management
of lung cancer: a national clinical guideline recommended
for use in Scotland. Available at: www.sign.ac.uk/guidelines/
published/index.html. Accessed December 20, 2002
14 Improving outcomes in lung cancer: the manual. Guidance
on Commissioning Cancer Services. UK: National Health
Service Executive, 1998
15 American College of Radiology. ACR Appropriateness Cri-
teria. Neoadjuvant therapy for marginally resectable (clinical
N2) non-small cell lung cancer. Available at: www.acr.org/
dyna/docdepartments/appropriateness_criteria/text.html.
Accessed November 26, 2002 (13311345)
teria. Postoperative radiotherapy in non-small cell lung
cancer. Available at: www.acr.org/dyna/docdepartments/
appropriateness_criteria/text.html. Accessed November 26,
2002 (12951318)
teria. Non-small cell lung cancer, nonsurgical, aggressive
therapy. Available at: www.acr.org/dyna/docdepartments/
appropriateness_criteria/text.html. Accessed November 26,
2002 (113)
18 Fernando HC, Goldstraw, P. The accuracy of clinical eval-
uative intrathoracic staging in lung cancer as assessed by
postsurgical pathologic staging. Cancer 1990; 65:25032506
19 Goldstraw P, Mannam, GC, Kaplan, DK, et al. Surgical
management of non-small cell lung cancer with ipsilateral
mediastinal node metastases (N2 disease). Thorac Cardio-
vasc Surg 1994; 107:1927
20 Should subcarinal nodes be routinely examined in patients
with non-small cell lung cancer? The Lung Cancer Study
Group. Thorac Cardiovasc Surg 1988; 95:883887
21 Ginsberg RJ. Multimodality therapy for Stage IIIA (N2)
lung cancer: an overview. Chest 1993; 103:356S359S
22 Gozetti G, Mastorilli M, Bragaglia RB, et al. Surgical
management of N2 lung cancer [abstract]. Lung Cancer
1986; 2:96
23 Martini N, McCaughan B, Bains MS, et al. Improved
survival in resection for selected Stage III lung cancer
[abstract]. Lung Cancer 1986; 2:97
24 Naruke T, Suemasu K, Ishikawa S, et al. Significance of
surgery in lung cancer [abstract]. Lung Cancer 1986; 2:96
25 Sawamura K, Mori T, Hasimoto S, et al. Results of surgical
treatment of N2 disease [abstract]. Lung Cancer 1986; 2:96
26 van Klaveren RJ, Festen J, Otten HJ, et al. Prognosis of
unsuspected but completely resectable N2 non-small cell
27 Watanabe Y, Shimizu J, Hayashi Y, et al. Results of surgical
treatment on patients with stage IIIA non-small lung cancer.
28 Martini N, Flehinger BJ. The role of surgery in N2 lung
cancer. Surg Clin North Am 1987; 67:10371049
29 Maggi G, Casadio C, Mancuso M, et al. Resection and
radical lymphadenectomy for lung cancer: prognostic signif-
icance of lymphatic metastases. Int Surg 1990; 75:1721
30 Izbicki R, Thetter O, Habkost M, et al. Radical systematic
mediastinal lymphadenectomy in non-small cell lung cancer:
a randomized controlled trial. Br J Surg 1994; 81:229235
31 Keller SM, Adak S, Wagner H, et al. A randomized trial of
postoperative adjuvant therapy in patients with completely
resected stage II or IIIA non-small-cell lung cancer. N Engl
J Med 2000; 343:12171222
32 Keller SM, Adak S, Wagner H, et al. Mediastinal lymph
node dissection improves survival in patients with stages II
and IIIA non-small cell lung cancer. Eastern Cooperative
Oncology Group. Ann Thorac Surg 2000; 70:358365
33 Choi N, Grillo H, Gardiella M, et al. Basis of new strategies
in postoperative radiotherapy of bronchogenic carcinoma.
Int J Radiat Oncol Biol Phys 1980; 6:3135
34 Green N, Kurohara S, George F, et al. Postresection
radiation for primary lung cancer. Radiology 1975; 116:405
407
35 Kirsch M, Sloan H. Mediastinal metastasis in bronchogenic
carcinoma: influence of postoperative irradiation, cell type,
and location. Ann Thorac Surg 1982; 33:459463
36 Effects of postoperative mediastinal radiation on completely
resected stage II and stage III epidermoid carcinoma of the
lung. Lung Cancer Study Group. N Engl J Med 1986;
315:13771381
37 Paterson R, Russel MH. Clinical trials in malignant disease:
Part IV. Lung cancer; value of postoperative radiation
therapy. Clin Radiol 1962; 13:141144
38 Bagma PJ. Post-operative radiotherapy in carcinoma of the
bronchus. In: Deeley TJ, ed. Carcinoma of the bronchus.
London, UK: Butterworths, 1971; 163170
39 Van Houtte PV, Rocmans P, Smets P, et al. Postoperative
radiation therapy in lung cancer: a controlled trial after
resection of curative design. Int J Radiat Oncol Biol Phys
1980; 6:983986
40 Weisenberger T. Effects of postoperative mediastinal radi-
ation on completely resected stage II and stage III epider-
moid cancer of the lung. Lung Cancer Study Group 773.
N Engl J Med 1985; 315:13771381
41 Stephens R, Girling D, Bleehen N, et al. The role of
postoperative radiotherapy in non-small-cell lung cancer: a
multicentre randomised trial in patients with pathologically
staged T12, N12, M0 disease. Br J Surg 1996; 73:632639
42 Debevec M, Bitenc M, Vidmar S, et al. Postoperative
radiotherapy for radically resected N2 non-small cell lung
cancer (NSCLC): randomised clinical study 19881992.
Lung Cancer 1996; 13:99107
43 Dautzenberg B, Arriagada R, Chammard AB, et al. A
controlled study of postoperative radiotherapy for patients
with completely resected non-small cell lung cancer. Cancer
1999; 86:265273
44 Postoperative radiotherapy in non-small-cell lung cancer:
systematic review and meta-analysis of individual patient
data from nine randomised controlled trials. PORT Meta-
analysis Trialist Group. Lancet 1998; 352:257263
45 Chemotherapy in non-small cell lung cancer: a meta-analysis
using updated data on individual patients from 52 random-
ized clinical trials. Non-Small Cell Lung Cancer Collabora-
tive Group. BMJ 1995; 311:899909
46 Niiranen A, Niitamo-Korhonen S, Kouri M, et al. Adjuvant
chemotherapy after radical surgery for non-small-cell lung
cancer. J Clin Oncol 1992; 10:19271932
47 Ohta M, Tsuchujz, R, Shimoyama, M, et al. Adjuvant
chemotherapy for completely resected stage III non-small
cell lung cancer. Thorac Cardiovasc Surg 1993; 106:703708
48 Feld R, Rubinstein L, Thomas PA. Adjuvant chemotherapy
with cyclophosphamide, doxorubicin, and cisplatin in pa-
tients with completely resected stage I non-small cell lung
cancer. Lung Cancer Study Group. J Natl Cancer Inst 1993;
85:299306
49 Figlin RA, Piantodosi S. A phase 3 randomized trial of
immediate combination chemotherapy vs delayed combina-
tion chemotherapy in patients with completely resected
stage II and III non-small cell carcinoma of the lung. Lung
Cancer Study Group. Chest 1994; 106:310S312S
50 A randomized trial of postoperative chemotherapy in non-
small-cell lung cancer (the second cooperative study). Study
Group of Adjuvant Chemotherapy for Lung Cancer. J Surg
Oncol 1995; 21:6977
51 Wada H, Hitomi S, Teramatsu T. Adjuvant chemotherapy
after complete resection in non-small cell lung cancer. West
Japan Study Group for Lung Cancer Surgery. J Clin Oncol
1996; 14:10481054
52 Le Chevalier T. Adjuvant chemotherapy in non-small cell
lung cancer. Semin Oncol 1998; 25:6265
53 Lad T, Rubinstein L, Sadeghi A. The benefit of adjuvant
treatment for locally advanced lung cancer. The Lung
Cancer Study Group. J Clin Oncol 1988; 6:917
54 Sawamura K, Mori T, Doi O, et al. A prospective random-
ized controlled study of the postoperative adjuvant therapy
for non-small cell lung cancer [abstract]. Lung Cancer 1988;
4:A166
55 Pisters KMW, Kris MG, Gralla RJ, et al. Randomized trial
comparing postoperative chemotherapy with vindesine and
cisplatin plus thoracic irradiation with irradiation alone in
stage III (N2) non-small cell lung cancer. J Surg Oncol 1994;
56:236241
56 Dautzenberg B, Chastang C, Arriagada C, et al. Adjuvant
radiotherapy vs combined sequential chemotherapy fol-
lowed by radiotherapy in the treatment of resected non-
small cell lung carcinoma: a randomized trial of 267 patients.
Cancer 1995; 76:779786
57 Schiller JH, Adak S, Feins RH, et al. Lack of prognostic
significance of p53 and K-ras mutations in primary resected
non-small-cell ling cancer on E4592: a laboratory ancillary
study on an Eastern Cooperative Oncology Group prospec-
tive randomized trial of postoperative adjuvant therapy.
J Clin Oncol 2001; 19:448457
58 Rusch VW. Surgery for stage III non-small cell lung cancer.
Cancer Control 1994; 1:455466
59 Pearson FG, DeLarue NC, Ilves R, et al. Significance of
positive superior mediastinal nodes identified at mediasti-
noscopy in patients with resectable cancer of the lung.
60 Naruke T, Suemasu K, Ishikawa S. Lymph node mapping
and curability at various level of metastasis in resected lung
61 Should subcarinal lymph nodes be routinely examined in
patients with non-small cell lung cancer? The Lung Cancer
Study Group. Thorac Cardiovasc Surg 1988; 95:883887
62 Miller DL, McManus KG, Allen MS, et al. Results of
surgical resection in patients with N2 non-small cell lung
cancer. Ann Thorac Surg 1994; 57:10951100
63 Okada M, Tsubota N, Yoshimura M, et al. Prognosis of
completely resected pN2 non-small cell lung carcinomas:
what is the significant node that affects survival? Thorac
64 Martini N, Kris MG, Flehinger BJ, et al. Preoperative
chemotherapy for stage IIIA (N2) lung cancer: The Sloan-
Kettering experience with 136 patients. Ann Thorac Surg
1993; 55:13651374
65 Pass HI, Pogrebniak HW, Steinberg SM, et al. Randomized
trial comparing neoadjuvant therapy for lung cancer: interim
analysis. Ann Thorac Surg 1992; 53:992998
66 Fleck J, Camargo J, Godot D, et al. Chemoradiation therapy
(CRT) vs chemotherapy (CT) alone as a neoadjuvant treat-
ment for stage III non-small cell lung cancer (NSCLC):
preliminary report of a phase III prospective randomized
trial [abstract]. Proc Am Soc Clin Oncol 1993; 12:A1108
67 Rosell R, Gomez-Codina J, Camps C, et al. Preresectional
chemotherapy in stage IIIA non-small cell lung cancer.
Lung Cancer 1999; 26:714
68 Rosell R, Gomez-Codina J, Camps C, et al. A randomized
trial comparing preoperative chemotherapy plus surgery
with surgery alone in patients with non-small cell lung
69 Roth JA, Fossella F, Komaki R, et al. A randomized trial
comparing perioperative chemotherapy and surgery with
surgery alone in resectable stage IIIA non-small cell lung
70 Roth JA, Atkinson EN, Fossella F, et al. Long-term fol-
low-up of patients enrolled in a randomized trial comparing
perioperative chemotherapy and surgery. Lung Cancer
1998; 21:16
71 Wagner H, Lad T, Piantadosi S, et al. Randomized phase 2
evaluation of preoperative radiation therapy and preopera-
tive chemotherapy with mitomycin, vinblastine, and cisplatin
in patients with technically unresectable stage IIIA and IIIB
non-small cell lung cancer. LCSG 881. Chest 1994; 106:
348S354S
72 Elias A, Herdon J, Kumar P, et al. A phase III comparison
of best local-regional therapy with or without chemother-
apy for stage IIIA T13N2 non-small cell ling cancer
(NSCLC): preliminary results [abstract]. Proc Am Soc Clin
Oncol 1997; 16:A1622
73 Ichinose Y, Tsuchiya R, Kato H. Randomized trial of
chemotherapy followed by surgery for stage IIIA non-small
cell lung cancer. The Japan Clinical Oncology Group (lung
cancer surgical study group) 9209 [abstract]. Lung Cancer
2000; 29:173
74 Depierre A, Milleron B, Moro-Sililot D, et al. Preoperative
chemotherapy followed by surgery compared with primary
surgery in resectable stage I (except T1N0), II, and IIIA
non-small lung cancer. J Clin Oncol 2002; 20:247253
75 Garland L, Robinson LA, Wagner H. Evaluation and man-
agement of patients with stage IIIA (N2) non-small-cell lung
cancer. Chest Surgery Clin North Am 2001; 11:69100
76 Roberts JR, Eustis C, DeVore RF, et al. Induction chemo-
therapy increases perioperative complications in patients
undergoing resection for non-small cell lung cancer. Ann
Thorac Surg 2001; 72:885888
77 Sonett JR, Krasna MJ, Suntharalingam M, et al. Safe
pulmonary resection after chemotherapy and high dose
thoracic radiation. Ann Thorac Surg 1999; 68:316320
78 Siegenthaler MP, Pisters KM, Merriman KW, et al. Preop-
erative chemotherapy for lung cancer does not increase
surgical morbidity. Ann Thorac Surg 2001; 71:11051111
79 Romano PS, Mark DH. Patient and hospital characteristics
related to in-hospital mortality after lung resection. Chest
1992; 101:13321337
80 Bach PB, Cramer LD, Schrag D, et al. The influence of
hospital volume on survival after resection for lung resec-
tion. N Engl J Med 2001; 345:181188
81 Silvestri GA, Handy J, Lackland D, et al. Specialists achieve
better outcomes that generalists for lung cancer surgery.
Chest 1998; 114:675680
82 Albain KS, Rusch VW, Crowley JT, et al. Concurrent
cisplatin/etoposide plus chest radiotherapy followed by sur-
gery in stages IIIA (N2) and IIIB non-small cell lung cancer:
mature results of Southwestern Oncology Group phase II
study 8805. J Clin Oncol 1995; 13:18801892
83 Bueno R, Richards W, Swanson S, et al. Nodal stage after
induction therapy for stage IIIA lung cancer determines
survival. Ann Thorac Surg 2000; 70:18261831
84 Margaritora S, Cesario A, Galetta D, et al. Ten year
experience with induction therapy in locally advanced non-
small cell lung cancer (NSCLC): is clinical re-staging pre-
dictive of pathologic staging? Eur J Cardiothorac Surg 2001;
19:894898
85 Vansteenkiste JF, Stroobants SG, De Leyn PR, et al.
Potential use of FDG-PET scan after induction chemother-
apy in surgically staged IIIA-N2 non-small-cell lung cancer:
a prospective pilot study. The Leuven Lung Cancer Group,.
Ann Oncol 1998; 9:11931198
86 Haberkorn U. Positron emission tomography (PET) of
non-small cell lung cancer. Lung Cancer 2001; 34:S115
SS121
87 Akhurst T, Downey RL, Ginsberg MS, et al. An initial
experience with FDG-PET in the imaging of residual
disease after induction therapy in lung cancer. Ann Thorac
Surg 2002; 73:259266
88 Cox J, Azaria N, Byhardt R, et al. A randomized phase I/II
trial of hyperfractionated radiation therapy with total doses
of 60.0 to 79.2 Gy: Possible survival benefits with 69.6 Gy
in favorable patients with stage III Radiation Therapy
Oncology Group 8311. J Clin Oncol 1990; 8:15431555
89 Saunders M, Dische, S, Barrett, A, et al. Continuous
hyperfractionated accelerated radiotherapy (CHART) vs
conventional radiotherapy in non-small-cell lung cancer: a
randomised multicentre trial. Lancet 1996; 350:161165
90 Mehta M, Tannehill S, Adak S, et al. Phase II trial of
hyperfractionated accelerated radiation therapy for non-
resectable non-small-cell lung cancer: Results of Eastern
Cooperative Oncology group 4593. J Clin Oncol 1998;
16:35183523
91 Auchter RM, Scholtens D, Adak S, et al. Quality of life
assessment in advanced non-small-cell lung cancer patients
undergoing an accelerated radiotherapy regimen: Report of
ECOG study 4593. Int J Radiat Oncol Biol Phys 2001;
50:11991206
92 Friess CG, Balkadi M, Harvey WH. Simultaneous cisplatin
and etoposide with radiation therapy in loco-regional non-
small cell lung cancer. In: Grall RJ, Einhorn LH, eds. Small
cell lung cancer and non-small cell lung cancer. London,
UK: Royal Society of Medicine Services Limited, 1989;
121126
93 Soresi E, Clerici M, Grilli R, et al. A randomized clinical trial
comparing radiation therapy vs radiation therapy plus cis-
dichlorodiammine platinum in the treatment of locally-
advanced non-small cell lung cancer. Semin Oncol 1988;
17:2025
94 Mattson K, Holsti LP, Holsti P, et al. Inoperable non-small
cell lung cancer: Radiation with or without chemotherapy.
Eur J Cancer Clin Oncol 1988; 24:477482
95 Ansari R, Tokara R, Fisher W, et al. A phase III study of
thoracic irradiation with or without concomitant cisplatin in
locally advanced unresectable non-small cell lung cancer:
Hoosier Oncology Group study [abstract]. Proc Am Soc Clin
Oncol 1991; 10:241
96 Morton RF, Jett JR, McGinnis WL, et al. Thoracic radiation
therapy alone compared with combined chemoradiotherapy
for locally unresectable non-small cell lung cancer: a ran-
domized phase III trial. Ann Intern Med 1991; 115:681686
97 Trovo NG, Minotel E, Fravelun G, et al. Radiotherapy vs
radiotherapy enhanced by cisplatin in stage III non-small
cell lung cancer. Int J Radiat Oncol Biol Phys 1992;
24:1116
98 Schaake-Koning C, Van den Bogaert W, Dalesio O, et al.
Effects of concurrent cisplatin and radiotherapy on inoper-
able non-small cell lung cancer. N Engl J Med 1992;
326:524530
99 Wolf M, Hans K, Becker H, et al. Radiotherapy alone vs
chemotherapy with ifosfamide/vindesine followed by radio-
therapy in unresectable non-small cell lung cancer. Semin
Oncol 1994; 21:4247
100 Le Chevalier T, Arriagada R, Quoix E, et al. Radiotherapy
alone vs combined chemotherapy and radiotherapy in unre-
sectable non-small cell lung cancer. Lung Cancer 1994;
10:S239S244
101 Sause WT, Scott C, Taylor S, et al. Radiation Therapy
Oncology Group (RTOG) 8809, and Eastern Cooperative
Oncology Group (ECOG) 4588. Preliminary results of a
phase III trial in regionally advanced, unresectable non-
small cell lung cancer. J Natl Cancer Inst 1995; 87:198205
102 Dillman RO, Herndon J, Seagren SL, et al. Improved
survival in stage III non-small cell lung cancer: a seven-year
follow-up of Cancer and Leukemia Group B (CALGB) 8433
trial. J Natl Cancer Inst 1996; 88:12101215
103 Jeremic B, Shibamoto Y, Acimovic L, et al. Hyperfraction-
ated radiation therapy with or without concurrent low-dose
daily carboplatin/etoposide for stage III non-small-cell lung
cancer: a randomized study. J Clin Oncol 1996; 14:1065
1070
104 Cullen MH, Billingham LG, Woodroffe CM, et al. Mitomy-
cin, ifosfamide, and cisplatin in unresectable non-small cell
lung cancer: effects on survival and quality of life. J Clin
Oncol 1999; 17:31883194
105 Le Chevalier T, Arriagada R, Quoix, E, et al. Radiotherapy
alone vs combined chemotherapy and radiotherapy in non-
resectable non-small cell lung cancer: first analysis of a
randomized trial of 353 patients. J Natl Cancer Inst 1991;
83:417423
106 Mariano P, Preatoni A, Cantoni A. Randomized trials of
radiotherapy alone vs combined chemotherapy and radio-
therapy in stages IIIA and IIIB non-small-cell lung cancer.
Cancer 1995; 76:593601
107 Prichard RS, Anthony SP. Chemotherapy plus radiotherapy
compared to radiotherapy alone in the treatment of locally
advanced, unresectable, non-small-cell lung cancer: a meta-
analysis. Ann Intern Med 1996; 125:723729
108 Furuse K, Fukuoka M, Kawahara M, et al. Phase III study of
concurrent vs sequential thoracic radiotherapy in combina-
tion with mitomycin, vindesine, and cisplatin in unresectable
stage III non-small cell lung cancer. J Clin Oncol 1999;
17:26922699
109 Langer C. Concurrent chemoradiation using paclitaxel and
carboplatin in locally advanced non-small cell lung cancer.
Semin Radiat Oncol 1999; 9:108116
2003;123;202-220 Chest
Lary A. Robinson, Henry Wagner, Jr. and John C. Ruckdeschel
Treatment of Stage IIIA Non-small Cell Lung Cancer
& Services
Updated Information
S
References
S#BIBL
free at:
Citations
S#otherarticles
Reprints
2003;123;221-225 Chest
James R. Jett, Walter J. Scott, M. Patricia Rivera and William T. Sause
Guidelines on Treatment of Stage IIIB Non-small Cell Lung Cancer
ISSN: 0012-3692.
Guidelines on Treatment of Stage IIIB
Non-small Cell Lung Cancer*
James R. Jett, MD, FCCP; Walter J. Scott, MD, FCCP;
M. Patricia Rivera MD, FCCP; and William T. Sause, MD, FACR
Stage IIIB includes patients with T4, any N, M0, and any T, N3, M0. Surgery may be indicated
only for carefully selected T4N0M0 patients with or without neoadjuvant chemotherapy or
chemoradiotherapy. Patients with N3 lymph node involvement are not considered as surgical
candidates. For patients with unresectable disease, good performance score, and minimal weight
loss, treatment with combined chemotherapy and radiotherapy has resulted in better survival
than treatment with radiotherapy alone. Multiple daily fractions of radiotherapy have not
resulted in improved survival compared with standard fractionation once daily. Concurrent
chemoradiotherapy appears to be associated with improved survival compared with sequential
chemotherapy and radiotherapy. Treatment of stage IIIB due to malignant pleural effusion is
addressed in the section that deals with stage IV disease. (CHEST 2003; 123:221S225S)
Key words: NSCLC stage IIIB; stage IIIB; unresectable lung cancer; unresectable NSCLC
Abbreviations: CHART continuous hyperfractionated accelerated radiotherapy; Fx fractions; NSCLC non-
small cell lung cancers; PS performance score; RTOG Radiation Therapy Oncology Group
S
tage IIIB disease includes patients with T4 tu-
mors, any N, M0, and any T, N3, M0. It is
estimated that 1015% of all patients are stage IIIB
at the time of diagnosis. The treatment options
depend on the extent of disease and include surgery
alone in carefully selected patients or a combination
of chemotherapy and radiotherapy. Surgical resec-
tion after induction therapy may be appropriate in
selected patients. Radiotherapy alone has been used
in the past but should be limited to patients with
poor performance score. Chemotherapy alone is not
a good treatment option except for patients with
malignant pleural effusion (discussed in the section
on stage IV disease).
Methods
This section of the evidence-based guidelines is
based on an extensive review of the medical litera-
ture, including 8 guidelines, 5 meta-analyses, and 20
manuscripts and abstracts, with an emphasis on
phase III randomized control trials. Selected key
references are included in the bibliography.
Limited Role of Surgery
The 5-year survival of patients with clinically
staged IIIB non-small cell lung cancer (NSCLC) is
3 to 7%.
1
Data on pathologically staged IIIB disease
was not available in the new Mountain International
Classification. Stage IIIB disease includes T4N0
3M0 and T14N3M0. This section will not address
treatment of patients with T4 disease due to malig-
nant pleural or pericardial effusions or IIIB Pancoast
tumors. Malignant effusions will be addressed in the
section on stage IV disease, and Pancoast tumors will
be discussed in the chapter on special treatment
issues.
Surgery may be indicated for stage IIIB disease
only in carefully selected situations.
2
Patients who
are T4N01 due solely to a satellite tumor nodule(s)
within the primary tumor lobe have a 5-year survival
of approximately 20% with surgery alone.
3,4
Individ-
uals with T4N01 disease due to main carinal in-
volvement have been treated with carinal resection
with or without pulmonary resection. Carinal resec-
tion carries an appreciable mortality of 10 to 15%,
with an increased risk of local recurrence.
5,6
The
5-year survival in these carefully selected series is
approximately 20%.
Neoadjuvant chemotherapy, or chemoradiother-
apy followed by surgical resection, has been used in
patients with N2 (IIIA) disease. However, few phase
II series have included carefully selected patients
*From the Mayo Clinic (Dr. Jett), Rochester, MN; Section of
Thoracic Surgical Oncology, Fox Chase Cancer Center (Dr.
Scott), Philadelphia, PA; University of North Carolina (Dr.
Rivera), Chapel Hill, NC; LDS Hospital, University of Utah
(Dr. Sause), Salt Lake City, UT.
Correspondence to: James R. Jett, MD, FCCP, Mayo Clinic, 200
First Street Southwest, Rochester, MN 55905; e-mail: jett.james@
mayo.edu
with T4 primary lesions or N3 nodes. A study by the
Southwestern Oncology Group employed concur-
rent chemoradiotherapy in 51 patients with IIIB
disease that excluded superior vena cava syndrome
and malignant effusions. That study observed a
resectability rate of 80%, with a median survival time
of 17 months and a 3-year survival rate of 24%.
7
These results were similar to those observed in
patients with IIIA disease reported in that same trial.
To date, however, there are no phase III trial data
that demonstrate that neoadjuvant treatment fol-
lowed by surgery in patients with IIIB disease results
in prolonged survival compared with treatment with
combination chemoradiotherapy.
Recommendations
1. Patients with clinical T4N0 NSCLC due to
either satellite tumor nodule(s) in the same
lobe or carinal involvement should be evaluated
by a thoracic surgeon for possible resection.
Level of evidence: fair; benefit: substantial;
grade of recommendation: B.
2. For patients with stage IIIB NSCLC due to T4
(excluding Pancoast tumors) or N3 disease,
treatment with neoadjuvant chemotherapy or
chemoradiotherapy followed by surgery has
been explored in limited phase II trials. At this
time, there are no phase III trial data available
to document that surgery adds to survival;
therefore, this approach should not be consid-
ered as standard therapy. Level of evidence:
poor; benefit: small/weak; grade of recommen-
dation: I.
Radiotherapy Alone vs Combination
Chemotherapy and Radiotherapy
The vast majority of patients with IIIB disease do
not benefit from surgery and are best managed with
chemotherapy plus radiotherapy or with radiother-
apy alone, depending on sites of tumor involvement
and performance score status. A trial by the Cancer
and Acute Leukemia Group B randomly assigned
patients with good performance score (PS 0, 1),
minimal weight loss ( 5%), and stage IIIA or IIIB
disease to treatment with thoracic radiotherapy alone
(60 Gy/30 Fx) or to treatment with identical radio-
therapy plus cisplatin and vinblastine chemother-
apy.
8
After more than 7 years of follow-up, the
median survival time was 13.7 months for combined
modality therapy and 9.6 months for radiotherapy
alone. The 5-year survival rates were 17% and 6%,
respectively (p 0.01).
9
Multiple randomized trials
of radiotherapy alone vs combined chemotherapy
and radiotherapy have included patients with inop-
erable IIIA and IIIB disease. Results of just the IIIB
patients are not independently available. At least 3
meta-analyses have been performed on the random-
ized trials.
1012
The non-small cell lung cancer col-
laborative group meta-analysis evaluated 11 trials
with cisplatin-based chemotherapy regimens.
10
The
results showed a significant overall benefit of che-
moradiotherapy. There was a 13% reduction in the
risk of death (hazard ratio of 0.87) and an absolute
benefit of 4% at 2 years and 2% at 5 years
(p 0.005). Trials using noncisplatin chemotherapy
were marginally, but not statistically, significant in
favor of chemoradiotherapy. Results of two separate
meta-analyses were similarly in favor of combination
chemotherapy and radiotherapy.
11,12
Clinical practice guidelines issued by the Ameri-
can Society of Clinical Oncology in 1997
13
recom-
mend the addition of a platinum-based regimen to
thoracic radiotherapy in patients with unresectable
IIIA/IIIB disease with a good performance score and
minimal weight loss. The National Comprehensive
Cancer Network practice guidelines and the Cancer
Care Ontario Practice Guidelines (No. 7-3) agree
with the American Society of Clinical Oncology
recommendations. Accordingly, patients with unre-
sectable stage IIIB NSCLC who have a good perfor-
mance score and minimal weight loss ( 5%) should
be treated with cisplatin-based chemotherapy in
combination with thoracic radiotherapy. With com-
bined modality therapy, the expected 5-year survival
is 10 to 15%.
Recommendations
3. For patients with stage IIIB disease without
malignant effusions, PS 0 or 1, and minimal
weight loss ( 5%), combined chemoradiother-
apy should be the standard of care. Level of
evidence: good; benefit: substantial; grade of
recommendation: A.
4. In patients with stage IIIB NSCLC and PS 2 or
those with substantial weight loss ( 10%),
combined modality treatment could be used
after careful consideration. Level of evidence:
poor; benefit: moderate; grade of recommen-
dation: C.
Altered Fractions of Radiotherapy
It appears from studies in several epithelial tumor
systems that the clinical effectiveness of radiation is
the total dose per unit time. The clinical advantage of
multiple daily fractions is a reduction in late tissue
damage. In a virulent tumor such as lung cancer, it is
unclear whether substantial clinical benefit can be
derived from radiation optimization.
14
There has been interest in altered fractionation
radiotherapy to increase control of the primary tu-
mor and decrease the toxicity to normal tissue.
Hyperfractionation is defined as the use of two or
more fractions daily of smaller-than-conventional
fraction size. The Radiation Therapy Oncology
Group (RTOG) conducted a three-arm randomized
control trial of standard once daily radiotherapy,
combination chemotherapy, and standard once daily
radiotherapy or hyperfractionated radiotherapy
given twice daily without chemotherapy.
15
Ninety-
five percent of patients had stage IIIA or IIIB
disease. The overall survival was statistically superior
for the combined chemotherapy and once daily
thoracic radiotherapy arm. Survival for patients re-
ceiving twice daily radiotherapy was not statistically
superior. The median survival for the standard radi-
ation was 11.4 months, 13.2 months for the com-
bined modality arm, and 12 months for the hyper-
fractionated irradiation treatment. The 5-year
survival rates for these groups were 5%, 8%, and 6%,
respectively.
Meta-analysis of this trial and two other smaller
trials was performed, and it reported improved
survival for hyperfractionated radiotherapy over
standard radiotherapy (odds ratio 0.69; 95% confi-
dence interval 0.510.95; p 0.02).
16
However, a
second meta-analysis of these same three studies,
conducted at Cancer Care Ontario Practice Guide-
lines Initiation Resource Group and based on 2-year
survival and a fixed effort model, did not demon-
strate a significant benefit of hyperfractionated ra-
diotherapy over standard radiotherapy (odds ratio
0.67; 95% confidence interval 0.421.07; p 0.09).
17
Recently, a phase III trial by the North Central
Cancer Treatment Group evaluated concurrent
combined treatment with etoposide/cisplatin chemo-
therapy (both arms) and once daily or twice daily
split course thoracic radiotherapy. There was no
significant difference in survival or toxicity for the
two arms. The median survival time was 14 vs 15
months, and the 2-year survival rates were 37% and
40%, respectively (p 0.6).
18
Accordingly, there is
absence of convincing data that hyperfractionated
(twice daily) thoracic radiotherapy is superior to
standard once daily radiotherapy.
Accelerated radiotherapy is defined as the use of
two or more fractions of standard fraction size daily
to the same conventional total dose as standard
radiotherapy, but increasing the number of fractions
per week and shortening the overall treatment time.
Hyperfractionated accelerated radiotherapy com-
bines the features of accelerated and hyperfraction-
ated regimen. It uses two or three fractions of
smaller fraction size daily, delivered over a shorter
period of time than conventional therapy. The goal is
to reduce long-term normal tissue toxicity by smaller
fraction size and to reduce repopulation in rapidly
proliferating tumors. One randomized phase III trial
in England
19,20
compared continuous hyperfraction-
ated accelerated radiotherapy (CHART) to standard
radiotherapy (60 Gy/30 Fx). CHART consisted of
three treatments per day (1.5 Gy/Fx), at least 6 h
apart, for 12 days without a break (54 Gy). Sixty-one
percent of patients were IIIA or IIIB. The 1- and
2-year survival rates for the CHART arm were 63%
and 29%, respectively, vs 55% and 20% for standard
radiotherapy. Overall, there was a 22% reduction in
the relative risk of death (p 0.008). Acute esoph-
agitis was more severe for patients receiving
CHART, but the incidence at 3 months was similar
to patients receiving standard radiotherapy, and
there was no difference in late morbidity. Recently,
a phase III trial of HART (same as CHART except
for no treatments on weekends) vs standard radio-
therapy after induction chemotherapy had to be
closed due to poor accrual in the Eastern Coopera-
tive Oncology Group.
Accordingly, at this time we have one phase III
trial of CHART that showed superior survival over
standard radiotherapy, but it had only 61% of pa-
tients with stage IIIA and IIIB disease, and no data
are available concerning the combination of CHART
and chemotherapy. Additionally, the schedule of
radiotherapy 3 times per day seems to have been
rejected by radiotherapists in North America. There-
fore, at this time, neither CHART nor HART can be
recommended as standard therapy. Optional dose,
volume, and fractionation schedules are evolving. An
RTOG randomized phase III trial in the 1980s
21
demonstrated that 60 Gy produced a nonstatistically
significant survival improvement compared with 50
Gy or two different schedules of 40 Gy. Accordingly,
the dose of 60 Gy in 6 weeks has been widely used
for 20 years. However, this dose and fractionation
schedule results in local control rates of 15 to
30%.
15,22
Therefore, higher doses may yield better
results. Guidelines of the American Society of Clin-
ical Oncology advise that definitive dose thoracic
radiotherapy should be at least 60 Gy in 1.8 to 2.0 Gy
fractions.
13
Recommendation
5. For stage IIIB NSCLC patients, there are no
convincing data that hyperfractionated (two or
more fractions daily) radiotherapy is superior to
standard once daily treatment. Continuous
hyperfractionated accelerated radiotherapy
(CHART) was demonstrated in one small trial
to be superior to standard once daily therapy,
but the logistics of three treatments daily have
not proven to be acceptable in a North Amer-
ican trial. No data are available in combining
CHART with chemotherapy. Level of Evi-
dence: poor; benefit: small/weak; grade of rec-
ommendation: I.
Concurrent vs Sequential
Chemoradiotherapy
Concurrent treatment with chemoradiotherapy
has become the standard in treatment of limited
stage small cell lung cancer. There is less information
available on treatment of NSCLC. The West Japan
Lung Cancer Group conducted a randomized phase
III trial of concurrent vs sequential thoracic radio-
therapy in combination with mitomycin, vindesine,
and cisplatin, with over 150 patients participating in
each arm.
23
Seventy-two percent had stage IIIB
disease. Radiation was begun on day 2 at a dose of
28 Gy (2 Gy/Fx 14), followed by a rest of 10 days
and then repeated for a total dose of 56Gy. In the
sequential arm, the same chemotherapy was given,
but radiotherapy was initiated after completing che-
motherapy and consisted of 56 Gy (2 Gy/Fx 28)
without a break. The median survival time was
superior for patients in the concurrent therapy arm
(16.5 vs 13.3 months), and the 5-year survival differ-
ence was 15.8% vs 8.9% (p 0.039). The RTOG
conducted a phase III trial of concurrent vs sequen-
tial chemoradiotherapy. The chemotherapy was vin-
blastine and cisplatin. Radiotherapy was begun on
day 1 of chemotherapy or on day 50 after chemo-
therapy, and the total dose was 63 Gy. The median
survival time was 17 months with concurrent therapy
and 14.6 months with sequential treatment.
24
This
difference was significant at the time of a follow-up
report in the fall of 2000.
A French cooperative group performed a phase
III randomized trial of sequential vs concurrent
chemoradiotherapy in unresectable IIIA/IIIB pa-
tients.
25
The chemotherapy was cisplatin and vi-
norelbine for 3 cycles, followed by thoracic radio-
therapy (66 Gy/33 Fx), or concurrent cisplatin/
etoposide, with thoracic radiotherapy followed by
cisplatin and vinorelbine. Seventy-five percent of
patients had IIIB disease, and over 100 patients
were enrolled in each arm. Incidence of grade 3/4
esophagitis was 26% in patients in the concurrent
therapy arm vs 0% in the sequential arm. The
median survival time was 13.8 months with se-
quential therapy and 15 months with concurrent
therapy. The 2-year survival rates were 23% and
35%, respectively.
25
While there was a trend in
favor of concurrent therapy, it was not statistically
significant at the time of this preliminary report.
Therefore, based on these three large phase III
randomized trials, concurrent chemoradiotherapy
appears to result in better survival than sequential
therapy. It is associated with some increased tox-
icity, mainly acute esophagitis, and should be
reserved for patients with PS 0 or 1 and minimal
weight loss.
Recommendation
6. For stage IIIB NSCLC patients with PS 0 or 1
and minimal weight loss, concurrent therapy
would be recommended. Concurrent chemora-
diotherapy is associated with an increase rate of
acute esophagitis compared to sequential ther-
apy. Concurrent therapy appears to be associ-
ated with improved survival over that of sequen-
tial therapy. Level of evidence: fair; benefit:
substantial; grade of recommendation: B.
1. Patients with clinical T4N0 NSCLC due to
either satellite tumor nodule(s) in the same
lobe or carinal involvement should be evalu-
ated by a thoracic surgeon for possible resec-
tion. Level of evidence: fair; benefit: substan-
tial; grade of recommendation: B.
2. For patients with stage IIIB NSCLC due to T4
(excluding Pancoast tumors) or N3 disease,
neoadjuvant chemotherapy or chemoradio-
therapy followed by surgery has been explored
in limited phase II trials. At the time this
article was written, there were no phase III
trial data available to document that surgery
would improve survival rates, so this approach
should not be considered as standard therapy.
Level of evidence: poor; benefit: small/weak;
grade of recommendation: I.
3. For patients with stage IIIB disease without
malignant effusions, PS 0 or 1, and minimal
weight loss ( 5%), combined chemoradio-
therapy should be the standard of care. Level
of recommendation: A.
4. For patients with stage IIIB NSCLC and PS 2
or those with substantial weight loss ( 10%),
combined modality treatment could be used
after careful consideration. Level of evidence:
poor; benefit: moderate; grade of recommen-
dation: C.
5. For stage IIIB NSCLC patients, there are no
convincing data that hyperfractionated (two or
more fractions daily) radiotherapy is superior
to standard once daily treatment. Continuous
hyperfractionated accelerated radiotherapy
(CHART) was demonstrated in one small trial
to be superior to standard once daily therapy,
but the logistics of three treatments daily have
not proven to be acceptable in a North Amer-
ican trial. No data are available in combining
CHART with chemotherapy. Level of evi-
dence: poor; benefit: small/weak; grade of
recommendation: I.
6. For stage IIIB NSCLC patients with PS 0 or 1
and minimal weight loss, concurrent therapy
would be recommended. Concurrent chemo-
radiotherapy is associated with an increased
rate of acute esophagitis compared to sequen-
tial therapy. Concurrent therapy appears to be
associated with improved survival vs sequential
therapy. Level of evidence: fair; benefit: sub-
stantial; grade of recommendation: B.
References
2 British Thoracic Society, of Cardiothoracic Surgeons of Great
Britain, and Ireland Working Party. Guidelines on the selec-
tion of patients with lung cancer for surgery. Thorax 2001;
56:89108
3 Deslauriers J, Brisson J, Cartier R, et al. Carcinoma of the
lung: evaluation of satellite nodules as a factor influencing
prognosis after resection. Thorac Cardiovasc Surg 1989;
97:504512
4 Urschel JD, Urschel DM, Anderson TM, et al. Prognostic
implications of pulmonary satellite nodules: are the 1997
staging revisions appropriate? Lung Cancer 1998; 21:8387
5 Mathieson DJ, Grillo HC. Crainal resection for bronchogenic
carcinoma. Thorac Cardiovasc Surg 1991; 102:1622
6 Darteville PG, Khalife J, Chapelier A, et al. Tracheal sleeve
pneumonectomy for bronchogenic carcinoma: a report of 55
cases. Ann Thorac Surg 1988; 46:6872
7 Albain KS, Rusch VW, Crowley JJ, et al. Concurrent cisplatin/
etoposide plus chest radiation therapy followed by surgery for
stage IIIA (N2) and IIIB non-small cell lung cancer: mature
results of Southwest Oncology Group Phase II Study 8805.
J Clin Oncol 1995; 13:18801892
8 Dillman RO, Seagren SL, Propert KJ, et al. A randomized
trial of induction chemotherapy plus high-dose radiation vs
radiation alone in stage III non-small cell lung cancer. N Engl
J Med 1990; 323:940945
9 Dillman RO, Herndon J, Seagren SL, et al. Improved survival
in stage III non-small cell lung cancer: seven-year follow-up
of cancer and leukemia group B (CALGB) 8433 Trial. J Natl
Cancer Inst 1996; 88:12101215
10 Non-small cell lung cancer collaborative group. Chemother-
apy in non-small cell lung cancer: a meta-analysis using
updated data on individual patients from 52 randomized
clinical trials. BMJ 1995; 311:899909
11 Pritchard RS, Anthony SP. Chemotherapy plus radiotherapy
compared with radiotherapy alone in the treatment of locally
advanced, unresectable, non-small cell lung cancer: a meta-
analysis. Ann Intern Med 1996; 125:723729
12 Marino P, Preatoni A, Cantoni A. Randomized trials of
radiotherapy alone vs combined chemotherapy and radiother-
apy in stages IIIa and IIIb non-small cell lung cancer: a
meta-analysis. Cancer 1995; 76:593601
guidelines for the treatment of unresectable non-small cell
14 Fu KK, Pajak TF, Trotti A, et al. A radiation therapy oncology
group phase III randomized study to compare hyperfraction-
ation with two variants of accelerated fractionation to stan-
dard fractionation radiotherapy for head and neck squamous
cell carcinomas: First Report of RTOG 9003. Int J Radiat Biol
Phys 2000; 48:716
15 Sause W, Kolesar P, Taylor S IV, et al. Final results of phase
III trial in regionally advanced unresectable non-small cell
lung cancer. Chest 2000; 117:358364
16 Stuschke M, Thames HD. Hyperfractionated radiotherapy of
human tumors: overview of the randomized clinical trials. Int
J Radiat Oncol Biol Phys 1997; 37:259267
17 Cancer Care Ontario Practice Guideline Initiative GPG 712.
Altered fractionation of radical radiation therapy in the
management of unresectable non-small cell lung cancer.
Available at http://hiru.mcmaster.ca/ccopgi/guidelines/lung/
cpg7_12f.html. Accessed December 18, 2002
18 Stella P, Marks R, Schild S, et al. Phase III trial of chemo-
therapy either standard radiotherapy or accelerated hyper-
fractionated thoracic radiotherapy for stage III non-small cell
lung cancer[abstract]. Proc Amer Soc Clin Oncol 2001;
20:312a (1245)
19 Saunders M, Dische S, Barrett A, et al. Continuous hyper-
fractionated accelerated radiotherapy (CHART) vs conven-
tional radiotherapy in non-small cell lung cancer: a random-
ized multicentre trial. Lancet 1997; 350:161165
20 Saunders M, Dische S, Barrett A, et al. Continuous, hyper-
fractionated, accelerated radiotherapy (CHART) vs conven-
tional radiotherapy in non-small cell lung cancer: mature data
from the randomized multicentre trial. Radiother Oncol
1999; 52:137148
21 Perez CA, Stanley K, Rubin P, et al. A prospective random-
ized study of various irradiation doses and fractionation
schedules in the treatment of inoperable non-oat cell carci-
noma of the lung. Cancer 1980; 45:27442753
22 LeChevalier T, Arriagada R, Quoix E, et al. Radiotherapy
alone vs combined chemotherapy and radiotherapy in non-
resectable non-small cell lung cancer: first analysis of a
randomized trial in 353 patients. J Natl Cancer Inst 1991;
83:417423
23 Furuse K, Fukuoka M, Kawahara M, et al. Phase III study of
concurrent vs sequential thoracic radiotherapy in combina-
tion with mitomycin, vindesine, and cisplatin in unresectable
stage III non-small cell lung cancer. J Clin Oncol 1999;
17:26922699
24 Curran WJ Jr, Scott C, Langer C, et al. Phase III comparison
of sequential vs concurrent chemoradiation for patients with
unresectable stage III non-small cell lung cancer: initial
report of the Radiation Therapy Oncology Group 9410
[abstract]. Proc Amer Soc Clin Oncol 2000; 19:484a (1891)
25 Pierre F, Maurice P, Gilles R, et al. A randomized phase III
trial of sequential chemoradiotherapy in locally advanced
non-small cell lung cancer [abstract]. Proc Amer Soc Clin
Oncol 2001; 20:312a
2003;123;221-225 Chest
James R. Jett, Walter J. Scott, M. Patricia Rivera and William T. Sause
Guidelines on Treatment of Stage IIIB Non-small Cell Lung Cancer
& Services
Updated Information
S
References
S#BIBL
free at:
Reprints
2003;123;226-243 Chest
Mark A. Socinski, David E. Morris, Gregory A. Masters and Rogerio Lilenbaum
Chemotherapeutic Management of Stage IV Non-small Cell Lung Cancer
ISSN: 0012-3692.
Chemotherapeutic Management of
Stage IV Non-small Cell Lung Cancer*
Mark A. Socinski, MD, FCCP; David E. Morris, MD;
Gregory A. Masters, MD, FCCP; and Rogerio Lilenbaum, MD
Stage IV non-small cell lung cancer (NSCLC) denotes the presence of metastatic disease and is
largely incurable using present-day therapies. Chemotherapy remains a therapeutic option in this
patient population, and there are many pertinent issues surrounding its use in patients with stage
IV NSCLC. Eleven questions were framed by the American College of Chest Physicians Lung
Cancer Guidelines Committee, and these were addressed by a systematic search of the available
literature. The issues addressed included the identification of prognostic factors in selecting
patients for chemotherapy and a critical analysis of the survival benefit provided by chemother-
apy. Given the development of several new chemotherapy agents over the past decade, the
impact that these agents have made was addressed as well as the definition of a standard of care
regarding chemotherapeutic regimens. Given the fact that chemotherapy does not represent a
curative option, other issues addressed were the optimal duration of treatment as well as its
impact on symptom relief and quality of life, the role of second-line therapy, and the outcomes
expectations from both first-line and second-line chemotherapy. The question of what specialty
delivered the chemotherapy also was addressed. Once the data were identified, a critical analysis
was undertaken attempting to objectively portray the data in support of answers for each of the
questions posed. We believe the data support the fact that properly selected patients benefit from
chemotherapy with regard to survival and palliation in both first-line and second-line settings. It
appears that in trials addressing the duration of first-line therapy, this survival and palliative
benefit occurs early, and prolonged therapy is not indicated. Therapy in this setting is
cost-effective, and there are several regimens that can be considered to be standard-of-care
options. Physicians involved in the diagnosis of these patients should be aware of the potential
benefits of chemotherapy, allowing them to give recommendations to patients that are based on
data derived from clinical trials. In addition, this awareness will allow them to make referrals,
when appropriate, to physicians who are trained in the administration of chemotherapy and the
management of patients undergoing such therapy. (CHEST 2003; 123:226S243S)
Key words: chemotherapy; evidence-based medicine; guidelines; non-small cell lung cancer
Abbreviations: BSC best supportive care; CI confidence interval; ECOG Eastern Cooperative Oncology
Group; ELVIS Elderly Lung Cancer Vinorelbine Italian Study; EORTC European Organization for Research and
Treatment of Cancer; FACT-L functional assessment of cancer therapy-lung cancer; HR hazard ratio;
NSCLC non-small cell lung cancer; OR odds ratio; PS performance status; QOL quality of life; RR relative
risk; TOI trial outcome index; TRT thoracic radiation therapy; V/I vinorelbine or ifosfamide
L
ung cancer remains the leading cause of cancer-
related death in the United States.
1
Approxi-
mately 169,400 new cases will be diagnosed in the
year 2002 with an estimated 154,900 deaths expected
from this disease.
1
The vast majority of these patients
have non-small cell lung cancer (NSCLC), which
comprises approximately 80% of all new cases.
2
In
early-stage NSCLC, surgical resection remains the
standard of care in fit patients. Ongoing trials in this
setting are addressing the role of both adjuvant and
neoadjuvant chemotherapy. In fit patients with un-
resectable, locally advanced, stage III NSCLC, che-
motherapy in combination with thoracic radiation
therapy (TRT) is the standard of care. Controversy
exists regarding the optimal strategy in combining
these two modalities (ie, sequential therapy, concur-
rent therapy, or both) as well as regarding the
optimal chemotherapeutic agents and optimal dose
*From the Multidisciplinary Thoracic Oncology Program (Drs.
Socinski and Morris), Lineberger Comprehensive Cancer Cen-
ter, University of North Carolina, Chapel Hill, NC; Northwestern
University Medical School (Dr. Masters), Evanston Northwest-
ern Healthcare, Evanston, IL; and the University of Miami
School of Medicine (Dr. Lilenbaum), Mount Sinai Comprehen-
sive Cancer Center, Miami Beach, FL.
Correspondence to: Mark A. Socinski, MD, FCCP, Director,
Multidisciplinary, Thoracic Oncology Program, CB No. 7305,
University of North Carolina, Chapel Hill, NC 27599; e-mail:
socinski@med.unc.edu
and schedule of TRT. In these two groups of pa-
tients, cure remains the goal in fit patients, and local
modalities (ie, surgery and TRT) remain a corner-
stone of therapy.
Stage IV NSCLC denotes the presence of meta-
static disease. The more common sites of metastatic
disease include the liver, bones, adrenal, brain, and
contralateral lung. In the 1997 revision of the staging
system,
3
a separate tumor nodule in the ipsilateral
nonprimary tumor lobes also was classified as meta-
static disease, although the possibility of multiple
primary lung cancers should be considered in se-
lected cases. In addition, there are subsets of pa-
tients with stage IIIB disease who are not appropri-
ate for combined therapeutic modality approaches,
such as those patients with a malignant pleural/
pericardial effusion and certain patients with ad-
vanced, palpable supraclavicular adenopathy. These
patients are typically included with stage IV patients
when the benefit of systemic chemotherapy is con-
sidered and have been included in most clinical trials
addressing this issue, as the prognosis of these
patients is similar to that of patients with stage IV
NSCLC. Where possible, we have clearly indicated
the percentage of stage IV patients in each of the
trials included in this analysis.
The purpose of this section is to review the
evidence supporting the role of systemic chemother-
apy in the management of stage IV NSCLC. As
noted above, most trials have included selected
patients with stage IIIB NSCLC in whom a systemic
therapeutic approach is appropriate. The 5-year
survival rate of this group of patients is 1%, and
therefore these patients are generally considered to
be incurable. Despite this, the important issues to
address include which patients are appropriate for
chemotherapy, the survival and palliative impact of
chemotherapy, the optimal chemotherapeutic ap-
proach, and its toxicity and outcomes expectations.
To accomplish this, we sought to identify the evi-
dence addressing these issues from primary data
sources that have been published in the existing
English literature. Eleven questions regarding the
role of chemotherapy in treating patients with stage
IV NSCLC were framed by the American College of
Chest Physicians Lung Cancer Guidelines Commit-
tee. A list of MEDLINE search terms that were used
to identify the primary evidence addressing these 11
questions is shown in the Appendix. In addition, the
primary evidence was supplemented by the authors
if data sources were identified outside the MED-
LINE search mechanism. Once appropriate data
sources were identified, a comprehensive review was
undertaken. The evidence addressing each question
is presented in detail followed by a summary state-
ment. All attempts to eliminate bias were made by
objective presentation of the evidence as it exists in
the primary data source.
Are There Identifiable Prognostic Factors
That Should Be Used When Selecting
Patients for Systemic Chemotherapy?
The prognosis of patients with advanced NSCLC
is poor. Most large phase III trials have shown a
median survival time of 8 to 10 months and a 1-year
survival rate of 30 to 35%.
2
Given the consistent
improvement in the survival of patients who have
been treated with chemotherapy over those receiv-
ing supportive care alone, clinicians struggle to strat-
ify these patients into different prognostic groups.
One would like to identify those patients who are the
most likely to benefit from aggressive chemotherapy
and to tolerate its side effects, and to identify another
group of patients who are unlikely to obtain any
meaningful advantage from such therapy. Ideally,
prognostic groups could help to stratify patients in
order to apply different approaches or levels of
aggression. This also could allow an appropriate
focus on quality of life (QOL) as a major end point.
Individual patient characteristics seem to influ-
ence survival in patients with advanced NSCLC. The
most important factor across all studies is perfor-
mance status (PS). Patients with stage IV NSCLC
who are compromised by their disease have poorer
survival compared to patients who are less compro-
mised. Two commonly used PS scales are shown in
Table 1. At least 10 trials evaluating prognostic
factors in patients with advanced NSCLC have
clearly identified PS at the time of diagnosis to be a
powerful predictor of survival.
413
In a landmark
analysis of 893 patients with stage IV NSCLC,
Finkelstein et al
7
documented the impact that PS
had on survival. In that study, the 1-year survival rate
was 36% for PS level 0 patients, 16% for PS level 1
patients, and 9% for PS level 2 patients (p 0.001).
Pretreatment weight loss is generally regarded as a
negative prognostic factor,
4,5,7,9
but not all trials have
reliably corroborated this.
6,8,10,13
The sex of NSCLC
patients also has been described as an important
prognostic variable, with most trials suggesting im-
proved survival for women.
68,12,13
Differences in
survival based on patient sex have generally been
small, usually 1 to 2 months.
Age may be a predictor of survival, with some
studies suggesting that elderly patients with ad-
vanced NSCLC have poor outcomes.
4,10,13
Other
studies looking at this variable have failed to confirm
this or have suggested that older patients have a
similar or even superior survival.
69,11,12
In a retro-
spective review,
14
no difference in survival was seen
for patients 40 years of age compared with a
matched group of patients who were 50 years of
age. Therefore, it is difficult to say that age is a
reliable independent prognostic factor for these pa-
tients.
A large Eastern Cooperative Oncology Group
(ECOG) trial
15
in the United States (ECOG 5592)
showed that QOL scores, as measured by the func-
tional assessment of cancer therapy-lung cancer
(FACT-L), were important predictive factors. This
study also showed that the absence of change in
cough or hoarseness, the absence of bone pain, the
absence of other symptoms from metastases, and the
absence of anorexia were all independent favorable
prognostic factors. Another retrospective study
16
showed pain level to correlate inversely with survival.
QOL, however, was not a predictive factor in this
analysis.
Pretreatment stage, even among those patients
with advanced NSCLC is prognostic, with stage IIIB
patients generally having a better survival rate than
those with overt metastatic or stage IV disease.
3,17
The total number of metastatic sites may influence
the prognosis.
410,12,13
Several nonrandomized trials
6
have suggested that patients with a solitary site of
metastasis may have superior outcomes and that
more aggressive therapy (including surgical resection
of the primary tumor and metastatic site) may
provide up to 20 to 30% of patients with long-term
survival. Studies also have suggested that specific
sites of metastatic disease may change the prognosis
in patients with advanced NSCLC. Specifically, pa-
tients with disease confined to the lungs may have
superior outcomes. Those with brain metastases may
have poorer outcomes,
4
but this conclusion is con-
troversial.
18,19
The presence of bone or liver metas-
tases has been found to confer a poor prognosis in
the retrospective analysis mentioned previously.
15
Histologic subtype does not reliably provide prog-
nostic importance in patients with advanced NSCLC,
despite the different clinical manifestations of adeno-
carcinoma compared to squamous histology.
46,10,11
Normal levels of serum lactate dehydrogenase, high
levels of albumin, and low levels of alkaline phospha-
tase all have, however, been associated with a better
prognosis in patients with advanced NSCLC.
6,13,20
The expression of neuroendocrine markers may
predict survival in patients with NSCLC. In one
study,
21
responding patients with two or more posi-
tive markers survived longer. Another study
22
found
that tumors containing 50% positively staining
cells were associated with shorter survival times.
Currently, neuroendocrine markers cannot be used
to reliably predict survival in patients with advanced
NSCLC.
Perhaps the most important prognostic factor, and
the one most clearly proven in randomized trials, is
whether patients receive chemotherapy.
2332
Nu-
merous randomized trials and several meta-analyses
have confirmed an improvement in the median
survival time of 6 to 8 weeks, translating to a 10%
improvement in the 1-year survival rate for those
advanced NSCLC patients who are receiving plati-
num-based chemotherapy.
Recommendation 1
1. When selecting patients for systemic chemo-
therapy, PS at the time of diagnosis should be
used because it is a consistent prognostic factor
for survival. Patients with a PS of ECOG level
0 or 1 should be offered chemotherapy. Level
recommendation, A
2. Data are not yet sufficient to routinely recom-
mend chemotherapy to patients with a PS of
ECOG level 2. Level of evidence, poor; benefit,
small/weak; grade of recommendation, I
3. Patients with a PS of ECOG level 3 or 4 should
not receive chemotherapy. Level of evidence,
fair; benefit, moderate; grade of recommenda-
tion, B
4. Other patient-related factors (eg, gender, age,
sites of metastases, and histology) have not
been consistent prognostic factors for survival.
Level of evidence, poor; benefit, small/weak;
Table 1Performance Status Scales
Status Scales Scale
ECOG
Normal 0
Fatigue without significant decrease in activity 1
Fatigue with significant impairment of daily activities
or bedrest 50% of waking hours
2
Bedrest 50% of waking hours 3
Bedridden or unable to care for self 4
Karnofsky
Normal, no complaints 100
Normal activity, minor signs of symptom disease 90
Normal activity with effort 80
Cares for self in daily activities, but unable to carry
on normal activity or work
70
Requires occasional assistance but able to care for
most needs
60
Requires considerable assistance and frequent
medical care
50
Disabled; requires special care and assistance 40
Severely disabled 30
Very sick; hospitalization necessary; active supportive
treatment necessary
20
Moribund 10
Dead 0
What Is the Evidence That Platinum-Based
Chemotherapy Improves Survival?
Ten randomized clinical trials have been pub-
lished
2332
comparing platinum-based chemotherapy
to best supportive care (BSC) [Table 2]. It should be
noted that BSC in these trials included aggressive
symptom management (eg, antitussive agents, sup-
plemental oxygen, and nonnarcotic and narcotic
analgesic agents) as well as palliative radiotherapy
when indicated. In all 10 trials, the median survival
time of the treated patients was numerically superior
to that of patients receiving BSC. The median
survival time of patients receiving BSC was 3.6
months (range, 2.4 to 4.9 months) in these 10 trials,
providing a benchmark for survival in patients with
untreated, advanced NSCLC. The median survival
time of the treated patients was 6.5 months (range,
4.7 to 8.5 months). The numeric survival advantage
seen in all 10 trials was statistically significant in 6 of
these trials.
Four meta-analyses
3336
have been published ex-
amining the effect of treatment vs BSC in patients
with advanced NSCLC. The studies have differed in
how trials were selected for review, in the number of
trials included, in the use of group or individual
patient data in the analysis, and in the statistical
methodology used (Table 3). Despite this, these four
meta-analyses were consistent in their conclusions.
The majority of trials included in these four studies
used cisplatin-based regimens. Souquet et al
33
in-
cluded seven trials, of which six were cisplatin-based.
The end point analyzed was the number of deaths at
3-month intervals up to 18 months. There was a
significant reduction in mortality for up to 6 months
for chemotherapy vs BSC. The odds ratio (OR) at 6
months was approximately 0.6 in favor of chemother-
apy. Grilli et al
34
analyzed six trials (of which five
were cisplatin-based) in terms of relative risk (RR).
The overall RR for death for chemotherapy vs BSC
0.87). Marino et al
35
included eight trials (of which
six were cisplatin-based) and determined the indi-
vidual and pooled odd ratios for death at 6 months.
The OR for death was 0.44 (95% CI, 0.32 to 0.50) in
favor of chemotherapy. The Non-Small Cell Lung
Cancer Collaborative Group used individual patient
data from 11 randomized trials (of which 8 were
cisplatin-based).
36
The other three trials used long-
term alkylating agents or vinca alkaloids and etopo-
side. Although a survival advantage for chemother-
apy was documented over BSC, there was a
difference between cisplatin-based trials and non-
cisplatin-based trials. In fact, the long-term admin-
istration of alkylating agents actually had a negative
impact on survival compared to BSC. When evalu-
ating trials employing only cisplatin-based regimens,
there was a 27% reduction in the risk of death
(hazard ratio, 0.73; 95% CI, 0.63 to 0.85; p 0.0001)
with chemotherapy vs BSC (Fig l).
Table 4 summarizes the recommendations from
several guidelines addressing this issue. As can be
seen, the organizations represent national as well as
Table 2Randomized Clinical Trials of Platinum-Based Chemotherapy vs BSC in Advanced NSCLC*
Study/Year
Patients,
No. Age, yr
PS
Male,
%
Stage
IV,
%
Chemotherapy
Regimen
Survival
Time,
mo
p Value T BSC T BSC T BSC
Rapp et al
23
/1988 198 53 56 58 58 74 86 CAP 5.7 3.9 0.05
(54% 60 yr) VdP 7.5 3.9 0.01
Ganz et al
24
/1989 22 6 (75% 65 yr) 85 90 100 VdP 4.7 3.1 NS
Woods et al
25
/1990 97 91 61 61 73 82 66 VdP 6.2 3.9 NS
Quoix et al
26
/1991 24 22 69 (3674) 62 (5273) 72 93 100 VdP 6.5 2.4 0.001
Cellerino et al
27
/1991 62 61 59 (3370) 62 (4572) 63 97 59 CErP43Mx
ECC
7.9 4.9 NS
Kaasa et al
28
/1991 44 43 62 62 49 100 EP 5.0 3.7 NS
Cartei et al
29
/1993 52 50 56 (3973) 57 (3971) 72 MCP 8.5 4.0 0.0001
Helsing et al
30
/1998 22 26 61 (3672) 65 (4478) 65 90 EC 6.7 2.6 0.003
Thongprasert et al
31
/1999 189 98 58 (3673) 60 (2873) NR NR NR IErP/MVbP 6.0 2.5 0.006
Cullen et al
32
/1999 175 176 62 (5668) 64 (5869) 62 72 NR MIP 6.7 4.8 0.03
*T treated; CAP cyclophosphamide, adriamycin, cisplatin; VdP vindesine, cisplatin; CErP cyclophosphamide, epirubicin, cisplatin;
MxECC methothexate, etoposide, lomustine; EP etoposide, cisplatin; MCP mitomycin, cyclophosphamide, cisplatin; EC etoposide,
carboplatin; IErP ifosfamide, epirubicin, cisplatin; MVbP mitomycin, vinblastine, cisplatin; MIP mitomycin, ifosfamide, cisplatin;
NS not significant; NR not reported.
Values given as median (range).
Values given as median.
Percentage of patients with either ECOG PS 01 or Karnofsky PS 70.
international societies from the United States, Can-
ada, and Europe. The guidelines are either evidence-
based or consensus guidelines and represent an
analysis of the cisplatin-based literature. All of these
organizations recommend chemotherapy in patients
with advanced stage IV NSCLC.
It is difficult to portray to patients the magnitude
of the survival effect in this setting. Table 2 shows
the differences in median survival times in trials
comparing platinum-based regimens to BSC. Stage
IV NSCLC still is a disease with a steeply downslop-
ing survival curve, making median survival a less
meaningful end point. In general, investigators in
this area have focused on the 1-year survival percent-
age. To evaluate this end point, the last four tri-
als
31,32,37,38
published in which the 1-year survival
percentage was reported and that included a BSC
control were analyzed. Fig 2 shows the percentage of
patients who are alive at 1 year as well as the absolute
number of patients who are alive in the United
States, comparing BSC to chemotherapy. The data
suggest a doubling of the 1-year survival rate and
approximately 10,000 more patients surviving be-
yond 1 year as a result of chemotherapy.
In conclusion, the evidence from both randomized
clinical trials and four separate meta-analyses sup-
port the fact that platinum-based chemotherapy
improves survival in patients with advanced stage IV
NSCLC.
Recommendation 2
Patients with a good PS (ie, ECOG level 0 or 1)
should be considered for a platinum-based chemo-
therapy regimen based on the survival advantage
provided over BSC. Level of evidence, good; benefit,
Do New Agents Improve Survival as
Single Agents Compared to BSC?
Since 1990, several new agents with significant
single-agent activity in NSCLC have been developed
including paclitaxel, docetaxel, vinorelbine, gemcit-
abine, and irinotecan. These agents are commonly
referred to as third-generation agents. Several ran-
domized trials have been reported in which these
new agents were tested against BSC using survival as
the primary end point (Table 5). Both taxanes (ie,
paclitaxel and docetaxel) have been compared to
BSC in a randomized trial.
37,39
Survival was superior
for patients who received both taxanes compared to
BSC (Table 5). The magnitude of the benefit mea-
sured either as an increase in median survival or as
1-year survival is similar to the benefit seen with
cisplatin-based chemotherapy regimens. Vinorelbine
has been compared to BSC in the elderly (defined as
persons 70 years of age) in the Elderly Lung
Cancer Vinorelbine Italian Study (ELVIS).
38
In this
select population of patients, a survival advantage
was seen with the estimated relative hazard of death
for patients receiving vinorelbine compared to BSC
being 0.65 (95% CI, 0.45 to 0.93). In addition to
Table 3Meta-analyses of Randomized Trials Comparing Chemotherapy vs BSC in Advanced NSCLC
Study/Year
Trials/Platinum-
Based Trials,
No.
Patients,
No. End Point Conclusion
Souquet et al
33
/1993 7/6 706 No. of deaths at 3, 6, 9,
12, and 18 mo
Statistically significant reduction of mortality for
up to 6 mo
Grilli et al
34
/1993 6/5 635 RR of death at 3, 6, 9,
and 12 mo
24% (95% CI, 1334%) reduction in probability
of death for chemotherapy vs BSC
Marino et al
35
/1994 8/6 712 OR of death at 6 mo OR 0.44 (95% CI, 0.320.50; p 0.05)
favored chemotherapy
NSCLC Collaborative
Group
36
/1995
11/8 1190 HR in death For cisplatin-based chemotherapy HR 0.73
(95% CI, 0.630.85; p 0.0001) vs BSC
Figure 1. Survival in trials employing cisplatin-based regimens
vs BSC only. Reprinted with permission from the Non-Small Cell
Lung Cancer Collaborative Group.
36
these three studies, further evidence of an impact of
the new agents comes from the randomized trial by
Crawford and colleagues.
40
In that study, vinorelbine
was compared to fluorouracil and leucovorin rather
than to BSC. The selection of fluorouracil and
leucovorin as the control arm of the study was
arbitrary, and this combination had not been tested
previously in patients with advanced NSCLC. The
response rate to fluorouracil and leucovorin was only
3%, the median survival time was 5.1 months, and
the 1-year survival rate was 16%. These survival
outcomes are almost identical to the BSC arms of the
other three trials shown in Table 5, suggesting that
fluorouracil and leucovorin had no impact on the
natural course of advanced NSCLC. The outcomes
associated with vinorelbine on the study were a
response rate of 12%, a median survival time of 7.0
months, and a 1-year survival rate of 25% (p 0.03
[compared to fluorouracil and leucovorin]). Thus,
this trial suggested a benefit to patients receiving
vinorelbine that was similar to that in the ELVIS.
38
This study is not included in Table 5 only because
patients in the control arm of the study received
ineffective chemotherapy rather than BSC. An-
other trial
41
compared the use of single-agent gem-
citabine to BSC but was designed to show a palliative
benefit. Since survival was not the end point, this
trial also is not included in Table 5.
The magnitude of the survival benefit from these
single agents is similar to that obtained with
platinum-based regimens. Whether the new active
single-agents yield survival outcomes that are similar
to the platinum-based regimens is not known. Sev-
eral ongoing randomized clinical trials are addressing
this issue.
Recommendation 3
Although the new agents demonstrate improved
survival compared to BSC (level of evidence, good;
benefit, moderate; grade of recommendation, B) in
elderly as well as nonelderly patients with advanced
NSCLC, the data are not yet sufficient to compare
the new single agents to platinum-based combina-
tions. Level of evidence, poor; benefit, small/weak;
Do the New Agents in Combination With
the Platinum-Based Agents Improve
Survival Over Second-Generation
Platinum-Based Regimens?
As noted above, a number of new chemotherapy
agents (ie, paclitaxel, docetaxel, gemcitabine, vinorel-
bine, and irinotecan) have been identified over the
last 10 years as having documented activity in pa-
tients with advanced NSCLC. These third-genera-
Figure 2. Estimated 1-year survival rates and absolute no. of
patients who are alive at 1 year in the United States.
31,32,37,39.
CT chemotherapy. Numbers on the right abscissa represent
thousands.
Table 4Summary of Recommendations From Other Guidelines in Stage IV NSCLC Patients*
Organization
Year of
Publication Recommendation
Grade of
Recommendation
American Society of Clinical Oncology 1997 Chemotherapy is recommended for good PS (01)
and possibly PS 2
A
British Thoracic Society 1998 Patients with NSCLC referred to pulmonologist or
oncologist for chemotherapy
C
National Cancer Guidance Group,
NHS Executive
1998 In advanced disease, platinum-based chemotherapy
indicated
A
ACCC 2000 Chemotherapy for PS 02 NSCLC patients C
LopezCCOPGI 2000 Chemotherapy indicated for survival and symptom
control/QOL
A
NCCN 2000 Chemotherapy indicated for PS 02 NSCLC patients C
PDQ Adult Treatment Editorial Board 2001 Chemotherapy indicated in stage IV NSCLC C
*C consensus guidelines; ACCC Association of Community Cancer Centers; CCOPGI Cancer Care Ontario Practice Guidelines Initiative;
NCCN National Comprehensive Cancer Network.
tion agents have been incorporated into clinical trials
and have been reported to have an improved toxicity
profile, but is there proof that these new drugs also
improve survival compared to older standard ther-
apies?
The first of the new drugs to be studied in
randomized trials was vinorelbine. A large French
trial
42
compared the European standard regimen of
cisplatin and vindesine to vinorelbine alone or vi-
norelbine in combination with cisplatin. This study
showed a significant improvement in patient survival
for the new drug combination of cisplatin/vinorel-
bine, with a median survival of 40 weeks vs 32 weeks
for cisplatin-vindesine. Survival for patients receiving
vinorelbine alone was not significantly different from
that for patients in the standard treatment arm. A
Southwest Oncology Group trial
43
showed an advan-
tage for patients receiving cisplatin with vinorelbine
over those receiving cisplatin alone. The median
survival time for the patients receiving cisplatin and
vinorelbine was 8 months compared to 6 months for
those receiving cisplatin alone.
Other new drugs have been studied in randomized
trials as well. A trial from ECOG showed improved
survival for patients receiving cisplatin and paclitaxel
(at lower or higher doses with granulocyte colony-
stimulating factor support) over the older combina-
tion of cisplatin and etoposide.
44
In this study,
patients receiving the combination of cisplatin and
paclitaxel showed a median survival time of 9.6 and
10 months, respectively, for the lower and higher
doses, compared to 7.7 months for the cisplatin and
etoposide combination. A trial by Belani and col-
leagues
45
failed to show a survival advantage for
patients receiving carboplatin and paclitaxel over
those receiving cisplatin and etoposide. This trial did
slow some of the momentum for the new drug
combinations but did not dampen the enthusiasm for
these agents, since the study suggested an improved
QOL for patients receiving paclitaxel and carboplatin
over the older combination of cisplatin and etopo-
side. A trial
46
comparing patients receiving cisplatin
alone to those receiving cisplatin plus paclitaxel
failed to show a survival advantage. Another trial
47
showed equivalent survival in a trial comparing
patients receiving cisplatin and paclitaxel vs those
receiving cisplatin plus teniposide.
Gemcitabine is another of the new agents studied
in randomized trials. A trial
48
showed a survival
advantage for patients receiving cisplatin and gem-
citabine over those receiving cisplatin alone. This
study showed a median survival time of 9.1 months
for patients receiving the two-drug combination
compared to 7.6 months for patients receiving the
older drug alone. No survival advantage was shown
for patients receiving the combination of cisplatin
and gemcitabine compared to those receiving mito-
mycin, ifosfamide, and cisplatin or cisplatin and
etoposide in two trials from Europe.
49,50
Two tri-
als
51,52
showed equivalent survival for patients re-
ceiving gemcitabine as a single agent to those receiv-
ing cisplatin and etoposide. Both studies also found
significantly less toxicity for the new drug compared
to the older combination.
Two trials from Japan
53,54
have compared a stan-
dard regimen of cisplatin and vindesine to the new
combination of cisplatin and irinotecan. One such
study found a median survival time of 52 weeks for
patients receiving the new regimen vs 47 weeks for
patients receiving the older standard (difference not
significant), with a trend in the 1-year survival rate
favoring the newer regimen at 49% compared to
40%. The other trial
55
did not show significant
differences between the two arms. A subset analysis
including only stage IV patients suggested a signifi-
cant survival advantage for patients receiving cispla-
tin/irinotecan compared to those receiving cisplatin/
vindesine.
Despite these positive trial results, over the last 20
years there still has been evidence that the progress
is slow in improving the treatment for patients with
advanced NSCLC. A retrospective study
56
from the
Dana Farber Cancer Institute looked at outcomes in
North American randomized chemotherapy trials
Table 5New Single Agents vs BSC in Advanced NSCLC*
Survival
Study/Year
Agent
Tested
Patients,
No. Age, yr
PS
01,
%
Male,
%
Stage
IV, %
Response
Rate, %
Median
Duration,
mo
1 Year
Rate, %
p Value T BSC T BSC T BSC T BSC
Ranson et al
39
/2000 Paclitaxel 79 78 65 (3778) 64 (2382) 82 75 55 16 6.8 4.8 NR NR 0.037
Roszkowski et al
37
/2000 Docetaxel 137 70 59 (3675) 60 (3174) 77 84 53 13.1 6.0 5.7 25 16 0.026
ELVIS Group
38
/1999 Vinorelbine 76 78 74 (7086) 74 (7085) 76 87 73 19.7 6.5 4.9 32 14 0.03
*See Table 2 for abbreviations not used in the text.
Values given as median (range).
and found slow progress in improving survival in the
studies they reviewed. Of 33 randomized phase III
trials of chemotherapy for advanced NSCLC be-
tween 1973 and 1994, only 5 showed a significant
difference in survival. There was a median prolon-
gation of survival time of 2 months in these positive
trial results. It is important to note, however, that
this review omitted most of the trials with the newer
drugs, which were introduced in the early 1990s.
Baggstrom and colleagues
57
have performed a meta-
analysis of the published literature comparing plati-
num-based regimens including a third-generation
agent to older standard platinum-based regimens.
Eight trials published since 1994 were identified,
which included 3,296 patients. In an analysis of
heterogeneity, the results of the trials were thought
to be consistent, allowing a summary analysis. When
examining the impact on 1-year survival, the new
third-generation regimens increased patient survival
compared to the older regimens (RR, 1.14; 95% CI,
1.01 to 1.29). There was an absolute increase in the
1-year survival rate of 4% using the newer combina-
tion regimens compared to the older regimens
(p 0.04). Also, patient response rates were im-
proved with the newer regimens (RR, 1.80; 95% CI,
1.51 to 2.15) with an absolute increase of 13%. There
was no difference in the rates of treatment-related
deaths comparing the newer regimens to the older
platinum-based regimens. This analysis suggests that
there has been a significant, albeit small, improve-
ment in survival with the use of the newer third-
generation regimens compared to the older standard
regimens.
In conclusion, the data suggest that progress is
being made in the management of patients with
advanced NSCLC through newer and more effective
chemotherapy. The magnitude of improvement,
however, is small. This emphasizes the importance of
continuing active clinical research into new treat-
ments. We need to identify more effective treatment
options that have less toxicity, including targeted and
biological agents, for this disease.
Recommendation 4
Combination chemotherapy regimens incorporat-
ing the new single agents with a platinum-based
agent should be considered the standard of care.
Level of evidence, fair; benefit, moderate; grade of
recommendation, B
Is There a Standard of Care Regarding
the Choice of Chemotherapy in the
First-Line Setting?
Two large randomized trials that were reported
within the last 2 years compared several of the
new-generation regimens in the treatment of
patients with advanced NSCLC. The first trial,
58
conducted by the Southwest Oncology Group, com-
pared the use of cisplatin-vinorelbine with carbopla-
tin-paclitaxel. There was no difference in objective
response, median survival time, or 1-year survival
rates between patients receiving the two combina-
tions. The second trial,
59
by ECOG, compared cis-
platin-paclitaxel as the standard regimen to the
following three new combinations: cisplatin-gemcit-
abine; cisplatin-docetaxel; and carboplatin-paclitaxel.
No significant difference in response rates or survival
was observed among the four arms. There was a
small, but statistically significant, difference in
progression-free survival in favor of patients receiv-
ing the cisplatin-gemcitabine combination. Both of
these trials also included a comprehensive analysis of
toxicity. In the first trial, the combination of cispla-
tin-vinorelbine produced more nausea and hemato-
logic toxicity. Carboplatin-paclitaxel produced more
alopecia and peripheral neuropathy. A formal QOL
analysis was undertaken and demonstrated compa-
rable parameters between the two regimens. In the
second trial, the combination of cisplatin-gemcitab-
ine caused more thrombocytopenia, whereas the
cisplatin-docetaxel combination produced more neu-
tropenia. Overall, patients receiving carboplatin-
paclitaxel had the lowest incidence of life-threaten-
ing or lethal toxicities. Platinum-based combination
regimens that have been tested and reported in
published phase III trials are shown in Table 6.
Given the number of active agents available, non-
platinum-containing combination regimens have
been evaluated in many phase I and II trials. A
randomized trial
60
comparing patients receiving the
combination of cisplatin-docetaxel to those receiving
gemcitabine-docetaxel has been published. There
Table 6Platinum-Based Combination Regimens
Tested in Published Phase III Trial and Considered
Standard of Care*
Regimen Doses Schedule References
Cisplatin 7580 mg/m
2
day 1 44, 47
Paclitaxel 135 mg/m
2
/24 day 1
or
175 mg/m
2
/3 h day 1
Every 21 d
Carboplatin AUC 6 day 1 45, 58, 59
Paclitaxel 225 mg/m
2
day 1 Every 21 d
Cisplatin 100 mg/m
2
day 1 42, 43, 58
Vinorelbine 25 mg/m
2
/wk Every 28 d
Cisplatin 100 mg/m
2
day 1 48, 49, 59
Gemcitabine 1,000 mg/m
2
/wk Every 28 d
Cisplatin 75 mg/m
2
day 1 59
Docetaxel 75 mg/m
2
day 1 Every 21 d
*AUC area under the concentration curve.
was no significant difference in response or survival
rates. The toxicity profile was slightly more favorable
for patients receiving gemcitabine-docetaxel, with
less neutropenia, nausea, and diarrhea. No other trial
that has been published to date addresses the issue
of nonplatinum doublets in stage IV NSCLC.
Recommendation 5
No one regimen has been demonstrated to be
superior in the first-line therapy for patients with
advanced NSCLC. A cisplatin-based or carboplatin-
based combination regimen that includes one of the
new agents remains the standard of care for first-line
therapy in patients with stage IV NSCLC. Level of
evidence, good; benefit, substantial; grade of recom-
mendation, A
Is There an Optimal Duration of
Chemotherapy?
Given the noncurative nature of stage IV NSCLC,
the duration of chemotherapy must be weighed
against the toxicity it engenders. Until recently, few
trials addressed this issue, and chemotherapy would
be administered for six or more cycles. In the 1997
guideline issued by the American Society of Clinical
Oncology,
61
the lack of data pertaining to this issue
was cited. The consensus of the expert panel was that
chemotherapy should be administered for no more
than eight cycles in patients with stage IV NSCLC.
The only trial referenced in that guideline was
reported in 1989 by Buccheri and colleagues.
62
This
controlled trial used a non-cisplatin-based chemo-
therapy regimen and randomized patients with sta-
ble disease after two or three cycles of chemotherapy
to continued treatment or BSC. No survival benefit
was seen with continuous treatment in this subset of
patients with stable disease.
In 2001, Smith and colleagues
63
reported a ran-
domized trial of three cycles vs six cycles of mitomy-
cin, vinblastine, and cisplatin in 308 patients with
advanced NSCLC, of whom 54% had stage IV
disease. Of the patients randomized to three cycles,
72% completed therapy, while only 31% of the
patients randomized to six cycles completed therapy.
The median survival times and 1-year survival rates
were similar for patients in both study arms, with no
advantage seen for the longer duration of therapy. In
addition, the median duration of symptom relief was
similar in both study arms. QOL parameters were
the same or were improved in patients receiving
three cycles of therapy as less fatigue, nausea, and
vomiting (typical cumulative toxicities of cisplatin)
was reported compared to patients receiving the
longer duration (six cycles) of therapy.
Another trial
64
randomized 230 patients with stage
IIIB/IV NSCLC to either four cycles of carboplatin-
paclitaxel or to continuous treatment until they
experienced objective progression of the disease.
Survival and QOL were the primary end points of
this trial. Of interest is the fact that in the continu-
ous-treatment arm of that trial, a median of only four
cycles of treatment was administered. No benefit
was seen in survival, QOL, or response rates be-
tween the two arms of that randomized trial. An
increasing rate of peripheral neuropathy (a known
cumulative toxicity of the regimen used) was seen in
patients receiving more than four cycles of therapy
with no significant increase in survival rates.
The results of these two randomized trials suggest
that the survival and palliative benefit that patients
receive from chemotherapy occurs in the first 3 to 4
cycles. Prolonged therapy may increase cumulative
toxicities that are specific to the regimen used
without increasing survival, thereby decreasing the
therapeutic benefit of therapy.
Recommendation 6
The duration of first-line therapy in patients with
stage IV NSCLC should be brief, consisting of three
to four cycles or fewer if there are signs of progres-
sive disease. Level of evidence, good; benefit, sub-
Does Second-Line Chemotherapy Improve
Survival?
Since the first-line therapy used in patients with
stage IV NSCLC is not curative, patients will even-
tually experience disease progression unless they
develop another fatal comorbid illness. Once the
disease progresses, the median survival time is ap-
proximately 3 months. The proportion of patients
receiving second-line therapy following disease pro-
gression after receiving first-line platinum-based
therapy has not been well-described but is generally
50%.
42,48
Many of these patients retain a good PS
at the time of disease progression and are motivated
to receive further therapy.
Two randomized clinical trials
65,66
have been re-
ported that addressed the issue of second-line treat-
ment in patients with advanced NSCLC that
progresses after they have received first-line plati-
num-based chemotherapy. Both trials used docetaxel
because this agent had shown significant activity in
this patient population in single-arm phase II trials.
67
The eligibility requirements for both trials were also
very similar, resulting in similar groups of patients
being entered into each trial. Patients could have
received more than one previous chemotherapy reg-
imen, but 65 to 77% of the patients entered into
these studies had received only one regimen. In the
trial of Shepherd et al,
65
previous treatment with a
taxane was also an exclusion criteria. The majority of
patients had a good PS (ECOG level 0 to 1 in 75 to
83% of patients), had stage IV NSCLC (80 to 90%),
and were men (72 to 82%).
The first trial
65
compared two doses of docetaxel
(100 mg/m
2
and 75 mg/m
2
every 3 weeks) to BSC in
patients who were taxane-na ve but had previously
been treated with a platinum-based regimen. The
original design was a two-arm trial comparing do-
cetaxel, 100 mg/m
2
, to BSC. The trial was halted
when a 6% death rate was noted in 49 patients
secondary to febrile neutropenia. Also, a median of
only two cycles of therapy was delivered. The dose of
docetaxel was subsequently reduced to 75 mg/m
2
. At
this dose, the median number of cycles delivered was
four, and no febrile neutropenic deaths occurred. In
an analysis of all patients, both the median survival
times (chemotherapy arm, 7.0 months; BSC arm, 4.6
months) and the 1-year survival rates (chemotherapy
arm, 29%; BSC arm, 19%) were significantly better
for patients receiving second-line docetaxel vs those
receiving BSC (p 0.047 [log rank test]). The me-
dian survival time and the 1-year survival rate for the
patients treated with docetaxel, 75 mg/m
2
, were 7.5
months and 37%, respectively (p 0.003 compared
to BSC). The overall response rate was 7.1%, with
42.7% of patients having disease stabilization on
treatment. Clinical benefit was shown in a QOL
study
68
in which all QOL parameters favored the
docetaxel-treated patients. Specifically, a significant
reduction in pain and fatigue scale scores among the
docetaxel-treated patients and a reduction in the
need for narcotics, nonmorphine analgesic agents,
and radiotherapy were documented in this group of
patients.
The second trial
66
randomized 373 patients who
had previously received platinum-containing chemo-
therapy to one of the three following arms: docetaxel,
100 mg/m
2
; docetaxel, 75 mg/m
2
; or a control regi-
men of either vinorelbine or ifosfamide (V/I). The
overall response rates were 6.7 to 10.8% among
patients in the two docetaxel arms vs 0.8% among
patients in the V/I arm (p 0.05). The time to
progression and the progression-free survival at 26
weeks also significantly favored patients in the two
docetaxel arms. Although the median survival time
was not significantly different between the groups,
the 1-year survival rate was significantly greater with
docetaxel, 75 mg/m
2
, (32%) compared to V/I (19%;
p 0.025). The 1-year survival rate for patients
receiving docetaxel at 100 mg/m
2
was 21%.
Several of the other new agents have been studied
in the second-line setting, including paclitaxel, gem-
citabine, irinotecan, and vinorelbine.
66,69,70
In gen-
eral, these trials have included small numbers of
patients with variable results. Response rates have
ranged from 0 to 20% and median survival rates
(when reported) of 4 to 8 months. No randomized
trials including these other agents exist with the
exception of vinorelbine, as noted above.
Recommendation 7
Patients with a good PS who are experiencing
disease progression after receiving platinum-based
chemotherapy should be offered second-line chemo-
therapy. Level of evidence, good; benefit, moderate;
Is There Evidence To Support the Use of
Chemotherapy To Relieve Symptoms and
Improve QOL?
The majority of patients with advanced NSCLC
are symptomatic at some point as a result of their
disease.
71
Symptoms may be either disease-specific
(eg, cough, hemoptysis, chest pain, or dyspnea) or
disease-nonspecific (eg, weight loss, malaise, or de-
clining PS).
At least seven studies
7278
have documented palli-
ation of symptoms by chemotherapy in patients with
advanced NSCLC (Table 7). These phase II studies
generally have reported percentages of patients with
a specific symptom in whom any improvement was
noted. A substantial percentage of patients derived
symptomatic benefit from treatment for both non-
specific symptoms and organ-specific symptoms.
The rate of symptom relief appears to be higher than
the objective response rates in all the studies re-
ported, suggesting that palliation can be achieved
with tumor shrinkage that does not meet the stan-
dard criteria for objective responses.
The duration of symptom relief has not been
reported consistently but ranges from 1.5 to 3.5
months.
72,75,77
The review by Thatcher et al,
78
exam-
ining the palliative impact of gemcitabine, suggested
that some symptoms may be more effectively palli-
ated than others (ie, median relief of 3 to 5 months
for dyspnea, cough, and chest pain, and median
relief of 2 to 3 months for anorexia and hemoptysis).
When the median duration of symptom relief is
compared with the median survival time of this
population, symptom relief can be achieved for
approximately 25 to 50% of the patients life span. It
is important to provide patients with this information
regarding the relief of symptoms.
In addition to symptom relief, the impact of
chemotherapy on the patients overall PS also has
been reported. Eight nonrandomized trials involving
770 patients were reported in a review by Thatcher
et al.
78
These trials included chemotherapy regimens
consisting of platinum combinations, nonplatinum
single agents, and combinations. Almost all of the
trials excluded patients with a PS of 2. The response
rates to the chemotherapy regimens ranged from 20
to 59%, and the median survival time ranged from 5
to 13 months. Approximately one third of patients
experienced an improvement in PS (range, 4 to
52%), and another one third of patients had a stable
PS while receiving treatment (range, 30 to 67%).
Two other trials
24,79
have confirmed these findings.
The duration of improvement has not been reported
consistently. However, when it has been re-
ported, improvement typically lasted from 4 to 6
months.
73,78,79
A randomized phase III trial comparing gemcitab-
ine with BSC has been reported and highlights the
palliative aspect of chemotherapy on disease-related
symptoms.
41
Symptom control was the primary end
point of this study, which involved 299 symptomatic
patients with advanced or metastatic NSCLC. Im-
provements were noted in terms of the need for
palliative radiotherapy for progressive symptoms. At
2 months, 42.3% of patients in the BSC arm required
radiotherapy compared with 7.3% of patients in the
gemcitabine arm. Also, the median time before
radiotherapy was needed was 7 months for patients
receiving gemcitabine vs 1 month for those receiv-
ing BSC (p 0.0001). QOL questionnaires and a
patient-assessed symptom scale noted improvements
significantly favoring the gemcitabine arm of the
study. The overall response rate for gemcitabine was
17%. No difference in survival was noted; however,
this was not the end point of the study.
QOL is an important aspect of treatment that
differs from symptom relief or assessment of PS.
QOL attempts to define the patients perception of
how the disease and its treatment affect his or her
well-being in all of lifes domains (ie, functional,
social, psychological, spiritual, disease-related symp-
toms, and treatment side effects). Several question-
naires have been developed and validated for clinical
use.
80,81
In the past, it has been difficult to obtain
carefully conducted serial QOL measurements in
patients receiving chemotherapy. This may be due to
the deterioration of the patients condition and to the
unwillingness or inability of the patient to complete
the questionnaires.
24,82,83
One of the early QOL
instruments developed was the functional living
index-cancer.
84
Using this instrument, two reports
suggested that initial functional living index-cancer
QOL scores were more predictive of survival than
were other parameters.
85,86
This suggests that QOL
measurements could further refine the assessment of
PS, as has been suggested in breast cancer patients.
87
Three randomized studies demonstrated im-
proved QOL with chemotherapy compared with
BSC.
31,32,39
In one study,
39
single-agent vinorelbine
was compared with BSC in patients 70 years of
age. The QOL instrument used was the European
Organization for Research and Treatment of Cancer
(EORTC) questionnaire and its lung cancer-specific
module. Although longitudinal compliance was not
optimal, EORTC functional scales were consistently
better for the patients receiving vinorelbine than for
control patients. Also, symptom improvement scores
were clearly better for some lung cancer-specific
items (ie, pain and dyspnea), which was consistent
with the results of the trial of gemcitabine vs BSC
41
that was discussed above. In the other two trials,
31,32
cisplatin-based combination chemotherapy improved
QOL compared with BSC.
Cella et al
88
suggested that prognosis can be
Table 7Relief of Symptoms by Chemotherapy in Advanced NSCLC*
Variables
Ellis
et al
72
(n 120)
Cullen
et al
73
(n 74)
Osoba
et al
74
(n 53)
Tummarello
et al
75
(n 46)
Fernandez
et al
76
(n 31)
Hardy
et al
77
(n 24)
Thatcher
et al
78
(NR)
Regimen used MVbP MIP BEP MVbP PVbMVI MVbP Gemcitabine
Objective response rate 32 56 44 33 42 21 20
Median survival time, mo 5.0 9.8 5.0 6.5 6 8.19.2
Symptom improvement
Cough 66 70 68 40 45 71 44
Hemoptysis 92 78 100 91 63
Pain 60 77 68 39 47 63 32
Dyspnea 59 46 31 66 78 65 26
Weight loss 44 30
Anorexia 58 50 29
Malaise 53 53 62
*Values given as %, unless otherwise indicated. BEP bleomycin, etoposide, and platinum; PVbMVI platinum, vinblastine, mitomycin,
vincristine, and ifosfamide. See Table 2 for abbreviations not used in the text.
Percentage of patients with a specific symptom who had relief or improvement in this symptom with treatment.
predicted from baseline QOL as well as from early
changes in a patients QOL. Using the FACT-L,
baseline, 6-week, 12-week, and 6-month QOL mea-
surements were obtained for 571 patients who were
enrolled in a three-arm, randomized, phase III trial
comparing a cisplatin-etoposide regimen with two
different schedules of cisplatin-paclitaxel. The study
showed a survival advantage for patients in the
paclitaxel-containing arms.
44
There were no differ-
ences in any of the QOL scores among patients in
the three arms of the study. The FACT-L physical
well-being emerged second (p 0.01) behind treat-
ment with a paclitaxel regimen (p 0.01) in predict-
ing a response to treatment in a model that consid-
ered multiple clinical factors (ie, disease stage, PS,
weight loss, comorbidity, presence of symptoms, and
FACT-L scores). The baseline trial outcome index
(TOI), which combines physical, functional, and lung
cancer symptom scores, correlated highly with sur-
vival, but a change from baseline to subsequent
assessment proved to be more important. The fol-
lowing four groups of patients were identified:
(1) those with a high baseline TOI who improved at
6 weeks; (2) those with a high baseline TOI who did
not improve at 6 weeks; (3) those with a low baseline
TOI who improved at 6 weeks; and (4) those with a
low baseline TOI who did not improve at 6 weeks.
The median survival times for the four groups were
16, 11, 11, and 5 months, respectively. The authors
suggested that a change in QOL may be able to
predict survival and may aid in decisions that are
made during the course of chemotherapy. Compli-
ance with the collection of QOL data was not
reported, but one has to wonder how this may have
influenced outcomes. Although interesting, these
findings need corroboration in future studies.
There may be differences in QOL among patients
receiving various treatment regimens. An EORTC
study
47
randomized patients with advanced disease
to receive either cisplatin-teniposide or cisplatin-
paclitaxel. The response rates were 28% for patients
receiving cisplatin-teniposide vs 41% for those re-
ceiving cisplatin-paclitaxel. However, the 1-year sur-
vival rate was the same. Using the EORTC instru-
ment, QOL measurements favored patients in
the paclitaxel-containing arm. In another study
45
comparing a cisplatin-etoposide regimen with a
carboplatin-paclitaxel regimen, the QOL using the
FACT-L significantly favored patients receiving the
carboplatin-paclitaxel regimen during the first 6
weeks of treatment. Patients in the carboplatin-
paclitaxel arm also had a significantly higher re-
sponse rate. In two other phase III studies
48,49
comparing cisplatin-gemcitabine regimens, either
with mitomycin-ifosfamide-cisplatin or with cisplatin
alone, no difference in global QOL was reported
between the two arms, despite a significantly higher
response rate in the cisplatin-gemcitabine arm in
both studies.
Baseline QOL appears to be an important prog-
nostic factor, and differences in QOL may occur
during treatment with different chemotherapy regi-
mens. More data are needed regarding the longitu-
dinal measurement of QOL during chemotherapy to
ascertain how this correlates with baseline PS mea-
surements and response rates.
87
Also, further refine-
ment and simplification of QOL instruments will
help to capture data on the chronic toxicity of
chemotherapy and its impact on the patients QOL.
Recommendation 8
Data from case series and randomized trials show
that chemotherapy can have a palliative effect on
disease-related symptoms and can improve QOL
compared to BSC in stage IV NSCLC patients who
are deemed suitable for treatment. Level of evi-
dence, good; benefit, moderate; grade of recommen-
dation, B
What Are Patients Preferences and
Attitudes Toward Chemotherapeutic
Treatment Options for Advanced NSCLC?
Two descriptive studies utilizing cancer patients
who previously had been treated addressed the issue
of patient preferences and attitudes toward receiving
palliative cisplatin-based chemotherapy compared to
BSC for survival and/or QOL benefit. Using a time-
tradeoff technique, 60 patients were interviewed to
address attitudes toward the improved median sur-
vival time and 1-year survival rate for the addition of
cisplatin chemotherapy and BSC compared to BSC
alone. Attitudes ranged from willingness to choose
more toxic treatment with no survival advantage to
declining chemotherapy regardless of the survival
benefit. Over half of the participants (57%) would
choose chemotherapy if it improved the 1-year sur-
vival rate by 10%.
89
A second descriptive study of
81 patients with advanced NSCLC who previously
had been treated with a cisplatin-based chemother-
apy regimen was performed to measure the survival
threshold that patients would be willing to accept for
receiving chemotherapy.
90
As in the prior study, the
response varied between a survival benefit of 1 week
to not accepting treatment even if survival were to be
improved by 24 months. A median survival threshold
for accepting chemotherapy with mild toxicity was
4.5 months, and for severe toxicity it was 9 months.
Only 22% of patients would be willing to choose
chemotherapy for a survival benefit of 3 months.
However, over half (68%) would choose chemother-
apy if it substantially improved QOL.
No study has been able to demonstrate or predict
a necessary minimum threshold improvement in
survival or QOL based on age, sex, education, PS, or
role in the treatment decision making. One study
91
has suggested that patients with cancer are much
more likely to opt for radical treatment that could
prolong life and relieve symptoms with minimal
chance for benefit than people who do not have
cancer, including health professionals. Therefore,
the different attitudes of health professionals toward
the benefits of chemotherapy for the patient need to
be considered.
Recommendation 9
Patient preferences need to be considered and
respected with regard to the decision to treat with
chemotherapy. Most patients would not choose che-
motherapy for a likely survival time of 3 months or a
10% improvement in the 1-year survival rate
unless there was an improvement in QOL. No
patient variables have been identified to determine
an individual patients minimum threshold to accept
chemotherapy, and therefore the decision to treat
with chemotherapy needs to be discussed with each
patient individually. Level of evidence, fair; benefit,
Is There Any Evidence That Would Support
Who Administered the Chemotherapy
Made a Difference?
The MEDLINE search addressing this issue
yielded no citations that were relevant with regard to
addressing this question. Since the evaluation of
NSCLC patients for chemotherapy requires an un-
derstanding of its indication as well as the proper
selection of patients, physicians performing these
duties should have experience and specialized train-
ing. This specialized training also should include
experience with the proper administration of chemo-
therapy protocols as well as a working knowledge of
the toxicity (both acute and chronic) of all chemo-
therapeutic agents used in this disease setting. In
addition, physicians should have the proper infra-
structure necessary for the administration of com-
monly used chemotherapeutic agents/regimens and
the management of the common complications of
these agents. Although no data exist on this issue, it
seems intuitive that patients who are potential can-
didates for chemotherapy should be referred to
physicians with specialized training in its administra-
tion and with experience in the management of such
patients.
Recommendation 10
Patients with stage IV NSCLC should be referred
to a physician with specialized training in oncology.
If chemotherapy is considered to be appropriate,
adequate resources to administer chemotherapy
safely must be available. Level of evidence, poor;
benefit, substantial; grade of recommendation, C
What Are the Outcome Expectations and
Adverse Effects Seen With Chemotherapy
and How Do They Compare With the
Natural History?
The natural history of untreated stage IV NSCLC
is best documented in the randomized trials of
chemotherapy vs BSC (Tables 2 and 5). The impact
that chemotherapy has on survival is significant and
has been discussed in the previous sections. When
QOL has been examined, patients receiving chemo-
therapy report better scores compared to patients
receiving only BSC,
31,32,39
supporting the contention
that the disease is worse than the treatment. The
expectations regarding survival and toxicity when
using modern chemotherapy regimens are shown in
Table 8. The trials shown in Table 8 predominantly
include patients with good PS with stage IV disease
who have been studied in large phase III trials that
have been published in peer-reviewed journals. Only
platinum-based regimens are described as those
regimens represent the standard of care in patients
with good PS and stage IV NSCLC. As is shown in
Table 8, the median survival time and the 1-year
survival rate expectations are 8.0 to 9.9 months and
30 to 43%, respectively. The major toxicities are
hematologic, with neutropenia being the predomi-
nant adverse effect. Despite this, the clinical conse-
quences of severe neutropenia (ie, sepsis) occur in
10% of patients, and the treatment-related death
rates in these studies ranged from 0 to 4%. Severe
anemia occurs in 7 to 30% of patients. Severe
thrombocytopenia varies depending on the regimen
used as well as on the dose and schedule of the
agents in the regimen. However, bleeding complica-
tions are unusual. Nonhematologic toxicity consists
mainly of nausea/vomiting, fatigue, and alopecia. In
the more recent trials employing modern antiemetic
regimens or using carboplatin rather than cisplatin,
the rates of severe nausea/vomiting range from 7 to
20%. The toxicity profiles of single agents are some-
what less than those of combination regimens and
can be found in the individual references provided.
It should also be noted that the risk of toxicity
increases in patients with PS 2.
92
Recommendation 11
Combination platinum-based chemotherapy can
be administered safely with acceptable and manage-
able toxicity profiles in patients with good PS who
have stage IV NSCLC. Level of evidence, good;
benefit, substantial; grade of recommendation, A
Conclusion
Chemotherapy improves survival and palliates
symptoms, thereby improving QOL in patients with
stage IV NSCLC in both the first-line and second-
line setting. Selecting patients based on PS is impor-
tant as patients significantly compromised by their
disease may not benefit from therapy and may
experience excessive toxicity. Both platinum-based
regimens as well as individual single-agent regimens
have an impact on survival. However, platinum-
based combination regimens using the new third-
generation agents represent standard-of-care regi-
mens. The impact that chemotherapy has on patient
survival probably occurs early in the treatment, and
the prolonged administration of therapy is not indi-
cated. Physicians involved in the evaluation and
management of patients with stage IV NSCLC
should be aware of the potential benefits of chemo-
therapy, allowing them to make appropriate recom-
mendations for patients under their care.
1. When selecting patients for systemic chemo-
therapy, PS at the time of diagnosis should be
used because it is a consistent prognostic
factor for survival. Patients with a PS (PS) of
ECOG 0 or 1 should be offered chemotherapy
(level of evidence, good; benefit, substantial;
grade of recommendation, A). Data are not
yet sufficient to routinely recommend chemo-
therapy to patients with a PS of ECOG level 2
(level of evidence, poor; benefit, small/weak;
grade of recommendation, I). Patients with a
PS of ECOG level 3 or 4 should not receive
chemotherapy (level of evidence, fair; benefit,
moderate; grade of recommendation, B).
Other patient-related factors (eg, gender, age,
sites of metastases, or histology) have not been
consistent prognostic factors for survival.
Level of evidence, poor; benefit, small/weak;
2. Patients with good PS (ie, ECOG level 0 or 1)
should be considered for a platinum-based
chemotherapy regimen based on the survival
advantage provided over BSC. Level of evi-
recommendation, A
3. Although new agents demonstrate improved
survival compared to BSC (level of evidence,
good; benefit, moderate; grade of recommen-
dation, B) in elderly patients as well as in
nonelderly patients with advanced NSCLC,
the data are not yet sufficient to compare the
new single agents to platinum-based combina-
tion therapies. Level of evidence, poor; bene-
fit, small/weak; grade of recommendation, I
4. Combination chemotherapy regimens incor-
porating the new single agents with a platinum
Table 8Survival and Toxicity Expectations with Modern Chemotherapy Regimens*
Expectations
Cisplatin-
Vinorelbine
42,43,58
Carboplatin-
Paclitaxel
58
Cisplatin-
Paclitaxel
44,46,47
Cisplatin-
Gemcitabine
4850
Survival
Median time, mo 8.09.3 8.6 8.19.9 8.69.1
1-year rate, % 36 38 3043 3239
Toxicity, %
Neutropenia 7681 57 4569 4064
Anemia 724 13 1020 2230
Thrombocytopenia 36 10 12 5064
Renal 56 01
Neurologic 57 13 440 01
Nausea/Vomiting 2058 7 1012 1239
Diarrhea 711 1 1 12
Hepatis 02 12
Sepsis 410 1 29 15
TRD 24 2 03 01
*These regimens were chosen because they represent regimens tested in randomized phase III trials and published in prereview journals.
The quoted percentages represent the rates of grade 3 or 4 toxicity reported.
Treatment-related deaths.
agent should be considered the standard of
care. Level of evidence, fair; benefit, moder-
5. No one regimen has been demonstrated to be
superior in the first-line therapy for patients
with advanced NSCLC. A cisplatin-based or
carboplatin-based combination regimen that
includes one of the new agents remains the
standard of care for first-line therapy in pa-
tients with stage IV NSCLC. Level of evi-
recommendation, A
6. The duration of first-line therapy in patients
with stage IV NSCLC should be brief, con-
sisting of 3 to 4 cycles or fewer if there are
signs of progressive disease. Level of evi-
recommendation, A
7. Patients with a good PS in whom disease
progresses after receiving platinum-based
chemotherapy should be offered second-line
chemotherapy. Level of evidence, good; ben-
8. Data from case series and randomized trials
show that chemotherapy can have a palliative
effect on disease-related symptoms and can
improve QOL compared to BSC in stage IV
NSCLC patients who are deemed suitable for
treatment. Level of evidence, good; benefit,
9. Patient preferences need to be considered
and respected with regard to the decision to
treat with chemotherapy. Most patients would
not choose chemotherapy for a likely survival
of 3 months or a 10% improvement in the
1-year survival rate unless there was an im-
provement in QOL. No patient variables have
been identified to determine an individual
patients minimum threshold to accept che-
motherapy, and therefore the decision to treat
with chemotherapy needs to be discussed with
each patient individually. Level of evidence,
fair; benefit, moderate; grade of recommen-
dation, B
10. Patients with stage IV NSCLC should be
referred to a physician with specialized train-
ing in oncology. If chemotherapy is consid-
ered to be appropriate, adequate resources to
administer chemotherapy safely must be avail-
able. Level of evidence, poor; benefit, sub-
11. Combination platinum-based chemotherapy
can be administered safely and with accept-
able and manageable toxicity profiles in pa-
tients with good PS who have stage IV
NSCLC. Level of evidence, good; benefit,
ACKNOWLEDGMENT: Thanks to Lenka Cook for the creation
of Figure 2.
Appendix
The following search terms were used in the study: age and
lung cancer; antineoplastic agents, combined; carcinoma, non-
small cell lung; carcinoma, non-small cell lung/drug therapy;
carcinoma, non-small cell lung/therapy; chemotherapy; clinical
trials; combination chemotherapy; duration of therapy; lung
neoplasms; lung neoplasms/drug therapy; lung neoplasms/thera-
py; outcomes; performance status and lung cancer; prognosis
factors and lung cancer; prognosis and lung cancer; prognosis and
non-small cell lung cancer; quality of life and lung cancer;
randomized trials; sex and lung cancer; stage IV non-small cell
lung cancer; weight loss and lung cancer.
References
1 American Cancer Society. Cancer facts and figures, 2002.
Available at: www.cancer.org. Accessed October 28, 2002
2 Socinski MA. Chemotherapy for stage IV non-small cell lung
cancer. In: Detterbeck FC, Rivera MP, Socinski MA, et al,
based guide for the practicing clinician. Philadelphia, PA: WB
Saunders, 2001; 307325
4 Lanzotti VJ, Thomas DR, Boyle LE, et al. Survival with
inoperable lung cancer: an integration of prognostic variables
based on simple clinical criteria. Cancer 1977; 39:303313
5 Stanley KE. Prognostic factors for survival in patients with
inoperable lung cancer. J Natl Cancer Inst 1980; 65:2532
6 OConnell JP, Kris MG, Gralla RJ, et al. Frequency and
prognostic importance of pretreatment clinical characteristics
in patients with advanced non-small cell lung cancer treated
with combination chemotherapy. J Clin Oncol 1986; 4:1604
1614
7 Finkelstein DM, Ettinger DS, Ruckdeschel JC. Long-term
survivors in metastatic non-small cell lung cancer: an Eastern
Cooperative Oncology Group study. J Clin Oncol 1986;
4:702709
8 Ruckdeschel JC, Finkelstein DM, Ettinger DS, et al. A
randomized trial of the four most active regimens for meta-
static non-small cell lung cancer. J Clin Oncol 1986; 4:1422
9 Bonomi PD, Finkelstein DM, Ruckdeschel JC, et al. Combi-
nation chemotherapy vs single agents followed by combina-
tion chemotherapy in stage IV non-small cell lung cancer: a
study of the Eastern Cooperative Oncology Group. J Clin
Oncol 1989; 7:16021613
10 Klastersky J, Sculier JP, Bureau G, et al. Cisplatin vs cisplatin
plus etoposide in the treatment of advanced non-small cell
11 Leudke DW, Einhorn L, Omura GA, et al. Randomized
comparison of two combination regimens vs minimal chemo-
therapy in non-small cell lung cancer: a Southeastern Cancer
Study Group Trial. J Clin Oncol 1990; 8:886891
12 Miller TP, Chen TT, Coltman CA, et al. Effect of alternating
combination chemotherapy on survival of ambulatory patients
with metastatic large cell and adenocarcinoma of the lung: a
Southwest Oncology Group Study. J Clin Oncol 1986; 4:502
508
13 Albain KS, Crowley JJ, LeBlanc M, et al. Survival determi-
nants in extensive-stage non-small cell lung cancer: the
Southwest Oncology Group Experience. J Clin Oncol 1991;
9:16181626
14 Liu NS, Spitz MR, Kemp BL, et al. Adenocarcinoma of the
lung in young patients: the MD Anderson experience. Cancer
2000; 88:18371841
15 Kim K, Bonomi P, Kargler J, et al. Two-year survival and
prognostic factors for overall survival in advanced NSCLC: an
ECOG experience [abstract]. Proc Am Soc Clin Oncol 1998;
17:461a
16 Herndon JE, Fleishman S, Kornblith AT, et al. Is quality of
life predictive of survival in patients with advanced NSCLC?
Cancer 1999; 85:333340
17 Donnadieu N, Paesmans M, Sculier J-P. Chemotherapy of
non-small cell lung cancer according of disease extent: a
meta-analysis of the literature. Lung Cancer 1991; 7:243252
18 Kelly K, Bunn PA Jr. Is it time to reevaluate our approach to
the treatment of brain metastases in patients with non-small
cell lung cancer. Lung Cancer 1998; 20:8591
19 Minotti V, Crino L, Meacci ML, et al. Chemotherapy with
cisplatin and teniposide for cerebral metastases in non-small
20 Vigano A, Binera E, Jhangri GS, et al. Clinical survival
predictors in patients with advanced cancer. Arch Intern Med
2000; 160:861868
21 Graziano SL, Mazid R, Newman N, et al. The use of
neuroendocrine immunoperoxidase markers to predict che-
motherapy response in patients with non-small cell lung
22 Berendsen HH, de Leij L, Poppema S, et al. Clinical
characterization of non-small cell lung cancer tumors showing
neuroendocrine differentiation features. J Clin Oncol 1989;
7:16141620
23 Rapp E, Pater JL, Willan A, et al. Chemotherapy can prolong
survival in patients with advanced non-small cell lung cancer:
report of a Canadian multicenter randomized trial. J Clin
Oncol 1988; 6:633641
24 Ganz PA, Figlin RA, Haskell CM, et al. Supportive care vs
supportive care and combination chemotherapy in metastatic
non-small cell lung cancer. Cancer 1989; 63:12711278
25 Woods RL, Williams CJ, Levi J, et al. A randomized trial of
cisplatin and vindesine vs supportive care only in advanced
non-small cell lung cancer. Br J Cancer 1990; 61:608611
26 Quoix E, Dietemann A, Charbonneau J, et al. La chimio-
therapie comportant du cisplatine est elle utile dans le cancer
bronchique non microcellulaire au stade IV?: resultats dune
etude ranodmisee. Bull Cancer 1991; 78:341346
27 Cellerino R, Tummarello D, Guidi F, et al. A randomized
trial of alternating chemotherapy vs best supportive care in
advanced non-small cell lung cancer. J Clin Oncol 1991;
9:14531461
28 Kaasa S, Lund E, Thorud E, et al. Symptomatic treatment vs
combination chemotherapy for patients with extensive non-
small cell lung cancer. Cancer 1991; 67:24432447
29 Cartei G, Cartei F, Canotne A, et al. Cisplatin-cyclophos-
phamide-mitomycin combination chemotherapy with sup-
portive care vs supportive care alone for treatment of meta-
static non-small cell lung cancer. J Natl Cancer Inst 1993;
85:794800
30 Helsing M, Bergman B, Thaning L, et al. Quality of life and
survival in patients with advanced non-small cell lung cancer
receiving supportive care plus chemotherapy with carboplatin
and etoposide or supportive care only: a multicenter random-
ized phase III trial. Eur J Cancer 1998; 34:10361044
31 Thongprasert S, Sanuganmitra P, Juthapan N, et al. Relation-
ship between quality of life and clinical outcomes in advanced
non-small cell lung cancer: best supportive care (BSC) vs
BSC plus chemotherapy. Lung Cancer 1999; 24:1724
32 Cullen MH, Billingham LJ, Woodroffe CM, et al. Mitomycin,
ifosfamide, and cisplatin in unresectable non-small cell lung
cancer: effects on survival and quality of life. J Clin Oncol
1999; 17:31883194
33 Souquet PJ, Chauvin F, Boissel JP, et al. Polychemotherapy
in advanced non-small cell lung cancer: a meta-analysis.
Lancet 1993; 11:18661872
34 Grilli R, Oxman AD, Julian JA. Chemotherapy for advanced
non-small cell lung cancer: how much benefit is enough?
J Clin Oncol 1993; 11:18661872
35 Marino P, Pampallona S, Preatoni A, et al. Chemotherapy vs
supportive care in advanced non-small cell lung cancer:
results of a meta-analysis of the literature. Chest 1994;
106:861865
36 The Non-Small Cell Lung Cancer Collaborative Group.
Chemotherapy in non-small cell lung cancer: a meta-analysis
using updated data on individual patients from 52 random-
ized clinical trials. BMJ 1995; 311:899909
37 Roszkowski K, Pluzanska A, Krzakowski M, et al. A multi-
center, randomized phase III study of docetaxel plus best
supportive care vs best supportive care in chemotherapy-
nave patients with metastatic or non-resectable localized
non-small cell lung cancer (NSCLC). Lung Cancer 2000;
27:145157
38 The Elderly Lung Cancer Vinorelbine Italian Study (ELVIS)
Group. Effects of vinorelbine on quality of life and survival in
elderly patients with advanced non-small cell lung cancer.
J Natl Cancer Inst 1999: 91:6672
39 Ranson M, Davidson N, Nicolson M, et al. Randomized trial
of paclitaxel plus supportive care vs supportive care for
patients with advanced non-small cell lung cancer. J Natl
Cancer Inst 2000; 92:10741080
40 Crawford J, ORourke M, Schiller JH, et al. Randomized trial
of vinorelbine compared with fluorouracil plus leucovorin in
patients with stage IV non-small cell lung cancer. J Clin Oncol
1996; 14:27742784
41 Anderson H, Hopwood R, Stephens RJ, et al. Gemcitabine
plus best supportive care (BSC) vs BSC in inoperable non-
small cell lung cancer: a randomized trial with quality of life
as the primary outcome. Br J Cancer 2000; 83:447453
42 Le Chevalier T, Brisgand D, Douillard J-Y, et al. Randomized
study of vinorelbine and cisplatin vs vindesine and cisplatin
versus vinorelbine alone in advanced non-small cell lung
cancer: results of a European multicenter trial including 612
patients. J Clin Oncol 1994; 12:360367
43 Wozniak AJ, Crowley JJ, Balcerzak SP, et al. Randomized trial
comparing cisplatin with cisplatin plus vinorelbine in the
treatment of advanced non-small cell lung cancer: a South-
west Oncology Group Study. J Clin Oncol 1998; 16:2459
2465
44 Bonomi P, Kim KM, Fairclough D, et al. Comparison of
survival and quality of life in advanced non-small cell lung
cancer patients treated with two dose levels of paclitaxel
combined with cisplatin vs etoposide with cisplatin: results of
an Eastern Cooperative Oncology Group trial. J Clin Oncol
2000; 18:623631
45 Belani CP, Natale RB, Lee JS, et al. Randomized phase III
trial comparing cisplatin/etoposide vs carboplatin/paclitaxel in
advanced non-small cell lung cancer (NSCLC) [abstract].
Proc Am Soc Clin Oncol 1998; 17:455a
46 Gatzemeier U, von Pawel J, Gottfried M, et al. Phase III.
comparative stud of high-dose cisplatin vs a combination of
paclitaxel and cisplatin in patients with advanced non-small
cell lung cancer. J Clin Oncol 2000; 18:33903399
47 Giaccone G, Splinter TAW, Debruyne C, et al. Randomized
study of paclitaxel-cisplatin vs cisplatin-teniposide in patients
with advanced non-small cell lung cancer. J Clin Oncol 1998;
16:21332141
48 Sandler AT, Nemunaitis J, Denham C, et al. Phase III trial of
gemcitabine plus cisplatin vs cisplatin alone in patients with
locally advanced or metastatic non-small cell lung cancer.
J Clin Oncol 2000; 18:122130
49 Crino L, Scaglotti GV, Ricci S, et al. Gemcitabine and
cisplatin vs mitomycin, ifosfamide, and cisplatin in advanced
non-small cell lung cancer: a randomized phase III study of
the Italian Lung Cancer Project. J Clin Oncol 1999; 17:3522
3530
50 Cardenal F, Lopez-Cabrerizo M, Anton A, et al. Randomized
phase III study of gemcitabine-cisplatin vs etoposide-cisplatin
in the treatment of locally advanced or metastatic non-small
cell lung cancer. J Clin Oncol 1999; 17:1218
51 Perng R-P, Chen Y-M, Ming-Liu J, et al. Gemcitabine vs the
combination of cisplatin and etoposide in patients with
inoperable non-small cell lung cancer in a phase II random-
ized study. J Clin Oncol 1997; 15:20972102
52 Manegold C, Bergman B, Chemaissani A, et al. Single-agent
gemcitabine vs cisplatin-etoposide: early results of a random-
ized phase II study in locally advanced or metastatic non-
small cell lung cancer. Ann Oncol 1997; 8:525529
53 Masuda N, Fukuoka M, Negoro S, et al. Randomized trial
comparing cisplatin (CDDP) and irinotecan (CPT-11) vs
CDDP and vindesine (VDS) vs CPT-11 alone in advanced
non-small cell lung cancer (NSCLC), a multicenter phase III
study [abstract]. Proc Am Soc Clin Oncol 1999; 18:492a
54 Niho S, Nagao K, Nishiwaki Y, et al. Randomized multicenter
phase III trial of irinotecan (CPT-11) and cisplatin (CDDP)
vs CDDP and vindesine (VDS) in patients with advanced
non-small cell lung cancer (NSCLC) [abstract]. Proc Am Soc
55 Fukuoka M, Nagao K, Ohashi Y, et al. Impact of irinotecan
(CPT-11) and cisplatin (CDDP) on survival in previously
untreated metastatic non-small cell lung cancer (NSCLC)
[abstract]. Proc Am Soc Clin Oncol 2000; 19:495a
56 Breathnach OS, Freidlin B, Conley B, et al. Twenty-two years
of phase III trials for patients with advanced non-small cell
lung cancer: sobering results. J Clin Oncol 2001; 19:1734
1742
57 Baggstrom MQ, Socinski MA, Hensing TA, et al. Third
generation chemotherapy regimens (3GR) improve survival
over second generation regimens (2GR) in stage IIIB/IV
non-small cell lung cancer (NSCLC): a meta-analysis of the
published literature [abstract]. Proc Am Soc Clin Oncol 2002;
21:306a
58 Kelly K, Crowley J, Bunn PA Jr, et al. Randomized phase III
trial of paclitaxel plus carboplatin vs vinorelbine plus cisplatin
in the treatment of patients with advanced non-small cell lung
cancer: a Southwest Oncology Group Trial. J Clin Oncol
2001; 19:32103218
59 Schiller JH, Harrington D, Belani CP, et al. Comparison of
four chemotherapy regimens for advanced non-small cell lung
60 Georgoulias V, Papadakis E, Alexopoulos A, et al. Platinum-
based, and non-platinum based chemotherapy in advanced
non-small lung cancer: a randomized multicenter trial. Lancet
2001; 357:14781484
guidelines for the treatment of unresectable non-small cell
62 Buccheri GF, Ferrigno D, Curcio A, et al. Continuation of
chemotherapy vs supportive care alone in patients with
inoperable non-small cell lung cancer and stable disease after
two or three cycles of MACC. Cancer 1989; 63:428432
63 Smith IE, OBrien MER, Talbot DC, et al. Duration of
chemotherapy in advanced non-small cell lung cancer: a
randomized trial of three vs six courses of mitomycin, vinblas-
tine, and cisplatin. J Clin Oncol 2001; 19:13361343
64 Socinski MA, Schell MJ, Peterman A, et al. A phase III trial
comparing a defined duration of therapy vs continuous
therapy followed by second-line therapy in advanced stage
IIIB/IV non-small cell lung cancer. J Clin Oncol 2002;
20:13351343
65 Shepherd FA, Dancey J, Ramlau R, et al. Prospective ran-
domized trial of docetaxel vs best supportive care in patients
with non small cell lung cancer previously treated with
platinum-based chemotherapy. J Clin Oncol 2000;18:2095
2103
66 Fossella FV, DeVore R, Kerr RN, et al. Randomized phase
III trial of docetaxel vs vinorelbine or ifosfamide in patients
with advanced non-small cell lung cancer previously treated
with platinum-containing chemotherapy regimens: the TAX
320 Non-Small Cell Lung Cancer Study Group. J Clin Oncol
2000; 18:23542360
67 Fossella FV, Lee JS, Murphy WK, et al. Phase II trial of
docetaxel for recurrent or metastatic non-small cell lung
68 Dancey J, Shepherd F, Ramlau R, et al. Quality of Life
(QOL) assessment in randomized study of taxotere (TAX)
verus best supportive care (BSC) in non-small cell lung
cancer (NSCLC) patients (pts) previously treated with plati-
num-based chemotherapy [abstract]. Proc Am Soc Clin Oncol
1999; 18:491a
69 Huisman C, Smit EF, Giaccone G, et al. Second-line chemo-
therapy in relapsing or refractory non-small cell lung cancer:
a review. J Clin Oncol 2000; 18:37223730
70 Belani CP. Single agents in the second-line treatment of
non-small cell lung cancer. Semin Oncol 1998; 25:1014
71 Hollen PJ, Gralla RJ, Kris MG, et al. Quality of life assess-
ment in individuals with lung cancer: testing the Lung Cancer
Symptom Scale (LCSS). Eur J Cancer 1993; 3:14781485
72 Ellis PA, Smith IE, Hardy JR, et al. Symptom relief with MVP
(mitomycin C, vinblastine and cisplatin) chemotherapy in
advanced non-small cell lung cancer. Br J Cancer 1995;
71:336370
73 Cullen MH, Joshi R, Chetiyawardana AD, et al. Mitomycin,
ifosfamide and cisplatin in non-small cell lung cancer: treat-
ment good enough to compare. Br J Cancer 1998; 58:359
361
74 Osoba D, Rusthoven JJ, Turnbull KA, et al. Combination
chemotherapy with bleomycin, etoposide, and cisplatin in
metastatic non-small cell lung cancer. J Clin Oncol 1985;
3:14781485
75 Tummarello D, Graziano F, Isidori P, et al. Symptomatic,
stage IV, non-small cell lung cancer (NSCLC): response,
toxicity, performance status change and symptom relief in
patients treated with cisplatin, vinblastine and mitomycin-C.
Cancer Chemother Pharmacol 1995; 35:249253
76 Fernandez C, Rosell R, Abad-Esteve A, et al. Quality of life
during chemotherapy in non-small cell lung cancer patients.
Acta Oncol 1989; 28:2933
77 Hardy JR, Noble T, Smith IE. Symptom relief with moderate
dose chemotherapy (mitomycin-C, vinblastine and cisplatin)
in advanced non-small cell lung cancer. Br J Cancer 1989;
60:764766
78 Thatcher N, Anderson H, Betticher DC, et al. Symptomatic
benefit from gemcitabine and other chemotherapy in ad-
vanced non-small cell lung cancer: changes in performance
status and tumour-related symptoms. Anticancer Drugs 1995;
6(suppl):3948
79 Vinante O, Bari M, Segati R, et al. The combination of
mitomycin, vinblastine and cisplatin is active in the palliation
of stage IIIB/IV non-small cell lung cancer. Oncology 1993;
50:14
80 Montazeri A, Gillis CR, McEwen J. Quality of life in patients
with lung cancer: a review of literature from 1970 to 1995.
Chest 1998; 113:467481
81 Cella DF, Tulsky DS, Gray G, et al. The functional assess-
ment of cancer therapy scale: development and validation of
the general measure. J Clin Oncol 1993; 11:570579
82 Finkelstein DM, Cassileth BR, Bonomi PD, et al. A pilot
study of the functional living index-cancer (FLIC) scale for
the assessment of quality of life for metastatic lung cancer
patients: an Eastern Cooperative Oncology Group study.
Am J Clin Oncol 1988; 11:630633
83 Bonomi P. Non-small cell lung cancer chemotherapy. In: Pass
HI, Mitchel JB, eds. Lung cancer: principles and practice.
Philadelphia, PA: Lippincott-Raven, 1996; 811823
84 Schipper H, Clinch J, McMurray A, et al. Mesuring the
quality of life of cancer patients: the functional living index-
cancer; development and validation. J Clin Oncol 1984;
2:472478
85 Ganz PA, Lee JJ, Sian J. Quality of life assessment: an
independent prognostic variable for survival in lung cancer.
Cancer 1991; 67:31313139
86 Ruckdeschel JC, Piantadosi S. Assessment of quality of life by
Functional Living Index-Cancer (FLIC) is superior to per-
formance status for prediction of survival in patients with lung
cancer [abstract]. Proc Am Soc Clin Oncol 1989, 8:311
87 Weeks J. Quality-of-life assessment: performance status up-
staged? J Clin Oncol 1992; 10:18271829
88 Cella D, Fairclough DL, Bonomi PB, et al. Quality of life
(QoL) in advanced non-small cell lung cancer (NSCLC):
results from Eastern Cooperative Oncology Group (ECOG)
study E5592 [abstract]. Proc Am Soc Clin Oncol 1997; 16:2a
89 Brundage MD, Feldman-Stewart D, Cosby R. Cancer pa-
tients attitudes toward treatment options for advanced non-
small cell lung cancer: implications for patient education and
decision support. Patient Educ Couns 2001; 45:149157
90 Silvestri G, Pritchard R, Welch HG. Preferences for chemo-
therapy in patients with advanced non-small cell lung cancer:
descriptive study based on scripted interviews. BMJ 1998;
317:771775
91 Slevin ML, Stubbs L, Plant HJ, et al. Attitudes to chemother-
apy: comparing views of patients with cancer with those of
doctors, nurses, and general public. BMJ 1990; 301:1458
1460
92 Sweeney C, Zhu J, Sandler A, et al. Outcome of patients with
a performance status of 2 in Eastern Cooperative Oncology
Group Study E1594: a phase III trial in patients with
metastatic non-small cell lung carcinoma. Cancer 2001; 92:
26352647
2003;123;226-243 Chest
Mark A. Socinski, David E. Morris, Gregory A. Masters and Rogerio Lilenbaum
Chemotherapeutic Management of Stage IV Non-small Cell Lung Cancer
& Services
Updated Information
S
References
S#BIBL
free at:
Reprints
2003;123;244-258 Chest
Frank C. Detterbeck, David R. Jones, Kemp H. Kernstine and Keith S. Naunheim
Special Treatment Issues
ISSN: 0012-3692.
Special Treatment Issues*
Frank C. Detterbeck, MD, FCCP; David R. Jones, MD, FCCP;
Kemp H. Kernstine, MD, PhD, FCCP; and Keith S. Naunheim, MD, FCCP
This chapter of the Lung Cancer Guidelines addresses patients with particular forms of non-small
cell lung cancer that require special considerations. This includes patients with Pancoast tumors,
T4N0,1M0 tumors, satellite nodules in the same lobe, synchronous and metachronous multiple
primary lung cancers (MPLC), and solitary metastases. For patients with a Pancoast tumor, a
multimodality approach, involving chemoradiotherapy and surgical resection, appears optimal
provided appropriate staging has been carried out. Patients with central T4 tumors that do not
have mediastinal node involvement are uncommon. When carefully staged and selected,
however, such patients appear to benefit from resection as part of the treatment as opposed to
chemoradiotherapy alone. Patients with a satellite lesion in the same lobe as the primary tumor
have a good prognosis and require no modification of the approach to evaluation and treatment
from what would be dictated by the primary tumor alone. On the other hand, it is difficult to know
how best to treat patients with a focus of the same type of cancer in a different lobe. Although
MPLC do occur, the survival results after resection for either a synchronous presentation or a
metachronous presentation with an interval of < 4 years between tumors are variable and
generally poor, suggesting that many of these patients may have had a pulmonary metastasis
rather than a second primary lung cancer. A thorough and careful evaluation of these patients is
warranted to try to differentiate between patients with a metastasis and those with a second
primary lung cancer, although criteria to distinguish them have not been defined. Finally, some
patients with a solitary focus of metastatic disease in the brain or adrenal gland appear to benefit
substantially from resection. (CHEST 2003; 123:244S258S)
Key words: adrenal metastasis; brain metastasis; carina; metachronous primary lung cancers; multiple primary lung
cancer; Pancoast tumor; satellite nodules; superior sulcus tumor; superior vena cava; synchronous primary lung cancers;
T4N0,1M0 tumor
Abbreviations: ACCP American College of Chest Physicians; MPLC multiple primary lung cancer;
NSCLC non-small cell lung cancer; PET positron emission tomography; WBRT whole brain radiotherapy
I
n general, patients with an early stage non-small
cell lung cancer (NSCLC) without mediastinal
nodal involvement (stage I and II) are treated pri-
marily with surgery, whereas those with a locally
advanced lung cancer with mediastinal nodal in-
volvement (stage IIIA and IIIB) are treated with
chemotherapy and radiation. However, there are
several relatively unusual presentations of NSCLC,
in which the anatomic and biological issues appear to
dictate a different approach. In addition, the pres-
ence of an isolated second focus of cancer in a
patient with lung cancer presents a situation where
the biology of this phenomenon is often not clear,
and therefore the approach to treatment is difficult.
This section addresses patients with particular
forms of NSCLC that require special considerations.
These include patients with Pancoast tumors,
T4N0,1M0 tumors, satellite nodules in the same
lobe, synchronous and metachronous multiple pri-
mary lung cancer (MPLC), and solitary metastases.
The scope of this project did not allow inclusion of
special histologic types of lung cancer, such as typical
and atypical carcinoid tumors, mucoepidermoid tu-
mors, or bronchioloalveolar carcinomas.
Methods
A formal meta-analysis was not available for any of the
particular forms of NSCLC that are the subject of this chapter,
and resources did not permit the American College of Chest
Physicians (ACCP) to conduct such an analysis independently.
Clinical guidelines from other organizations were available only
with regard to Pancoast tumors. These involve primarily consen-
*From the Multidisciplinary Thoracic Oncology Program, Divi-
sion of Cardiothoracic Surgery (Dr. Detterbeck), University of
North Carolina, Chapel Hill, NC; Division of Thoracic and
Cardiovascular Surgery (Dr. Jones), University of Virginia, Char-
lottesville, VA; Division of Cardiothoracic Surgery (Dr. Kerns-
tine), University of Iowa Hospitals and Clinics, Iowa City, IA;
Division of Cardiothoracic Surgery (Dr. Naunheim), St. Louis
University Health Sciences Center, St. Louis, MO.
The work for these guidelines was performed at the University of
North Carolina at Chapel Hill.
Correspondence to: Frank C. Detterbeck, MD, FCCP, Division of
Cardiothoracic Surgery, Department of Surgery, University of
North Carolina at Chapel Hill, CB #7065, 108 Burnett-Womack
Building, Chapel Hill, NC 27599-7065; e-mail: fdetter@med.
unc.edu
sus opinion statements and are discussed in the section on
Pancoast tumors.
15
However, a systematic review of literature in
each of these areas is available, published in the year 2001.
6
The
recommendations in this section rely heavily on the data from this
review.
The data regarding the approach to these special situations was
reviewed, summarized, and used to define management recom-
mendations by the writing committee. This document was then
reviewed by three independent reviewers, and further changes
were made. The revised document and recommendations were
further reviewed by the entire ACCP Guidelines committee to
assure that it met the requirements of a balanced, accurate, and
generally acceptable representation of the issues with regard to
these particular forms of NSCLC.
Pancoast Tumors
Definition
Lung cancers that occur in the apex of the chest
and invade apical chest wall structures are called
superior sulcus tumors or Pancoast tumors. The
classic description of such patients involves a syn-
drome of pain radiating down the arm as a manifes-
tation of brachial plexus involvement. With improve-
ments in radiographic techniques, earlier diagnosis,
and a more detailed understanding of the anatomy, a
tumor can be classified as a Pancoast tumor if it
invades any of the structures at the apex of the chest,
including the most superior ribs or periostium, the
lower nerve roots of the brachial plexus, the sympa-
thetic chain near the apex of the chest, or the
subclavian vessels. These tumors are now divided
into anterior, middle, and posterior compartment
tumors depending on the location of the chest wall
involvement in relation to the insertions of the
anterior and middle scalene muscles on the first rib.
7
A syndrome of pain radiating down the arm is no
longer a prerequisite for an apical tumor to be
designated a Pancoast tumor.
Workup
There are no data that specifically address the
reliability of the clinical examination in patients with
Pancoast tumors with regard to the presence of
distant metastases. In the absence of data to the
contrary, the panel thought that Pancoast tumors
should be treated like most other resectable lung
cancers, meaning that imaging tests for distant me-
tastases are not routinely necessary in the presence
of a negative clinical evaluation. There is also no
evidence regarding the reliability of CT or positron
emission tomography (PET) scans for mediastinal
node involvement specifically in patients with Pan-
coast tumors. The consensus of the panel was that, in
the face of normal-sized lymph nodes by CT, medi-
astinoscopy should be performed, although a nega-
tive PET scan in the mediastinum may obviate this.
In addition, mediastinoscopy should be performed in
the presence of enlarged or PET-positive lymph
nodes. The argument for this approach to intratho-
racic staging is that it is consistent with the general
recommendation for accurate staging before initia-
tion of a major intervention, such as resection, and
consistent with data demonstrating that N2,3 node
involvement is a major negative prognostic factor.
No firm recommendation can be made about
whether mediastinoscopy should be done before or
after preoperative therapy. An MRI demonstrates
involvement of apical chest wall structures better
than a CT scan,
8
but CT provides more information
about the presence of nodal enlargement, and pul-
monary, hepatic, and adrenal metastases. Therefore,
both a chest CT and an MRI are indicated to assess
the resectability of a Pancoast tumor.
Treatment
The classic approach to curative treatment of
Pancoast tumors has been preoperative radiotherapy
followed by surgical resection. This dates back to an
experience published in 1961 by Shaw et al,
9
in
which 12 of 18 patients treated with this approach
were still alive at the time the paper was written.
However, the follow-up was 2 years in 90% of the
patients.
9
Alternatives are treatment with radiation
alone, preoperative chemoradiotherapy and resec-
tion, or chemoradiotherapy without resection.
Treatment with radiation alone has achieved good
palliation of pain in approximately 75% of patients.
10
In general, very few patients treated with radiation
alone are long-term survivors (approximately 5%).
11
However, many of these series have included pa-
tients with advanced stage tumors. Among studies
that have involved primarily patients who had a
reasonable chance of cure, the average median sur-
vival time was 16 months, and the average 5-year
survival was 20% (range 1523%).
10,1214
Treatment with preoperative radiation and resec-
tion has resulted in an average median survival time
of 22 months and a 5-year survival of 27%.
11
In these
series, approximately one third of patients under-
went an incomplete (R
1
or R
2
) resection, and approx-
imately one third of the resections involved only a
limited resection of the affected lobe of the lung.
11
Retrospective analysis has found that a complete
resection with negative margins (R
0
) and a pulmo-
nary resection involving at least a lobectomy are
major factors associated with better survival.
15
Fur-
thermore, N2,3 lymph node involvement is a major
negative prognostic factor and should generally be
considered a contraindication to surgery.
11
Patients
with vertebral body or subclavian vessel involvement
have traditionally been considered to be unresect-
able, but it appears that, with improved surgical
approaches to these structures, a few experienced
centers have been able to achieve reasonable survival
in such patients.
16,17
The presence of Horner syn-
drome is also associated with poor survival.
11
A large phase II study of preoperative chemora-
diotherapy in patients with Pancoast tumors has
shown a complete resection rate of 92% and a good
2-year survival rate compared with historical controls
of radiotherapy followed by surgery.
18
Furthermore,
local recurrences were seen in only 33% of those
patients with a recurrence, whereas in series involv-
ing preoperative radiotherapy alone, the majority of
recurrences involved the tumor bed. These data, in
combination with the data for non-Pancoast stage III
NSCLC, suggest that preoperative chemoradiother-
apy is a significant improvement over preoperative
radiotherapy, particularly in light of the fact that
there are insufficient numbers of patients with Pan-
coast tumors to complete a randomized comparison.
Other published guidelines have recommended
that patients with Pancoast tumors should be evalu-
ated by a thoracic surgeon.
3
If there is no evidence of
mediastinal node involvement
1
or extensive local
invasion,
5
patients should undergo resection in
combination with radiotherapy or chemoradio-
therapy.
1,4,5
Patients with inoperable Pancoast tu-
mors should be treated with radiotherapy.
2
The
latter two recommendations were rated grade B,
whereas the strengths of the other statements were
rated grade C. Thus other guidelines have reached
the same conclusions as this ACCP document, al-
though the recommendations in those other docu-
ments have been less detailed and more vaguely
worded.
In summary, the available data suggests that the
best survival is achieved by preoperative chemora-
diotherapy followed by surgical resection in carefully
selected patients. Preoperative radiotherapy fol-
lowed by surgical resection is a reasonable alterna-
tive. Involvement of subclavian vessels or the verte-
bral column is associated with poor survival after
resection. However, a few centers have gained expe-
rience with improved surgical approaches to these
structures and have reported reasonable survival
rates after resection. Involvement of mediastinal
nodes is associated with poor survival after resection.
At the time of resection, it is important to carry out
a complete resection, which should involve at least a
lobectomy. There is no data on how unresectable, yet
potentially curable, patients with Pancoast tumors
should be treated. However, extrapolation from the
data for non-Pancoast stage III NSCLC suggests that
chemoradiotherapy is the best approach. For pa-
tients in whom cure is not believed possible, radio-
therapy offers good palliation of pain.
Recommendations: Pancoast tumor
1. For patients with a Pancoast tumor, a tissue
diagnosis should be obtained prior to the initi-
ation of therapy. Level of evidence, poor; ben-
efit, substantial; grade of recommendation, C
2. Patients with a Pancoast tumor without evi-
dence of mediastinal node involvement or dis-
tant metastases should be evaluated by an
experienced thoracic surgeon for potential re-
section. Level of evidence, fair; benefit, sub-
3. Patients with a Pancoast tumor being consid-
ered for resection should undergo evaluation
with an MRI of the thoracic inlet and brachial
plexus, in addition to a CT of the chest. Level of
recommendation, B
4. Resection of patients with a Pancoast tumor
with involvement of the subclavian vessels or
the vertebral column should not be routinely
undertaken (outside of specialized centers).
Level of evidence, poor; benefit, moderate;
ered for curative resection should undergo a
cervical mediastinoscopy. Involvement of me-
diastinal nodes represents a contraindication to
resection. Level of evidence, good; benefit,
6. Patients with a potentially resectable, nonmeta-
static Pancoast tumor (and good performance
status) should undergo preoperative chemoradio-
therapy prior to resection. A reasonable alterna-
tive for such patients is preoperative radiother-
apy. Level of evidence, fair; benefit, moderate;
7. At the time of resection of a Pancoast tumor,
every effort should be made to achieve a
complete resection. Level of evidence, good;
tion, A
8. Resection of a Pancoast tumor should consist of
a lobectomy (instead of a wedge), as well as
removal of the involved chest wall structures.
9. For patients with a Pancoast tumor, postoper-
ative radiotherapy is not recommended, in
either completely or incompletely resected
patients, because of lack of a demonstrated
survival benefit. Level of evidence, poor; ben-
efit, none; grade of recommendation, D
10. Patients with a good performance status and
an unresectable but nonmetastatic Pancoast
tumor should be considered for combination
chemotherapy and radiotherapy with intent to
cure. Level of evidence, poor; benefit, mod-
erate; grade of recommendation, C
11. Palliative radiotherapy should be considered
in patients who are not candidates for treat-
ment with curative intent (ie, surgery, chemo-
radiotherapy etc.). Level of evidence, fair;
benefit, moderate; grade of recommenda-
tion, B
T4N0,1M0 Tumors
Patient Selection and Workup
Most patients with involvement of T4 structures
have mediastinal node involvement as well. These
patients should be treated with chemoradiotherapy,
as is generally recommended for patients with stage
IIIB NSCLC. However, selected patients with T4
involvement, but without mediastinal node involve-
ment, can be viewed as candidates for surgery.
Although many reports have demonstrated the tech-
nical feasibility of resection of T4 structures, fewer
series have provided long-term survival data. The
largest experience of resection for T4 involvement
involved carinal resections, usually together with a
right pneumonectomy (327 patients in whom long-
term survival data were reported).
19
A moderate
experience is available with left atrial involvement
(85 patients), and a smaller experience has been
reported with involvement of the superior vena cava
(40 patients).
19
The fact that so few patients have
been reported with long-term survival statistics un-
derscores the fact that patients who are candidates
for a surgical approach are extremely rare and highly
selected.
A mediastinoscopy should be performed even if a
CT suggests no N2,3 involvement in patients with T4
tumors being considered for a surgical approach.
This argument is based on the fact that CT evalua-
tion of the mediastinum in central tumors has a high
false negative rate. Furthermore, the operative mor-
tality and the poor survival of T4N2,3M0 patients
argues that surgical resection be undertaken only in
carefully selected patients. For patients who are
being considered for carinal resection, it may be best
to perform mediastinoscopy at the same time as
resection, in order to prevent scarring and therefore
lack of mobility of the airways at the time of recon-
struction.
Outcomes After Surgery
In a fairly large series from Japan involving an
aggressive approach to T4 tumors, approximately
one third of patients were able to undergo complete
(R
0
) resection, one third a microscopically incom-
plete resection (R
1
), and one third a grossly incom-
plete resection (R
2
).
20
The 5-year survival rates for
these groups were 22%, 18%, and 0%, respectively.
20
No data are available regarding how often a resection
can actually be carried out among patients with
involvement of specific T4 structures.
The data regarding the outcome after resection in
patients with carinal involvement shows an average
5-year survival of 26%. However, the survival comes
at a price of an average operative mortality of 18%
(range 729%). It should be noted, however, that the
survival statistics have included all operative deaths
as well. The fact that the best reported 5-year
survival (42%) comes from the largest series,
21
which
also reported an operative mortality of only 7%, can
be interpreted to suggest that such resections should
be undertaken only in experienced centers. Survival
data for resections involving other T4 structures have
involved fewer patients, making interpretation of the
data difficult (Table 1). The survival of patients with
left atrial involvement has been poor. In general,
however, the survival of patients with involvement of
other T4 structures has been similar to that reported
for patients with carinal involvement.
Patients with involvement of T4 structures should
Table 1Results of Resection of Patients With T4 Involvement From NSCLC*
Structure
Studies
(No.)
No. of
Patients
Hospital
Mortality
5-yr Survival (%)
Average Highest Lowest
Any 1 101 13 13 23 (R
0
) 0 (R
2
)
Carina 8 327 18 26 42 13
Left atrium 5 85 8 8 22 0
SVC 3 40 8 23 31 15
Vertebral bodies 2 29 0 42 nd nd
Aorta 2 27 0 nd nd nd
Esophagus 1 7 nd 14 nd nd
Main PA 1 7 nd 0 nd nd
*nd, no data; PA, pulmonary artery; R
0
, complete resection; R
2
, incomplete resection with gross residual disease; SVC superior vena cava.
be very carefully selected before undertaking surgi-
cal resection, because of the limited survival and the
high mortality. This means that these patients should
have a high likelihood of being able to tolerate a
major operation from a general medical standpoint.
This also means that the evaluations to rule out
either mediastinal or extrathoracic metastases should
be especially thorough and that the threshold for
pursuing subtle abnormalities seen on imaging tests
should be low.
Preoperative chemotherapy or chemoradiotherapy
in patients with T4 tumors has been reported in
several trials. A 5-year survival of 20% was reported
among all patients in the largest trial (57 patients;
62% of whom underwent complete resection).
22
These results are encouraging, given that 60% of the
patients entered in the study had T4N2M0 tumors
by careful surgical staging. By comparison, 5-year
survival results for chemoradiotherapy without sur-
gery in patients with stage IIIA,B tumors have been
approximately 9% and 14% in large randomized
trials involving sequential or concurrent chemoradio-
therapy, respectively.
23
However, these latter series
have included both stage IIIA and IIIB patients and
have not reported data separately or reported any
data specifically in patients with T4N0,1M0 tumors.
A retrospective analysis of the Southwest Oncology
Group experience suggested that patients with
T4N0,1M0 tumors benefited from preoperative che-
moradiotherapy and surgery compared with treat-
ment with chemoradiotherapy alone (2-year survival
of 64% vs 33%).
24
Recommendations: T4N0,1M0 tumors
12. Patients with a clinical T4N0,1M0 NSCLC
should be carefully evaluated (with imaging
studies) for distant metastatic disease prior to
considering surgical resection. Level of evi-
ommendation, B
13. Resection of T4N0,1M0 tumors in selected
patients may result in better survival than
chemoradiotherapy without resection. Level
of evidence, poor; benefit, moderate; grade of
recommendation, C
14. Mediastinoscopy should be done prior to sur-
gical resection of patients with clinical
T4N0,1M0 tumors. Level of evidence, fair;
tion, B
Satellite Nodules and MPLC
Definitions
Occasionally, patients present with more than one
focus of cancer within the lung. The American Joint
Committee on Cancer staging system classifies a
second focus of cancer within the same lobe as T4,
whereas a second focus in another lobe is classified
as M1. However, the classification does not help in
grouping tumors according to similar biological situ-
ations. Although the secondary focus may represent
a hematogenously spread metastasis, it may also be a
second primary lung cancer or a second focus that is
a manifestation of local spread. Distinguishing these
situations is difficult. In this section, these tumors
are classified according to clinical presentation, a
method that at least has practical relevance in defin-
ing an approach to these patients. Thus, this section
will distinguish a synchronous lesion within the same
lobe as the primary tumor, two synchronous foci of
cancer in different lobes, and two metachronous foci
of cancer in the lung. Circumstances can be identi-
fied for each of these clinical presentations to rea-
sonably allow them to be defined as satellite lesions,
or synchronous and metachronous MPLC. In this
document, as in the published literature, a satellite
lesion is any additional focus of lung cancer of the
same histologic type within the same lobe, regardless
of the relative size or location in different segments,
and regardless of whether it is discovered by the
radiologist, the surgeon, or the pathologist.
Definitions for satellite lesions within the same
lobe as the primary tumor, synchronous second
primary lung cancers, and metachronous second
primary lung cancers are given in Table 2. In
general, these criteria are relatively well accepted,
but some authors have varied slightly in some details
(eg, the minimum interval between metachronous
MPLC). Many data are available regarding the inci-
dence of a second primary lung cancer and the
Table 2Definition of Satellite Nodules, MPLC, and
Pulmonary Metastases
Satellite nodules
Satellite nodules from primary tumor
Same histology and same lobe as primary cancer and no
systemic metastases
Multiple primary lung cancers
Same histology, anatomically separated
Cancers in different lobes and no N2,3 involvement and no
systemic metastases
Same histology, temporally separated
4-yr interval between cancers and no systemic metastases
from either cancer
Different histology
Different histologic type or different molecular genetic
characteristics or arising separately from foci of carcinoma
in situ
Hematogenously spread pulmonary metastases
Same histology and multiple systemic metastases
Same histology, in different lobes
and presence of N2,3 involvement, or 2-yr interval
recurrence rates and patterns of resected lung can-
cer. Therefore, the incidence of a second primary
cancer and the incidence of a solitary pulmonary
metastasis can be estimated for different stages of
the primary lung cancer and by location of the
second focus of cancer, as shown in Figure 1.
Although such estimates are based on extrapolations
from known data, the resulting incidences and dis-
tributions between synchronous and metachronous
presentations or same histology and different histo-
logic types are both internally consistent and very
close to what is actually observed. Analysis of these
rates suggests that the biological situation (ie, new
primary vs locally or hematogenously spread metas-
tasis) can be defined clearly in some clinical presen-
tations (ie, satellite lesions, MPLC of different his-
tologic types, and metachronous tumors with an
interval of 4 years). In other clinical presenta-
tions, the biological situation is very unclear.
Small pulmonary lesions are frequently seen in
addition to the primary tumor on the chest CT. This
occurred in 16% of clinical stage I-IIIA, potentially
operable patients with NSCLC in one large study.
25
The lesions were not calcified, and they ranged from
4 to 12 mm. A definitive diagnosis (biopsy or
follow-up of 24 months) was established in only
20% of the patients, the remainder being lost to
follow-up or having unavailable pathology reports.
Of the lesions for which a definitive diagnosis was
available, 86% were found to be benign. In another
study, 10% of patients had a second lesion detected
preoperatively, of which nearly 60% were found to be
benign.
26
Therefore, a patient should not be denied a
curative approach on the basis of a second pulmonary
nodule without a definitive tissue diagnosis.
In the following paragraphs, a prospective ap-
proach is formulated for patients with clinical stage
I-III NSCLC in whom a second intraparenchymal
focus of cancer is not only identified radiographically
but is also proven to be malignant by cytologic
studies. Patients with disseminated disease (extratho-
racic metastases) are excluded. In addition, the 30%
of patients with synchronous MPLC, in which the
second cancer was found incidentally at thoracot-
omy, are excluded for obvious reasons. Patients with
bronchioloalveolar carcinoma should also be consid-
ered separately. Finally, it must be emphasized that
the majority (5786%) of additional nodules seen
radiographically in patients with clinical stage I-III
NSCLC are benign lesions.
25,26
Therefore, the con-
siderations noted in the following discussion are
relevant only when a histologic diagnosis of a MPLC
has been made.
Satellite Nodules of Cancer in the
Same Lobe
Workup and Treatment Results
Studies that have reported on long-term survival
specifically of patients with satellite nodules in the
same lobe as the primary tumor have generally
reported good survival. The overall 5-year survival
rate of all patients, of whom approximately 60% have
N1 or N2 involvement, is 34%.
27
The 5-year survival
for patients with satellite nodules and no node
involvement is 64% (range, 54 to 70%), which is
similar to the survival for patients with stage I
NSCLC without satellite nodules.
27
Direct compar-
isons have generally demonstrated a slightly inferior
survival in patients with satellite nodules, stage for
stage, compared with patients without satellite nod-
ules.
28
Nevertheless, the survival of patients with
satellite nodules in the same lobe is consistently
higher than that reported for patients with a second
cancer nodule in a separate lobe (5-year survival
approximately 10%, range, 0 to 23% for all pa-
tients).
27
In general, no additional diagnostic workup is
necessary in patients with a secondary lesion in the
same lobe. First of all, the available data indicates
that most secondary lesions in the same lobe as the
primary tumor were found to be benign. Further-
more, the prognosis in those patients who are found
to have a satellite nodule of cancer is only slightly
inferior to those without a satellite focus, which
argues that resection should be undertaken even in
those patients who do, in fact, have a satellite focus
of cancer. Therefore, there is little reason to attempt
to definitively diagnose a second lesion preopera-
tively in patients with clinical stage I,II tumors with
a second radiographic nodule in the same lobe.
Figure 1. Estimated incidence of MPLC, solitary pulmonary
metastases, and satellite lesions in different clinical presentations.
These estimates are based on data concerning recurrence rates
by stage and time interval, location of metastases, and the
observed incidence of MPLC and satellite lesions for each
clinical presentation. Pulm Metspulmonary metastasis.
Adapted with permission from Detterbeck et al.
27
Furthermore, there is little reason to perform any
additional preoperative staging investigations (eg,
mediastinoscopy, CT of the head, bone scan) in
patients with a second nodule in the same lobe as the
primary tumor, other than what is dictated by the
patients clinical status and the primary tumor.
Recommendations: Satellite Nodule in the
Same Lobe
15. No further diagnostic workup of a satellite
nodule is needed in patients with suspected or
proven lung cancer and a satellite nodule
within the same lobe. Level of evidence, fair;
tion, B
16. For patients with a satellite lesion within the
same lobe as a suspected or proven primary
lung cancer, distant organ scanning and con-
firmation of the mediastinal node status
should be carried out as dictated by the
primary lung cancer alone, and not modified
due to the presence of the satellite lesion.
17. For patients with NSCLC and a satellite focus
of cancer within the same lobe, resection via a
lobectomy is the preferred treatment. Level of
recommendation, B
Synchronous Second Primary Lung Cancer
Definition
A synchronous second focus of lung cancer in a
different lobe is easily defined as a second primary
lung cancer when the two sites are of different
histologic types. Cancers may also be distinguished
on the basis of different molecular genetic charac-
teristics. In the absence of molecular analysis, it is
difficult to distinguish two synchronous cancers that
are of the same histologic type as separate primary
lung cancers. One proposed requirement for classi-
fication as synchronous second primary lung cancers
is that there be no mediastinal node involvement and
no sites of distant metastases when the two cancers
are of the same histologic type.
27
It can be estimated
that the incidence of a second primary cancer using
this definition is slightly higher than the incidence of
a isolated pulmonary metastasis, given what is known
about the incidence of MPLC and the rate and sites
of spread of lung cancer.
27
On the other hand, when
mediastinal node involvement is present, the inci-
dence of an isolated pulmonary metastasis is higher
than that of a second primary cancer.
27
Although the
exact incidence of multiple primary cancers and
isolated pulmonary metastasis may not be fully de-
fined by these estimates, at the very least it is clear
that the identification of two synchronous foci of
cancer of the same histologic type is difficult.
Patient Selection and Treatment Results
The survival of patients with synchronous (differ-
ent lobe) MPLC (either same or different histologic
types) is highly variable, consistent with the difficulty
of reliably classifying these tumors.
27
The 5-year
survival for all patients ranges from 0 to 70%, and the
survival of patients in whom both tumors are classi-
fied as stage I ranges from 0 to 79%.
2932
This data
suggests that a great deal of caution is necessary in
classifying two synchronous foci of cancer as two
separate primary lung cancers. Approximately one
third of the second foci of cancer are found inciden-
tally at the time of resection.
27
Approximately 60% of
synchronous second primary lung cancers are squa-
mous cell cancers, and in approximately 60% of the
cases, the tumors are of the same histologic type.
27
The first issue to consider in approaching patients
with a synchronous second focus of lung cancer in a
different lobe is the accuracy of the diagnosis. If two
histologic types of primary NSCLC are diagnosed
preoperatively, it must be remembered that the
accuracy of determining lung cancer cell type by
cytologic studies is only 60 to 80%.
3336
A histologic
or core needle diagnosis should be obtained, espe-
cially if there is evidence of mediastinal lymph node
involvement, because this increases the probability
that the second focus is an isolated pulmonary
metastasis. Even when a diagnosis of synchronous
second primary lung cancers is secure, careful stag-
ing with distant organ scanning and mediastinoscopy
should be carried out because the survival of patients
with synchronous MPLC is poor, even in patients
who have cancers of different histologic types.
37
Patients with a synchronous second cancer of
similar histologic type present a conundrum. These
patients should undergo an extensive search for
mediastinal involvement, distant metastases, or an
extrapulmonary primary cancer. Genetic marker
analysis may be useful in distinguishing between
MPLC and a metastasis. In the absence of distant
metastases, lymph node involvement, or evidence
that the second focus of cancer is a metastasis,
resection is reasonable, although the reported long-
term survival is generally poor.
Occasionally, patients not suspected of having a
second primary cancer are found intraoperatively
to have a second cancer. It is usually difficult to
determine whether the histologic type of the two
cancers is the same or different on frozen section
examination. There are no published data that
address this specific situation. The panel believes
it is reasonable to proceed with a resection of each
lesion if each appears to be a resectable primary
lung cancer, given that the patient has already
been exposed to the morbidity of a thoracotomy.
However, this can only be recommended if the
patient has adequate pulmonary reserve to tolerate
the resection, if there is no mediastinal nodal
involvement, and if there is no clinical evidence of
distant metastases. Concerns about the adequacy
of pulmonary reserve may make it necessary to
perform a limited resection (segmentectomy or
wedge) of one or both of the lesions. Nevertheless,
the resection must be a complete resection (R
0
); if
this cannot be achieved, then nothing more than a
biopsy of the lesions for diagnosis is indicated. The
prognosis after resection in such situations has not
been defined, but is likely to be poor, similar to
the survival of patients with synchronous primary
lung cancers that are recognized or at least sus-
pected preoperatively.
Recommendations: Synchronous Second Primary
Lung Cancer
18. Patients suspected of having two synchronous
primary lung cancers should have a thoughtful
search for an extrathoracic primary cancer to
rule out the possibility that both of the lung
lesions represent metastases from an extratho-
racic primary. Level of evidence, poor; bene-
19. For patients suspected of having two synchro-
nous primary lung cancers, a careful and
thorough search for distant metastases should
be performed. Level of evidence, poor; bene-
nous primary lung cancers, the absence of
mediastinal node involvement should be con-
firmed (usually via mediastinoscopy) prior to
21. For patients with two synchronous primary
lung cancers, resection of both lung cancers is
reasonable provided a careful search for dis-
tant metastases or mediastinal lymph node
involvement has been carried out and is neg-
ative. Level of evidence, poor; benefit, mod-
22. If a patient (not suspected of having a second
focus of cancer) is discovered intraoperatively
to have a second cancer in a different lobe, it
is recommended that a resection of each
lesion be undertaken, provided the patient has
adequate pulmonary reserve and provided
there is no N2 nodal involvement. Level of
evidence, poor; benefit, moderate; C
Metachronous Second Primary
Lung Cancer
Definition
A metachronous second focus of lung cancer is
easily defined as a second primary lung cancer when
the two tumors are different histologic types. When
they are of the same type, the second focus can be
reliably defined as a second primary if there is no
evidence of systemic metastases and at least a 4-year
interval between the two.
27
Some authors have in-
cluded patients with 2-year interval,
38
but the
estimated incidence of a solitary pulmonary metas-
tasis from the previous lung cancer is practically the
same as the estimated incidence of a new primary
lung cancer.
27
Therefore, an interval of 2 to 4 years
represents a gray area, where it is difficult to deter-
mine whether a new lesion is a second primary. If the
interval is 2 years, it is much more likely that the
lesion is a metastasis from the original cancer than a
second primary lung cancer.
Among studies reporting on metachronous second
primary lung cancers, approximately two thirds of these
have been tumors of the same histologic type (most
often squamous cell).
27
The average time interval be-
tween tumors in these studies is 48 months. Approxi-
mately 80% of second primary lung cancers are found
on a routine chest radiograph, and approximately 75%
are stage I.
27
Approximately 65% of second primary
lung cancers are able to be resected, with approxi-
mately one third of the resections involving a limited
resection. The operative mortality for the resection has
been reported to average 7%.
27
The 5-year survival of
all patients who present with a second primary is
approximately 20%.
30,32,39,40
The survival of patients
able to undergo resection of the second primary is
36%.
3032,37,4144
The survival of patients who are
found to have a second primary lung cancer that
is pathology stage I is also only 36% (range 20
50%).
30,31,37,40,41,43
A careful search for sites of recurrence should be
conducted in patients who present with a nodule that
is suspected of being a metachronous second pri-
mary lung cancer. This is particularly important if
the histologic type is the same as the primary cancer
and if the interval between cancers has been
4 years. A new cancer appearing in 2 years
should be assumed to be a metastasis unless it is
clearly of a different histologic type. Although some
cancers appearing between 2 and 4 years after the
first primary lung cancer are probably MPLC, a fair
amount of doubt about this exists until the interval
has been 4 years. Resection of a second primary
lung cancer that is early stage should be undertaken,
although the prognosis is not as good as that of an
early stage single primary lung cancer.
Recommendations: Metachronous Second Primary
Lung Cancer
23. A careful and thorough search for distant
metastases should be performed in patients
suspected of having metachronous second pri-
mary lung cancers. Level of evidence, poor;
tion, C
24. Resection of a metachronous second primary
lung cancer is reasonable provided a careful
search for other distant metastases or medias-
tinal lymph node involvement has been car-
ried out and is negative. Level of evidence,
dation, C
Isolated Brain Metastasis
Patient Selection and Workup
Approximately 25% of patients with stage IV
NSCLC have a brain metastasis as well as other sites
of metastatic disease.
45
The median survival of pa-
tients with a brain metastasis is approximately
2 months when treated with steroids alone, and 3 to
6 months when treated with whole brain radiother-
apy (WBRT).
45
Because the survival of patients with
a brain metastasis is so short, there is reason to
consider aggressive treatment of the brain metastasis
with either surgical resection or radiosurgery as a
palliative treatment to prolong survival. However, a
subset of patients with stage IV disease has a brain
metastasis as the only site of metastatic disease. In
this group, it is reasonable to consider aggressive
therapy of both the primary lesion as well as the
isolated metastatic site as a potentially curative ther-
apy. This latter group is the focus of this section.
Patients with a brain metastasis who are treated with
surgery or radiosurgery of the brain metastasis as a
palliative treatment are discussed in the chapter on
palliative care.
Aggressive treatment of a brain metastasis may
involve either surgical resection of the metastasis or
ablation of the metastasis by radiosurgery. This latter
technique involves a precisely focused beam of
radiation with a steep fall-off of the dose outside of
the target area, hence the name radiosurgery. Al-
though no randomized trial of surgery vs radiosur-
gery has ever been completed, comparison of the
results of these techniques in patients treated pallia-
tively suggests that they are similar with regard to
survival, local control, morbidity, and mortality.
46,47
A number of technical issues often favor one of these
treatments over the other; therefore, they are best
viewed as complimentary modalities. In the discus-
sion in this section, they will be considered together
as similar methods of aggressive treatment of a brain
metastasis.
Patients with a brain metastasis should be selected
for curative treatment only after a thorough search
for other sites of disease has been negative. Further-
more, it is obvious that only patients in whom both
the brain metastasis and the primary tumor can be
completely resected can be considered candidates
for curative treatment (synchronous presentation). It
appears reasonable to assume that patients with N2,3
involvement and a brain metastasis are not good
candidates for curative therapy, although data dem-
onstrating this is lacking.
47
Therefore, it appears
reasonable to perform mediastinoscopy in selecting
patients for resection of the brain metastasis and the
primary lesion. The histologic subtype does not play
a role.
47
The number of brain metastases may not
play a role as long as the number is small ( 3), and
they can all be completely resected (as has been
demonstrated by several retrospective studies in
patients treated for palliation).
4851
The outlook is likely to be more optimistic for
patients who are younger, female, or have a meta-
chronous presentation.
47
The outlook may also be
better in patients with supratentorial lesions and
those with a brain metastasis 3 cm in diameter.
However, these considerations are relative and
should not necessarily exclude patients who are
otherwise fit and in whom a complete resection is
likely to be achieved.
Treatment Outcomes
Survival statistics of patients with a brain metasta-
sis who were treated with curative intent have been
reported by a number of studies.
47
The overall
survival for all patients is fairly consistent, and
averages 14% (range, 8 to 21%). The 5-year survival
for completely resected patients averages 21%
(range, 16 to 30%).
47
The operative mortality in
these studies has been low, averaging 2%.
47
Approx-
imately two thirds of the cases involved a metachro-
nous presentation.
47
There are conflicting data regarding the role of
adjuvant WBRT following resection of an isolated
brain metastasis. Retrospective analyses of patients
who were primarily treated with curative intent have
suggested either no survival benefit
52
or a significant
benefit.
53
The rate of intracranial recurrence among
patients treated primarily with palliative intent was
lower after WBRT in a randomized study,
54
while
retrospective analyses in such patients have shown
conflicting results.
47
It is likely that a benefit, might
only be seen in patients without other sites of
metastases, given the experience with prophylactic
cranial irradiation in patients with small cell lung
cancer. There is no data regarding the role of
adjuvant chemotherapy for patients who have under-
gone curative resection of a brain metastasis.
Recommendations: Isolated Brain Metastasis
25. Patients with an isolated brain metastasis from
NSCLC should be considered for a curative
approach. Level of evidence, poor; benefit,
26. For patients with an isolated brain metastasis
from NSCLC who are being considered for a
curative approach, a careful search for other
distant metastases should be carried out with
imaging tests. Level of evidence, poor; bene-
27. For patients with a synchronous presentation
of isolated brain metastases and a resectable
primary lung cancer, mediastinoscopy should
be done to rule out N2,3 involvement prior to
curative resection. Level of evidence, poor;
tion, C
28. For patients with no other sites of metastases
and a synchronous resectable N0,1 primary
NSCLC, resection or radiosurgical ablation of
an isolated brain metastasis should be under-
taken (as well as resection of the primary
tumor). Level of evidence, fair; benefit, sub-
and a previously completely resected primary
NSCLC (metachronous presentation), resec-
tion or radiosurgical ablation of an isolated
brain metastasis should be undertaken. Level
recommendation, B
30. For patients who have undergone a curative
resection of an isolated brain metastasis, ad-
juvant whole brain radiotherapy is reasonable,
although there is conflicting and insufficient
data regarding a benefit with respect to sur-
vival or the rate of recurrent brain metastases.
Level of evidence, poor; benefit, unclear;
resection of an isolated brain metastasis (and
resection of the primary tumor), adjuvant
chemotherapy can be neither recommended
nor recommended against because of insuffi-
cient data regarding this issue. Level of evi-
dence, poor; benefit, unclear; grade of recom-
mendation, I
Isolated Adrenal Metastasis
Highly selected patients have been reported who
have undergone resection of an adrenal metastasis
from NSCLC with intent to cure. The overall 5-year
survival for these patients has been 10 to 23%.
47,55,56
Survival after resection of the primary and the
adrenal metastasis appears to be good primarily in
patients without nodal involvement.
47,56
Other fac-
tors such as the histologic type, synchronous vs
metachronous presentation, and ipsilateral vs con-
tralateral location do not have prognostic value in the
limited number of reported patients who underwent
this treatment.
47,55,56
Recommendations: Isolated Adrenal Metastasis
32. Patients with an isolated adrenal metastasis
from NSCLC should be considered for a
curative approach. Level of evidence, poor;
tion, C
33. For patients with an isolated adrenal metasta-
sis being considered for curative therapy, a
careful search for other distant metastases
should be carried out with imaging tests.
of an isolated adrenal metastasis and a resect-
able primary lung cancer, mediastinoscopy
should be done to rule out N2,3 involvement
prior to resection. Level of evidence, poor;
tion, C
35. For carefully selected patients with no other
sites of metastases and a synchronous resect-
able N0,1 primary NSCLC, resection of an
isolated adrenal metastasis from NSCLC
should be undertaken (as well as resection of
the primary tumor). Level of evidence, poor;
tion, C
tion of an isolated adrenal metastasis should
be undertaken. Level of evidence, poor; ben-
efit, moderate; grade of recommendation, C
Summary
The available data for patients with Pancoast
tumors suggests that the best survival is achieved by
preoperative chemoradiotherapy followed by surgi-
cal resection in carefully selected patients. Preoper-
ative radiotherapy followed by surgical resection is a
reasonable alternative. Involvement of subclavian
vessels, vertebral column, or mediastinal lymph
nodes is associated with poor survival after resection.
At the time of resection, it is important to carry out
a complete resection that should involve at least a
lobectomy. There is no data on how unresectable, yet
still potentially curable, patients with Pancoast tu-
mors should be treated. However, extrapolation from
the data for non-Pancoast stage III NSCLC suggests
that chemoradiotherapy is the best approach. For
patients in whom cure is not believed to be possible,
radiotherapy offers good palliation of pain.
Although most patients with T4 NSCLC have
N2,3 or M1 involvement, surgical resection should
be pursued in highly selected patients with
T4N0,1M0 tumors. The survival of such patients in
whom a complete resection was achieved appears to
be better than after treatment with chemoradiother-
apy alone. However, the operative mortality is rela-
tively high, and patients must be carefully staged and
selected. Preoperative chemoradiotherapy may also
be beneficial.
An additional small pulmonary nodule is not an
infrequent finding on a CT scan in patients with a
NSCLC. Most of these lesions are benign. If the
lesion is within the same lobe as the lung cancer, no
special workup is necessary other than what would
usually be done, because lobectomy is associated
with good survival, even when a second focus of
cancer is present (satellite lesion). If a second lesion
in another lobe is suspected of being malignant, it is
difficult to define whether this represents a synchro-
nous second primary lung cancer or a manifestation
of systemic disease. The patient should undergo a
thorough investigation for evidence of metastatic
disease before making a decision regarding treat-
ment. The prognosis and whether resection should
be undertaken is difficult to define when two lesions
of the same histologic type are present in different
lobes. Resection of both lesions may be appropriate,
but the prognosis is likely to be much worse than for
similar staged isolated primary lung cancers.
A careful search for sites of recurrence should be
conducted in patients who present with a nodule that
is suspected to be a metachronous second primary
lung cancer. This is particularly important if the
histologic type is the same as the primary cancer and
if the interval between cancers has been 4 years. A
new cancer appearing in 2 years should be as-
sumed to be a metastasis unless it is clearly of a
different histologic type. Although some cancers
appearing between 2 and 4 years after the first
primary lung cancer may be MPLC, a fair amount of
doubt about this exists until the interval has been
4 years. Resection of an early stage second pri-
mary lung cancer should be undertaken, although
the prognosis is not as good as that for an early-stage
single primary lung cancer.
Patients who have previously undergone com-
plete resection of the primary tumor but are
subsequently found to have a solitary cranial or
adrenal metastasis should be evaluated for resec-
tion of the metastasis with curative intent. In
addition, patients who present with a resectable
primary lung cancer and a solitary metastasis to
the brain and possibly also to the adrenal gland
should be evaluated for possible resection of both
lesions with curative intent. It is necessary to
perform a careful search for other sites of metas-
tases, and patients with mediastinal node involve-
ment should be excluded from such an approach.
Five-year survival rates of 15 to 20% have consis-
tently been reported in patients who have under-
gone resection of a solitary metastasis (as well as
resection of the primary tumor).
Pancoast Tumors
1. For patients with a Pancoast tumor, a tissue
diagnosis should be obtained prior to the
initiation of therapy. Level of evidence, poor;
tion, C
2. Patients with a Pancoast tumor without evi-
dence of mediastinal node involvement or
distant metastases should be evaluated by an
experienced thoracic surgeon for potential
resection. Level of evidence, fair; benefit,
ered for resection should undergo evaluation
with an MRI of the thoracic inlet and brachial
plexus, in addition to a CT of the chest. Level
recommendation, B
4. Resection of patients with a Pancoast tumor
with involvement of the subclavian vessels or
the vertebral column should not be routinely
undertaken (outside of specialized centers).
Level of evidence, poor; benefit, moderate;
ered for curative resection should undergo a
cervical mediastinoscopy. Involvement of me-
diastinal nodes represents a contraindication
to resection. Level of evidence, good; benefit,
6. Patients with a potentially resectable, nonmeta-
static Pancoast tumor (and good performance
status) should undergo preoperative chemora-
diotherapy prior to resection. A reasonable al-
ternative for such patients is preoperative radio-
therapy. Level of evidence, fair; benefit,
7. At the time of resection of a Pancoast tumor,
every effort should be made to achieve a
complete resection. Level of evidence, good;
tion, A
8. Resection of a Pancoast tumor should consist
of a lobectomy (instead of a wedge), as well as
removal of the involved chest wall structures.
9. For patients with a Pancoast tumor, postoper-
ative radiotherapy is not recommended, either
in completely or incompletely resected pa-
tients, because of lack of a demonstrated
survival benefit. Level of evidence, poor; ben-
efit, none; grade of recommendation, D
10. Patients with a good performance status and
an unresectable, but nonmetastatic Pancoast
tumor should be considered for combination
chemotherapy and radiotherapy with intent to
cure. Level of evidence, poor; benefit, mod-
11. Palliative radiotherapy should be considered
in patients who are not candidates for treat-
ment with curative intent (ie, surgery, chemo-
radiotherapy etc.). Level of evidence, fair;
tion, B
T4N0,1M0 Tumors
12. Patients with a clinical T4N0,1M0 NSCLC
should be carefully evaluated (with imaging
studies) for distant metastatic disease prior to
considering surgical resection. Level of evi-
ommendation, B
13. Resection of T4N0,1M0 tumors in selected
patients may result in better survival than
chemoradiotherapy without resection. Level
recommendation, C
14. Mediastinoscopy should be done prior to sur-
gical resection of patients with clinical
T4N0,1M0 tumors. Level of evidence, fair;
tion, B
Satellite Nodules of Cancer in the Same Lobe
15. No further diagnostic workup of a satellite
nodule is needed in patients with suspected or
proven lung cancer and a satellite nodule
within the same lobe. Level of evidence, fair;
tion, B
16. For patients with a satellite lesion within the
same lobe as a suspected or proven primary
lung cancer, distant organ scanning and con-
firmation of the mediastinal node status
should be carried out as dictated by the
primary lung cancer alone, and not modified
due to the presence of the satellite lesion.
17. For patients with NSCLC and a satellite focus
of cancer within the same lobe, resection via a
lobectomy is the preferred treatment. Level of
recommendation, B
Synchronous Second Primary Lung Cancer
18. Patients suspected of having two synchronous
primary lung cancers should have a thoughtful
search for an extrathoracic primary cancer to
rule out the possibility that both of the lung
lesions represent metastases from an extratho-
racic primary. Level of evidence, poor; bene-
nous primary lung cancers, a careful and
thorough search for distant metastases should
be performed. Level of evidence, poor; bene-
nous primary lung cancers, the absence of
mediastinal node involvement should be con-
firmed (usually via mediastinoscopy) prior to
21. For patients with two synchronous primary
lung cancers, resection of both lung cancers is
reasonable provided a careful search for dis-
tant metastases or mediastinal lymph node
involvement has been carried out and is neg-
ative. Level of evidence, poor; benefit, mod-
22. If a patient (not suspected of having a second
focus of cancer) is discovered intraoperatively
to have a second cancer in a different lobe, it
is recommended that a resection of each
lesion is undertaken, provided the patient has
adequate pulmonary reserve and there is no
N2 nodal involvement. Level of evidence,
dation, C
Metachronous Second Primary Lung Cancer
23. A careful and thorough search for distant
metastases should be performed in patients
suspected of having metachronous second pri-
mary lung cancers. Level of evidence, poor;
tion, C
24. Resection of a metachronous second primary
lung cancer is reasonable provided a careful
search for other distant metastases or medias-
tinal lymph node involvement has been car-
ried out and is negative. Level of evidence,
dation, C
Isolated Brain Metastasis
25. Patients with an isolated brain metastasis from
NSCLC should be considered for a curative
approach. Level of evidence, poor; benefit,
26. For patients with an isolated brain metastasis
from NSCLC who are being considered for a
curative approach, a careful search for other
distant metastases should be carried out with
imaging tests. Level of evidence, poor; bene-
of isolated brain metastases and a resectable
primary lung cancer, mediastinoscopy should
be done to rule out N2,3 involvement prior to
curative resection. Level of evidence, poor;
tion, C
and a synchronous resectable N0,1 primary
NSCLC, resection or radiosurgical ablation of
an isolated brain metastasis should be under-
taken (as well as resection of the primary
tumor). Level of evidence, fair; benefit, sub-
tion or radiosurgical ablation of an isolated
brain metastasis should be undertaken. Level
recommendation, B
resection of an isolated brain metastasis, ad-
juvant whole brain radiotherapy is reasonable,
although there is conflicting and insufficient
data regarding a benefit with respect to sur-
vival or the rate of recurrent brain metastases.
Level of evidence, poor; benefit, unclear;
resection of an isolated brain metastasis (and
resection of the primary tumor), adjuvant
chemotherapy can be neither recommended
nor recommended against because of insuffi-
cient data regarding this issue. Level of evi-
dence, poor; benefit, unclear; grade of recom-
mendation, I
Isolated Adrenal Metastasis
32. Patients with an isolated adrenal metastasis
from NSCLC should be considered for a
curative approach. Level of evidence, poor;
tion, C
33. For patients with an isolated adrenal metasta-
sis being considered for curative therapy, a
careful search for other distant metastases
should be carried out with imaging tests.
of an isolated adrenal metastasis and a resect-
able primary lung cancer, mediastinoscopy
should be done to rule out N2,3 involvement
prior to resection. Level of evidence, poor;
tion, C
35. In carefully selected patients with no other
sites of metastases and a synchronous resect-
able N0,1 primary NSCLC, resection of an
isolated adrenal metastasis from NSCLC
should be undertaken (as well as resection of
the primary tumor). Level of evidence, poor;
tion, C
tion of an isolated adrenal metastasis should
be undertaken. Level of evidence, poor; ben-
efit, moderate; grade of recommendation, C
References
1 British Thoracic Society. Guidelines on the selection of
patients with lung cancer for surgery. Thorax 2001; 56:89
108
2 Scottish Intercollegiate Guidelines Network. Management of
lung cancer: a national clinical guideline recommended for
use in Scotland. SIGN, Edinburgh, Scotland, 1998
3 National Cancer Institute. Non-small cell lung cancer. PDQ
(Physician Data Query) (database) summary, 1999. Available
at: http://www.cancer.gov/cancerinfo/pdq/treatment/non-
small-cell-lung/healthprofessional. Accessed December 21,
2002
4 Ettinger DS, Bepler G, Cox JD, et al. National Comprehen-
sive Cancer Network (NCCN) practice guidelines for non-
small-cell lung cancer. Rockledge, PA: National Comprehen-
sive Cancer Network; 2000.
5 National Cancer Institute. Guidelines on the non-surgical
management of lung cancer. Clin Oncol (R Coll Radiol) 2001;
11:S1S53
6 Detterbeck FC, Rivera MP, Socinski MA, et al, eds. The
diagnosis and treatment of lung cancer: an evidence-based
guide for the practicing clinician. 1st ed. Philadelphia, PA:
WB Saunders, 2001
7 Dartevelle P, Macchiarini P. Resection of superior sulcus
tumors. In: Kaiser LR, Kron IL, Spray, TL, eds. Mastery of
cardiothoracic surgery. Philadelphia, PA: Lippencott-Raven,
1998; 257265
8 Heelan RT, Demas BE, Caravelli JF, et al. Superior sulcus
tumors: CT and MR imaging. Radiology 1989; 170:637641
9 Shaw RR, Paulson DL, Kee JL Jr. Treatment of the superior
sulcus tumor by irradiation followed by resection. Ann Surg
1961; 154:2940
10 Van Houtte P, MacLennan I, Poulter C, et al. External
radiation in the management of superior sulcus tumor. Can-
cer 1984; 54:223227
11 Detterbeck FC, Jones DR, Rosenman JG. Pancoast tumors.
In: Detterbeck FC, Rivera MP, Socinski MA, et al, eds.
Diagnosis and treatment of lung cancer: an evidence-based
guide for the practicing clinician. Philadelphia, PA: WB
Saunders, 2001; 233243
12 Ahmad K, Fayos JV, Kirsh MM. Apical lung carcinoma.
Cancer 1984; 54: 913917
13 Komaki R, Roh J, Cox JD, et al. Superior sulcus tumors:
results of irradiation of 36 patients. Cancer 1981; 48:1563
1568
14 Millar J, Ball D, Worotniuk V, et al. Radiation treatment of
superior sulcus lung carcinoma. Australas Radiol 1996; 40:
5560
15 Ginsberg RJ, Martini N, Zaman N, et al. Influence of surgical
resection and brachytherapy in the management of superior
sulcus tumor. Ann Thorac Surg 1994; 57:14401445
16 Gandhi S, Walsh GL, Komaki R, et al. A multidisciplinary
surgical approach to superior sulcus tumors with vertebral
invasion. Ann Thorac Surg 1999; 68:17781785
17 Dartevelle PG, Chapelier AR, Macchiarini P, et al. Anterior
transcervical-thoracic approach for radical resection of lung
tumors invading the thoracic inlet. Thorac Cardiovasc Surg
1993; 105:10251034
18 Rusch VW, Giroux DJ, Kraut MJ, et al. Induction chemora-
diation and surgical resection for non-small cell lung carcino-
mas of the superior sulcus: Initial results of Southwest
Oncology Group Trial 9416 (Intergroup Trial 0160). Thorac
19 Detterbeck FC, Jones DR. Surgery for stage IIIb non-small
cell lung cancer. In: Detterbeck FC, Rivera MP, Socinski
MA, et al, eds. Diagnosis and treatment of lung cancer: An
evidence-based guide for the practicing clinician. Philadel-
phia, PA: WB Saunders, 2001; 283289
20 Tsuchiya R, et al. Extended resection of the left atrium, great
vessels, or both for lung cancer. Ann Thorac Surg 1994;
57:960965
21 Dartevelle PG. Herbert Sloan Lecture: Extended operations
for the treatment of lung cancer. Ann Thorac Surg 1997;
63:219
22 Rendina EA, Venuta F, De Giacomo T, et al. Induction
chemotherapy for T4 centrally located non-small cell lung
23 Hensing TA, Halle JS, and Socinski MA. Chemoradiotherapy
for stage IIIa, b non-small cell lung cancer. In: Detterbeck
FC, Rivera MP, Socinski MA, et al, eds. Diagnosis and
treatment of lung cancer: An evidence-based guide for the
practicing clinician. Philadelphia, PA: WB Saunders, 2001;
291303
24 Albain KS, et al. Concurrent cisplatin/etoposide plus radio-
therapy (PERT) for pathologic stage (pathTN) IIIB non-
small cell lung cancer (NSCLC): A Southwest Oncology
Group (SWOG) phase II study (S9019). Proc ASCO 1997;
16:446a (abstract)
25 Keogan MT, Tung KT, Kaplan DK, et al. The significance of
pulmonary nodules detected on CT staging for lung cancer.
Clin Radiol 1993; 48:9496
26 Kunitoh H, Eguchi K, Yamada K, et al. Intrapulmonary
sublesions detected before surgery in patients with lung
cancer. Cancer 1992; 70:18761879
27 Detterbeck FC, Jones DR, and Funkhouser WK Jr. Satellite
nodules and multiple primary cancers. In: Detterbeck FC,
Rivera MP, Socinski MA, et al, eds. Diagnosis and treatment
of lung cancer: An evidence-based guide for the practicing
clinician. Philadelphia, PA: WB Saunders, 2001; 437449
28 Deslauriers J, Brisson J, Cartier R, et al. Carcinoma of the
lung: evaluation of satellite nodules as a factor influencing
prognosis after resection. Thorac Cardiovasc Surg 1989;
97:504512
29 Ferguson MK, DeMeester TR, DesLauriers J, et al. Diagno-
sis and management of synchronous lung cancers. Thorac
30 Rosengart TK, Martini N, Ghosn P, et al. Multiple primary
lung carcinomas: prognosis and treatment. Ann Thorac Surg
1991; 52:273279
31 Okada M, Tsubota N, Yoshimura M, et al. Operative ap-
proach for multiple primary lung carcinomas. Thorac Cardio-
vasc Surg 1998; 115:836840
32 Antakli T, Schaefer RF, Rutherford JE, et al. Second primary
Surg 1991; 102:266271
34 Payne CR, Hadfield JW, Stovin PG, et al. Diagnostic accuracy
of cytology and biopsy in primary bronchial carcinoma. J Clin
Pathol 1981; 34:773778
35 Truong LD, Underwood RD, Greenberg SD, et al. Diagnosis
and typing of lung carcinomas by cytopathologic methods: a
review of 108 cases. Acta Cytol 1985; 29:379384
36 Cataluna JJS, Perpina M, Greses JV, et al. Cell type accuracy
of bronchial biopsy specimens in primary lung cancer. Chest
1996; 109:11991203
37 Deschamps C, Pairolero PC, Trastek VF, et al. Multiple
primary lung cancers: results of surgical treatment. Thorac
38 Martini N, Melamed MR. Multiple primary lung cancers.
39 Mathisen DJ, Jensik RJ, Faber LP, et al. Survival following
resection for second and third primary lung cancers. Thorac
40 Verhagen AFTM, Tavilla F, van de Wal HJ, et al. Multiple
primary lung cancers. Thorac Cardiovasc Surg 1994; 42:
4044
41 van Bodegom PC, Wagenaar SS, Corrin B, et al. Second
primary lung cancer: importance of long term follow up.
Thorax 1989; 44:788793
42 Ribet M, Dambron P. Multiple primary lung cancers. Eur
J Cardiothorac Surg 1995; 9:231236
43 Adebonojo SA, Moritz DM, Danby CA. The results of
modern surgical therapy for multiple primary lung cancers.
Chest 1997; 112:693701
44 Wu S, Lin ZO, Xu CW, et al. Multiple primary lung cancers.
Chest 1987; 92:892896
45 Detterbeck FC, Jones DR, Molina PL. Extrathoracic staging.
In: Detterbeck FC, Rivera MP, Socinski MA, et al, eds.
Diagnosis and treatment of lung cancer: An evidence-based
guide for the practicing clinician. Philadelphia, PA: WB
Saunders, 2001; 94110
46 Bindal AK, Bindal RK, Hess KR, et al. Surgery vs radiosur-
gery in the treatment of brain metastasis. J Neurosurg 1996;
84:748754
47 Detterbeck FC, Bleiweis MS, Ewend MG. Surgical treatment
of stage IV non-small cell lung cancer. In: Detterbeck FC,
Rivera MP, Socinski MA, et al, eds. Diagnosis and treatment
of lung cancer: An evidence-based guide for the practicing
clinician. Philadelphia, PA: WB Saunders, 2001; 326338
48 Alexander E III, Moriarity TM, Davis RB, et al. Stereotactic
radiosurgery for the definitive, noninvasive treatment of brain
metastases. J Natl Cancer Inst 1995; 87:3440
49 Flickinger JC, Kondziolka D, Lundsford LD, et al. A multi-
institutional experience with stereotactic radiosurgery for
solitary brain metastasis. Int J Radiat Oncol Biol Phys 1994;
28:797802
50 Kim YS, Kondziolka D, Flickinger JC, et al. Stereotactic
radiosurgery for patients with nonsmall cell lung carcinoma
metastatic to the brain. Cancer 1997; 80:20752083
51 Bindal RK, Sawaya R, Leavens ME, et al. Surgical treatment
of multiple brain metastases. J Neurosurg 1993; 79:210216
52 Burt M, Wronski M, Arbit E, et al. Resection of brain
metastases from non-small-cell lung carcinoma: results of
therapy. Thorac Cardiovasc Surg 1992; 103:399411
53 Smalley SR, Laws ER Jr, OFallon JR, et al. Resection for
solitary brain metastasis. J Neurosurg 1992; 77:531540
54 Patchell RA, Tibbs PA, Regine WF, et al. Postoperative
radiotherapy in the treatment of single metastases to the
brain: a randomized trial. JAMA 1998; 280:14851489
55 Porte H, Siat J, Guibert B, et al. Resection of adrenal
metastases from non-small cell lung cancer: A multiinstitu-
tional study. Ann Thorac Surg 2001; 71:981985
56 Pham DT, Dean DA, Detterbeck FC. Adrenalectomy as the
new treatment paradigm for solitary adrenal metastasis from
lung cancer. Paper presented at: The Society of Thoracic
Surgeons Thirty-Seventh Annual Meeting, Jan. 30, 2001, New
Orleans, LA
2003;123;244-258 Chest
Frank C. Detterbeck, David R. Jones, Kemp H. Kernstine and Keith S. Naunheim
Special Treatment Issues
& Services
Updated Information
S
References
S#BIBL
free at:
Reprints
2003;123;259-271 Chest
George R. Simon and Henry Wagner
Small Cell Lung Cancer
ISSN: 0012-3692.
Small Cell Lung Cancer*
George R. Simon, MD, FCCP; and Henry Wagner, MD
Among patients with lung cancers, the proportion of those with small cell lung cancer (SCLC) has
decreased over the last decade. SCLC is staged as limited-stage disease and extensive-stage
disease. Standard staging procedures for SCLC include CT scans of the chest and abdomen, bone
scan, and CT scan or MRI of the brain. The role for positron emission tomography scanning in the
staging of SCLC has yet to be defined. Limited-stage disease is treated with curative intent with
chemotherapy and radiation therapy, with approximately 20% of patients achieving a cure. The
median survival time for patients with limited-stage disease is approximately 18 months.
Extensive-stage disease is treated primarily with chemotherapy, with a high initial response rate
of 60 to 70% and a complete response rate of 20 to 30%, but with a median survival time of
approximately 9 months. Patients achieving a complete remission should be offered prophylactic
cranial irradiation. Currently, there is no role for maintenance treatment or bone marrow
transplantation in the treatment of patients with SCLC. Relapsed or refractory SCLC has a
uniformly poor prognosis. In this section, evidence-based guidelines for the staging and
treatment of SCLC are outlined. (CHEST 2003; 123:259S271S)
Key words: carboplatin; chemotherapy; cisplatin; etoposide; irinotecan; paclitaxel; prophylactic cranial irradiation;
radiation therapy; small cell lung cancer
Abbreviations: CAV cyclophosphamide, adriamycin, and vincristine; CODE cyclophosphamide, vincristine,
doxorubicin, and etoposide; CR complete response; ECOG Eastern Cooperative Oncology Group;
G-CSF granulocyte colony-stimulating factor; NSCLC non-small cell lung cancer; PCI prophylactic cranial
irradiation; PET positron emission tomography; SCLC small cell lung cancer; TEP paclitaxel, etoposide, and
cisplatin; TRT thoracic radiotherapy
T
his document is the result of a comprehensive
review of the existing guidelines, meta-analyses,
and relevant randomized clinical trials on the subject
of small cell lung cancer (SCLC).
Among lung cancers, the proportion of patients
with SCLC has decreased from 17.4% in 1986 to
13.8% in 1998.
1
Like non-SCLC (NSCLC), it has a
strong association with tobacco use, but its clinical
characteristics tend to be more aggressive than
NSCLC, and median survival time without treat-
ment is between 2 and 4 months.
Staging of SCLC
Patients are staged according to a two-stage sys-
tem, which was developed by the Veterans Admin-
istration Lung Cancer Study Group, as having limited-
stage disease or extensive-stage disease. Patients
with limited-stage disease have involvement re-
stricted to the ipsilateral hemithorax within a single
radiation port. Extensive-stage disease is defined as
the presence of obvious metastatic disease. Patients
with limited-stage disease with the presence of con-
tralateral hilar and/or supraclavicular nodes and/or
with malignant pericardial and/or pleural effusions
are excluded from clinical trials for limited-stage
SCLC.
A complete evaluation of a patient newly diag-
nosed with SCLC should consist of a medical history
and physical examination, a review of the histopa-
thology specimens, a CT scan of the chest and upper
abdomen to include the whole liver and the adrenal
glands, a bone scan, and a CT scan or MRI exami-
nation of the brain. Additionally, complete blood
counts, measurement of electrolyte, BUN, and cre-
atinine levels, liver function tests, and measurement
of lactate dehydrogenase levels should be performed
in all patients at baseline. The utility of positron
emission tomography (PET) scanning in patients
with SCLC has been recently reported in two small
prospective studies.
2,3
In a study reported by Hauber
et al,
2
PET scans detected all primary lesions, lymph
node metastases, and distant metastases that had
been detected by other standard staging procedures.
In a second study,
3
30 patients with SCLC were
evaluated with 36 PET scan examinations, and the
Tampa, FL.
Correspondence to: George R. Simon, MD, FCCP, 12902 Mag-
nolia Dr, Suite 3170, Thoracic Oncology Program, H. Lee Moffitt
Cancer Center and Research Institute, Tampa, FL 33612; e-mail:
Simongr@moffitt.usf.edu
results were compared with the sum of the other
staging procedures. The results of 23 of the 36 PET
scan examinations were concordant with those of the
other staging procedures. In seven cases, the PET
scan examination resulted in upward staging of the
patient, and in one instance the PET scan revealed
the presence of a viable tumor when conventional
staging procedures had revealed no residual disease.
PET scan identified all areas of tumor involvement
detected by other staging procedures. A third
study
102
looked at the accuracy of PET scanning in
detecting bony metastases in patients with SCLC
and NSCLC, comparing the PET scans to bone
scans and single-photon emission CT scans. In this
study, PET scans were found to be the most accurate
whole-body imaging modality for the screening of
bone metastases. These studies suggested that PET
scanning is likely to be a useful staging tool in
patients with SCLC. However, all the above studies
were small, and the experience with PET scan as a
staging tool remains largely limited. Until larger
prospective studies become available, PET scanning
cannot be recommended for routine use in the
staging and restaging of patients with SCLC.
Recommendation
1. In all patients, the routine staging of SCLC
should include medical history and physical
examination, complete blood counts, compre-
hensive chemistry panels, CT scans of the chest
and abdomen, a CT scan or MRI of the brain,
and a bone scan. Level of evidence, good; benefit,
2. For the routine staging of patients with SCLC,
PET scanning is not recommended outside of a
clinical trial. Level of evidence, fair; benefit,
Treatment of Extensive Stage SCLC
First-Line Treatment
Platinum-based chemotherapy remains the main-
stay of treatment for extensive SCLC. In a meta-
analysis
4
of randomized trials (19 trials and 4,054
evaluable patients) comparing a cisplatin-based reg-
imen with a non-cisplatin-based regimen, patients
randomized to a regimen containing cisplatin had a
significant increase in the probability of response and
survival with no significant increase in toxicity. De-
tailed analyses of the roles of etoposide and cisplatin
in the treatment of SCLC have been performed by
Berghmans et al
5
and were reported in abstract form
in September 1999. Thirty-six eligible trials that
were performed between 1980 and 1998 were clas-
sified into the following four groups: (1) cisplatin vs
no cisplatin (1 trial); (2) etoposide (without cisplatin)
vs no etoposide (17 trials); (3) cisplatin/etoposide vs
no cisplatin/etoposide (9 trials); and (4) cisplatin/
etoposide vs etoposide (9 trials). The authors con-
cluded that the use of cisplatin and/or etoposide
offered a significant survival advantage to patients
with SCLC.
5
In another meta-analysis, Chute et al
6
evaluated all
21 cooperative group trials performed in North
America from 1972 to 1993. Patients with extensive-
stage SCLC who were treated during a similar time
interval and were listed in the Surveillance, Epide-
miology, and End Results database also were exam-
ined. Trends were tested in the number of trials and
the survival of patients over time. In this analysis, a
modest 2-month prolongation in median survival was
demonstrated in patients with extensive-stage SCLC.
This improvement in survival was independently
associated with both cisplatin-based therapy and in
the improvement of best supportive care. This meta-
analysis again establishes that cisplatin-based chemo-
therapy should be the cornerstone of first-line che-
motherapy for patients with extensive-stage SCLC.
The issue of carboplatin vs cisplatin was recently
reviewed by Brahmer and Ettinger,
7
who concluded
that carboplatin plus etoposide is as effective as
cisplatin plus etoposide but is less toxic (except for
increased myelosuppression). The Hellenic Oncol-
ogy Group conducted a phase III trial
103
comparing
cisplatin and etoposide with carboplatin and etopo-
side. In this study, containing patients with both
limited-stage and extensive-stage disease, the me-
dian survival time was 11.8 months for cisplatin plus
etoposide and 12.5 months for carboplatin plus
etoposide. The difference was not statistically signif-
icant, although the study was not powered to show
equivalence.
103
A recent Japanese trial
8
compared cisplatin and
irinotecan with cisplatin and etoposide. Patients
randomized to the cisplatin/irinotecan arm did (sta-
tistically) significantly better than the group that was
randomized to the cisplatin/etoposide arm (median
survival time, 420 vs 300 days, respectively). Confir-
matory trials are underway in the United States.
Several phase II trials with irinotecan, topotecan,
and paclitaxel in combination with either cisplatin or
etoposide have been reported, and these have been
summarized in Table 1.
The issue of adding a third drug to cisplatin and
etoposide has been investigated. The Hoosier On-
cology Group evaluated the addition of ifosfamide to
cisplatin and etoposide in a phase III trial of 171
patients with extensive-stage disease. At the expense
of increased toxicity, the 2-year survival rate in-
creased from 5 to 13% with the addition of ifos-
famide.
9
Mavroudis et al
10
compared the use of
paclitaxel, etoposide, and platinum (TEP) with the
use of etoposide and platinum. The study was ter-
minated early, secondary to a higher number of toxic
deaths in the TEP arm. Despite a statistically signif-
icant improvement in the time to progression for
TEP, there was no difference in overall survival.
Recently, another phase III intergroup trial (Cancer
and Leukemia Group B 9732) was reported
11
com-
paring cisplatin and etoposide with or without pac-
litaxel in patients with extensive-stage SCLC. No
significant survival advantage was seen with the
addition of paclitaxel to cisplatin and etoposide in
this study. On the other hand, there was an increased
incidence of deaths from toxicities in the paclitaxel
arm.
Recommendations
3. Patients with extensive-stage disease should
receive platinum-based chemotherapy. Level
recommendation, A
4. Patients achieving a complete remission (CR)
should be offered prophylactic cranial irradia-
tion (PCI). Level of evidence, fair; benefit,
small; grade of recommendation, C
Maintenance Treatment
The topic of maintenance therapy in patients
with SCLC was extensively reviewed in the journal
Lung Cancer in 1998.
12
Several randomized trials
have demonstrated that 4 to 6 months of treatment is
equal to prolonged treatment when survival is con-
sidered as the end point. In the meta-analysis re-
ported by Sculier et al,
12
13 published randomized
trials were included. One showed a statistically sig-
nificant difference in survival in favor of mainte-
nance therapy, 5 studies showed survival advantage
in subgroups of patients, 1 study showed significantly
shorter survival times with maintenance therapy, and
6 studies showed no difference. The Eastern Coop-
erative Oncology Group (ECOG) conducted a phase
III trial in which patients showing a response to
therapy or patients whose disease stabilized after
receiving four cycles of cisplatin and etoposide were
randomized to observation alone or to four cycles of
topotecan therapy.
13
Despite an improvement in
progression-free survival, there was no difference in
overall survival between the two groups.
Treatments other than chemotherapy for mainte-
nance are currently being investigated in ongoing
clinical trials. A phase III randomized trial is cur-
rently underway testing the efficacy of anti-GD3
immunization as maintenance treatment. Metallo-
proteinase inhibitors and inhibitors of angiogenesis
also are being investigated in this fashion.
Recommendation
5. For patients with extensive-stage or limited-
stage SCLC achieving a partial or CR, there is
no evidence, outside of a clinical trial, for
Table 1Phase II Trials of SCLC Patients Receiving Combination Chemotherapy vs Untreated Patients*
Treatment
Patients,
No.
Responders, No. Response Rate Survival
Reference CR PR Total % 95% CL Mo
1-Yr
Rate, %
CEC 46 10 32 42 91 7998 18 22 69
TEP 38 6 28 34 90 7597 12 70
TEP 23 5 14 19 83 6195 11 46 71
Cisplatin-irinotecan 35 10 20 30 86 7095 13 21.7 72
Cisplatin-vinblastine-MMC 30 1 21 22 73 6696 6 73
Cisplatin-etoposide-all-trans-RA 22 1 9 10 45 2468 11 41 74
Cisplatin-paclitaxel-G-CSF 34 3 20 23 61 538 8 75
Topotecan-paclitaxel 28 6 11 17 60 4179 14 76
Etoposide-irinotecan 50 33 66 5179 12 77
Paclitaxel-carboplatin 69 5 37 42 61 4872 12 78
Paclitaxel-irinotecan 11 4 1 5 45 1777 79
Paclitaxel-doxorubicin 16 1 3 4 25 752 80
Cisplatin-docetaxel 20 0 11 11 55 3277 81
Topotecan-paclitaxel 13 8 69 3991 14 82
Topotecan-paclitaxel 15 10 5 15 100 78100 83
CPT 18 3 10 13 72 4790 84
EPE 12 6 6 12 100 74100 85
*CEC cisplatin-etoposide-carboplatin; MMC mitomycin C; RA retinoic acid; CL confidence limits; CPT cisplatin-paclitaxel-topote-
can; EPE etoposide-paclitaxel-epirubicin; PR partial response.
Includes 14 patients with stage IIIB and stage IV disease.
maintenance treatment. Level of evidence,
good; benefit, none/negative; grade of recom-
mendation, D
Treatment of Relapsed or Refractory
SCLC
Despite high initial response rates to chemother-
apy (ie, 45 to 75% CRs) reported in patients with
limited-stage disease and 20 to 30% CRs in patients
with extensive-stage disease, the response duration is
usually short with a progression-free survival time of
approximately 4 months for patients with extensive-
stage disease and 12 months for patients with limited-
stage disease. Most patients are destined to relapse,
and the prognosis of this group of relapsed patients is
poor. Patients who relapse 3 months after first-
line therapy are commonly called refractory, and
patients who relapse 3 months after therapy are
called sensitive. Patients with late relapses after
receiving initial therapy may be retreated with the
same induction regimen used initially.
Patients whose disease progresses early after
induction therapy and who are in satisfactory
clinical condition, should be offered a second-line
regimen. In a randomized multicenter study, von
Pawel et al
101
compared cyclophosphamide, adria-
mycin, and vincristine (CAV) with topotecan as a
single agent in patients who relapsed at least 60
days after the completion of initial therapy. A total
of 211 patients were enrolled in the study. The
response rate was 24.3% in patients who were
treated with topotecan and was 18.3% in patients
treated with CAV (p 0.285). The median times
to disease progression were 13.3 weeks for pa-
tients in the topotecan arm and 12.3 weeks for
patients in the CAV arm. The median survival time
was 25 weeks for patients receiving topotecan and
24.7 weeks for those receiving CAV. The propor-
tion of patients who experienced symptom im-
provement was greater in the topotecan arm than
in the CAV group for four of the eight symptoms
evaluated. The authors concluded that topotecan
was at least as effective as CAV in the treatment of
patients with recurrent SCLC and resulted in
improved control of several symptoms. However,
toxicity rates were high in both arms of the study,
and alternative dose schedules of topotecan are
currently being evaluated. Several recently re-
ported phase II trials in patients with relapsed/
refractory SCLC are summarized in Table 2.
Recommendation
6. Patients with SCLC who have relapsed follow-
ing an initial response to treatment or who are
refractory to the initial treatment should be
offered further chemotherapy. The chemother-
apy offered will depend on the duration of
response after receiving first-line chemother-
apy or the lack of response to first-line chemo-
therapy (ie, sensitive relapses vs refractory pa-
tients). Level of evidence, fair; benefit, small/
weak; grade of recommendation, C
Treatment of Elderly Patients With
Extensive-Stage SCLC
Approximately 25% of patients with SCLC are
70 years of age (ie, elderly). The performance
status and the physiologic status of the patient should
guide treatment decisions rather than the patients
chronologic age. It is clear that patients with good
performance status (ECOG level 0 or 1) and normal
organ function should be treated with optimal che-
motherapy (and with radiotherapy, if indicated) as in
their younger counterparts. Similar outcomes of
elderly patients with limited-stage SCLC have been
shown in the intergroup trial 0096 in which cisplatin,
etoposide, and thoracic radiotherapy were adminis-
tered once a day or twice daily.
14
The National
Cancer Institute of Canada performed a retrospec-
Table 2Phase II Trials of SCLC Patients Receiving Combination Chemotherapy: Refractory or Relapsed Patients*
Treatment
Patients,
No.
Responders, No.
Response
Rate, %
Survival
Reference CR PR Total Mo
1 Yr
Rate, %
Etoposide-irinotecan 24 3 14 17 71 9 86
Cisplatin-topotecan 28 1 7 8 29 87
Etoposide-hexmethylmelamine 30 1 5 6 22 5 21 88
Irinotecan-paclitaxel 11 1 4 5 45 89
Carboplatin-paclitaxel 18 0 3 3 17 90
tive review
15
of their BR3 and BR6 trials and also
concluded that age did not appear to impact the
delivery, tolerance, or efficacy of thoracic irradiation
in the combined-modality management of patients
with limited-stage SCLC. Greater myelosuppression
is to be expected since equivalent exposure to a drug
will lead to more myelosuppression in the elderly
compared to their younger counterparts. This has
been shown to be the case at least for etoposide.
16
Greater ancillary support will be required in the
elderly. However, despite treatment delays elderly
patients with good performance status have similar
prognoses to those of younger patients.
Elderly patients with poor performance status or
with compromised organ function may be offered
single-agent chemotherapy or polychemotherapy in
attenuated doses. However, several randomized
studies
17,18
have indicated that such gentler che-
motherapy is inferior to optimal combination chemo-
therapy. Options available to these patients include
the following: oral etoposide for 14 days combined
with carboplatin on day 1 every 28 days
19
; abbrevi-
ated chemotherapy with CAV in full doses followed
3 weeks later by chemotherapy with cisplatin and
etoposide in optimal doses
20
; or chemotherapy with
platinum, adriamycin, vincristine, and etoposide in
reduced doses.
21
A recently reported phase III trial
22
compared carboplatin-gemcitabine therapy with
cisplatin-etoposide therapy in patients with SCLC
who had poor prognoses, with carboplatin-gemcitab-
ine therapy exhibiting a more favorable overall tox-
icity profile at the expense of increased myelotoxic-
ity. Another phase III trial compared the use of
single-agent carboplatin with CAV,
23
with carbopla-
tin producing response rates, relief of tumor-related
symptoms, and survival similar to that seen with
CAV. There was a lower risk of life-threatening
sepsis and less need for hospitalization in the group
that received carboplatin.
Recommendations
7. Elderly patients with good performance status
and with intact organ function should be
treated with platinum-based chemotherapy.
Level of evidence, good; benefit, moderate;
8. Elderly patients with poor prognostic factors
such as poor performance status or severe
concomitant comorbid disease still may be con-
sidered for chemotherapy. Level of evidence,
poor; benefit, small; grade of recommendation, C
9. Elderly patients achieving a CR should be
offered PCI. Level of evidence, fair; benefit,
Dose Intensity in SCLC
The issue of dose intensity has been subjected to
extensive clinical investigation in recent years pri-
marily due to the initially chemosensitive nature of
the disease and also owing to the almost universal
lack of durable responses and cure, despite the initial
responses. Several randomized trials,
2426
however,
have failed to show a survival advantage with dose-
intense chemotherapy. On the basis of promising
results from a pilot trial reported by Murray et al,
27
the Southwest Oncology Group reported the results
of their phase III trial comparing therapy with
cisplatin, vincristine, doxorubicin, and etoposide
(CODE) with alternating therapy using CAV then
cisplatin and etoposide. The study was stopped early
owing to an excessive mortality rate in the CODE
arm (8%) compared to the standard arm (1%). The
response rates with CODE were higher than that of
patients in the standard arm of the study, but
progression-free survival and overall survival were
not statistically significantly improved. Furuse et al
28
and Skarlos et al
29
compared more dose-intense
weekly regimens with standard or alternating non-
cross-resistant regimens and again failed to show a
survival advantage for the weekly, more dose-intense
arms. Pujol et al
30
compared therapy with cyclophos-
phamide, epirubicin, etoposide, and cisplatin in
higher doses with recombinant human granulocyte-
macrophage colony-stimulating factor vs the same
drugs in lower doses without granulocyte-macro-
phage colony-stimulating factor and failed to show a
survival advantage for patients in the more dose-
intense arm. However, Steward et al,
31
comparing
therapy with the combination of vincristine, ifos-
famide, carboplatin, and etoposide every 3 weeks vs
the same regimen every 4 weeks, showed a signifi-
cant survival advantage for patients receiving the
intensified regimen. Another trial, which was con-
ducted by Thatcher et al
32
and that also showed
positive results for the dose-intense approach, com-
pared standard therapy with adriamycin, cyclophos-
phamide, and etoposide administered every 3 weeks
with dose-dense therapy with adriamycin, cyclophos-
phamide, and etoposide administered every 2 weeks
along with granulocyte-colony stimulating factor (G-
CSF) support. The overall survival rate was statisti-
cally significantly better in the dose-dense group
(1-year survival rate, 47% vs 39%, respectively
[p 0.04]; 2-year survival rate, 13% vs 8%, respec-
tively). Progression-free survival, nonhematologic
toxicity, and quality of life were similar in the two
groups.
Interestingly, Arriagada et al
33
conducted a ran-
domized trial of 105 patients with limited-stage
SCLC who were randomized to receive a higher
dose of cisplatin (100 vs 80 mg/m
2
) or cyclophos-
phamide (300 vs 225 mg/m
2
) initially. All patients
then were treated with the same dose of doxorubicin
and etoposide, and radiotherapy. There was an im-
proved disease-free survival rate (2-year survival
rate, 28% vs 8%, respectively; p 0.02) and an
improved overall survival rate (2-year survival rate,
46% vs 26%, respectively; p 0.02) among patients
in the higher dose arm. The issue of dose intensity
may need further evaluation in the limited-stage
disease setting.
Recommendation
10. For patients with either extensive-stage SCLC
or limited-stage SCLC, there is no role for
dose-dense/intense initial/induction, or main-
tenance treatment outside of a clinical trial.
The Role of Growth Factor and the Use
of Stem Cell Support in SCLC
The use of G-CSF in the treatment of SCLC has
been analyzed carefully by Chouaid et al.
34
They did
a retrospective review of their experience and also
analyzed published data from three randomized
trials that evaluated the effectiveness of primary
therapy with G-CSF in patients with SCLC. The use
of G-CSF for primary prophylaxis was not found to
be cost-effective, did not improve palliation, and
hence is not recommended for routine use.
Recommendation
11. In patients with SCLC who are receiving
chemotherapy, the routine use of G-CSF is
not recommended. level of evidence, good;
tion, D.
Treatment of Limited-Stage SCLC
Radiation Therapy in SCLC
SCLC has long been recognized to be clinically
responsive to radiation, and in vitro radiation of
SCLC cell lines has shown that they often have a
greater intrinsic radiosensitivity than adenocarcino-
mas or squamous cell lung cancer cell lines. Because
of these observations, many early trials of combining
radiation with chemotherapy in patients with SCLC
used low total radiation doses. More recently, it has
become increasingly clear that higher doses than the
old regimens of 30 Gy in 10 fractions or 45 Gy in 25
fractions are needed to provide durable local control.
A number of trials conducted in the 1970s and
1980s compared chemotherapy alone to chemother-
apy plus thoracic radiation therapy (TRT) in patients
with limited SCLC. These varied in radiation dose,
timing, and choice of chemotherapeutic agents (but
most were performed with an alkylating agent and
doxorubicin-based therapy rather than with cisplatin
and etoposide). The analysis by Warde and Payne
35
showed improved local control and survival with the
addition of TRT, particularly in patients 60 years.
Pignon et al
36
obtained individual patient data from
these trials and were able to update analyses from
the time of the original publication. They found that
the addition of TRT resulted in an increase in the
3-year survival rate from 8.9 to 14.3%, an absolute
improvement of 5%, and a relative improvement of
nearly 50%.
36
With the publication of these two
meta-analyses, the debate shifted from whether or
not to employ TRT to how best to integrate it with
chemotherapy.
Sequencing and Timing of Thoracic Radiation and
Chemotherapy
The issues here are whether to administer radia-
tion therapy and chemotherapy concurrently, se-
quentially, or in an alternating fashion, and whether
radiation should be administered early or late in the
overall course of treatment. Murray and Coldman
37
performed a meta-analysis of trials that combined
chemotherapy and TRT, using progression-free sur-
vival at 3 years as a surrogate end point for long-term
survival. The best results were seen with TRT be-
ginning 3 to 5 weeks from the start of chemotherapy.
As radiation was further delayed, the benefit de-
creased and survival approached that seen with
chemotherapy alone.
There have been at least nine randomized trials
that have addressed the issue of the timing of
radiation in patients with limited-stage SCLC (Table
3). There are major differences in trial design, choice
of chemotherapeutic agents, radiation dose, and
fractionation schedules. With those caveats, several
reasonable conclusions emerge from these data, as
follows:
1. Trials that used alkylating agents and doxoru-
bicin-based chemotherapy showed little effect
of radiation timing and sequencing. They also
reported significant difficulty in delivering
planned treatment (both chemotherapy and
radiation) when radiation was given concur-
rently with or alternated between cycles of
chemotherapy. Long-term survival in most of
these trials was in the range of 10% that is
minimally different from that seen with chemo-
therapy alone.
2. When platinum-etoposide regimens are used
for chemotherapy, concurrent chemotherapy-
TRT is superior to sequential TRT, where TRT
is administered after chemotherapy.
3. When platinum-etoposide chemotherapy and
concurrent TRT are combined, the data are
divergent as to whether early TRT (ie, in week
1) is better than delayed TRT (ie, week 6 or
week 13).
Radiation Dose
Relatively few trials have addressed the issue of
optimal TRT dose in patients with SCLC. A retro-
spective analysis
38
of patients treated at the Massa-
chusetts General Hospital showed an improvement
in local control as radiation doses were increased
from 30 to 50 Gy. The Cancer and Leukemia Group
B
39
has tried to define the maximal tolerated dose for
concurrent TRT and cisplatin-etoposide chemother-
apy when these are administered after three courses
of induction chemotherapy with cyclophosphamide-
cisplatin-etoposide. They examined both a once-a-
day radiation therapy schedule with 2-Gy fractions
and a twice-daily schedule with 1.5-Gy fractions. The
limiting toxicity was acute esophagitis, and the max-
imum tolerated dose was reported to be 45 Gy in 3
weeks for twice-daily fractionation and 70 Gy in 7
weeks for daily fractionation. As the best local con-
trol rates in current trials do not exceed 70%, the
exploration of dose escalation seems warranted, and
a recent trial proposed by the ECOG to the National
Cancer Institute would compare the regimen of
twice-daily 45-Gy doses for 3 weeks to that of a daily
70-Gy dose for 7 weeks. Studies in patients with
NSCLC clearly have shown the feasibility of giving
even higher radiation doses to conformal planned
fields with concurrent chemotherapy, and this ap-
proach may also be feasible in patients with SCLC.
Radiation Fractionation
The rapid growth of many SCLC cell lines encour-
aged the exploration of treatment acceleration by
giving two fractions per day with a modest reduction
in fraction size from the usual 1.8 to 2.0 Gy to 1.5 Gy.
Two prospective trials have compared this approach
to conventional daily fractionation. The North Amer-
ican Intergroup trial 0096 compared doses of 45 Gy
administered in 25 fractions for 5 weeks to the
investigational arm of doses of 45 Gy administered in
Table 3Randomized Trials of Radiation-Chemotherapy Sequencing and Timing in Limited-Stage SCLC*
Study Comparison Chemotherapy Result Comment
Perry et al
91
TRT day 1 vs day 64
vs none
CAV/CEV Delayed TRT gave best survival Planned drug dose reduction in
arm with day 1 TRT
Murray et al
92
TRT cycle 2 vs cycle 6 CAV/PE
alternating
Median and 4-yr survival favored
early TRT
Greatest difference in patterns of
failure was in CNS relapse
Schultz et al
93
TRT day 1 vs day 120 CAV/PE
alternating
Delayed TRT nonsignificantly
better (MST, 12.9 vs 10.7 mo)
No long-term follow-up reported
Tsukada et al
94
TRT concurrent with
cycle 1 or following
cycle 4
PE 3-yr survival significantly better
for early concurrent vs late
sequential (30.9% vs 20.7%,
respectively)
Double comparison of timing and
concurrency
Gregor et al
95
TRT alternating
between cycles 2, 3,
4, and 5 or following
cycle 5
CAE Nonsignificant trend favoring
delayed TRT
More toxicity with resultant drug
dose reduction and delay in
alternating arm
LeBeau et al
96
TRT concurrent with
cycle 2 or alternating
among cycles 2, 3, 4,
and 5
CAE No difference in median survival No long-term data; poor
compliance in alternating arm
Work et al
97
TRT starting week 1 vs
week 18
CAV and PE No difference in median or 5-yr
survival
Early RT preceeded
chemotherapy and was not
concurrent with it
Samantas et al
98
week 13
CbpE No difference in median or 2-yr
survival
Reasonably good 2-year survival
rate (35%)
Jeremic et al
99
week 6
CbpE MST significantly better for
early TRT (34 vs 26 mo,
respectively)
Local control better with early
TRT
*A doxorubicin; C cyclophosphamide; Cbp carboplatin; E etoposide; P cisplatin; V vincristine; MST median survival time;
RT radiation therapy. Table modified and expanded from Wagner.
100
30 fractions for 3 weeks. Chemotherapy consisted
of four cycles of cisplatin-etoposide. The accelerated
regimen resulted in improved local control (intratho-
racic failure: accelerated therapy arm, 36%; standard
therapy arm, 52%) and long-term survival, which was
26% for the twice-daily regimen and 16% for the
standard regimen. There was an increased rate of grade
3 esophagitis (26% vs 11%, respectively), but there
were no other significant differences in toxicity.
40
The North Central Cancer Treatment Group
41
also compared twice-daily fractionation to daily frac-
tionation but with a significantly different twice-daily
regimen and overall study design. In both arms of
the study, radiation therapy was administered con-
currently with the fourth and fifth cycles of chemo-
therapy. The standard radiation regimen was 50.4 Gy
in 28 fractions for 5 weeks. Patients in the twice-
daily arm of the study received 48 Gy in three
fractions, with a 2.5-week split after the initial 24-Gy
dose. Thus, unlike the above-mentioned North
American Intergroup trial, there was no overall
acceleration of the radiation course. In this trial,
there were no differences in local control or survival
between patients in the two arms of the study.
Radiation Target Volume
SCLC often presents with bulky mediastinal ade-
nopathy and often with a confusing mixture of tumor
and atelectasis in lung parenchyma. Radiation target
volumes are often large, limiting the achievable
doses. In attempting to define the minimal appro-
priate dose, the following two issues can be consid-
ered:
1. Is elective radiation applied to normal-appear-
ing lymph nodes required? This has not been
studied prospectively. However, the North
American Intergroup trial, which produced the
best 5-year survival rate reported by a cooper-
ative group, limited elective radiation with no
intentional radiation to the contralateral hilum
or to supraclavicular nodes, unless there was
bulky superior mediastinal adenopathy.
42
2. If TRT is started after several cycles of chemo-
therapy, can the target volume comprise the
postchemotherapy rather than the prechemo-
therapy tumor volume? A retrospective re-
view
41,43,44
of data from the Mayo Clinic and
the North Central Cancer Treatment Group
suggests that this can be done, as recurrences
tended to be at the center of the tumor rather
that at the periphery. An earlier trial by the
Southwest Oncology Group,
45
which random-
ized patients having partial responses to radia-
tion to prechemotherapy or postchemotherapy
volumes, also reported no difference in recur-
rence rates.
Recommendations
12. Patients with limited-stage SCLC should be
referred to a radiation oncologist and a med-
ical oncologist for chemotherapy and radiation
therapy. Level of evidence, good; benefit,
13. Patients with limited-stage SCLC achieving a
CR should be offered PCI. Level of evidence,
good; benefit, substantial; grade of recom-
mendation, A
PCI
Brain metastases are common in SCLC. In pa-
tients who achieve a CR to induction therapy, CNS
metastases will emerge over the next 2 years in about
50 to 60% of patients, and 20 to 30% of these
metastases will be the sole site of disease recur-
rence.
46
Overt metastatic disease in the brain, while
often responding temporarily to radiation or chemo-
therapy, is rarely if ever cured. The hypothesis that
moderate doses of radiation given to patients without
detectable CNS involvement might eradicate occult
metastatic disease has been entertained for 20
years,
47
but only in the past few years have data
emerged to allow a reasonable consensus that PCI
can reduce the risk of CNS failure and improve
survival, and can do so without excessive toxicity.
4850
A meta-analysis of randomized trials of PCI in
complete responders (patients predominately with
limited disease) concluded that it significantly re-
duced CNS failure by about 50% and produced a
modest (about 5%) but also significant improvement
in the 3-year survival rate. There was a trend toward
better results with higher doses (ie, 30 to 36 Gy using
2-Gy fractions) than with 20-Gy doses, but this was
not a randomized comparison. An intergroup trial is
currently comparing 25-Gy doses administered in 10
fractions to 36-Gy doses administered in 18 fractions.
Earlier trials of PCI had variably reported late
neurotoxicity, with deterioration in memory, calcu-
lation ability, and quality of life. The relation of these
toxicities to treatment was unclear. In several more
recent trials
51
in which cognitive function was as-
sessed prospectively, significant differences between
SCLC patients and age-matched and gender-
matched control subjects have been observed prior
to any treatment, with up to 40% of patients showing
significant impairment. Significant further deteriora-
tion following PCI was not seen in a large 1997 trial
in the United Kingdom.
52
Van Oosterhout et al
53
performed careful neurologic and neurophysiologic
examinations of 59 survivors who were alive 2
years after receiving a diagnosis and who underwent
a cranial CT or MRI scan. Groups were neurophysi-
ologically compared with matched control subjects.
The authors concluded that although more inten-
sively systemically treated patients showed more
neurologic impairment, there was no statistical evi-
dence for additional neurotoxicity with PCI.
Recommendations
14. Patients with limited-stage SCLC achieving a
CR or patients who have undergone resection
who have stage I disease should be offered
PCI. Level of evidence, good; benefit, sub-
15. Patients with extensive-stage SCLC achieving
a CR should be offered PCI. Level of evi-
dence, fair; benefit, small; grade of recom-
mendation, C
Role of Surgery in Early Stage SCLC
The role of surgery in the treatment of early-
stage SCLC recently has been reviewed.
54
Surgery
as a primary modality of treatment was abandoned
after the British Medical Counsel
55
published the
results of their study comparing primary radiation
therapy with surgery in patients with resectable
SCLC with a 10-year follow-up. The overall sur-
vival was better for patients in the radiation
therapy-alone arm of the study, and there were no
long-term survivors among patients in the surgery
arm of the study. However, subsequent reports
published in the 1970s and early 1980s showed
long-term survival in patients who had been
treated with surgery alone who had very early-
stage disease. The most favorable subset of pa-
tients had T1N0 tumors that had been identified
either at the time of surgery or at the time of
postoperative pathologic examination.
5659
Even
though the role of adjuvant therapy has not been
evaluated in prospective randomized trials, there are
several reports
58,6066
suggesting a benefit for adju-
vant chemotherapy even in the earliest stages of the
disease.
The role of surgery in node-positive patients was
evaluated prospectively by the Lung Cancer Study
Group.
67
Patients with stage I disease (ie, T12,N0
tumors) were excluded from this trial. Patients were
initially treated with five cycles of CAV. Responding
patients were randomized to undergo surgery or not
to undergo surgery. All patients received radiation
therapy to the chest and brain. There was no differ-
ence in survival between the arms of the study. For
all patients, the median survival time was 15 months
and the 2-year survival rate was 20%.
All patients who are being worked up for surgery
should have undergo mediastinoscopy prior to
undergoing resection. The utility of mediastinos-
copy in SCLC patients has been validated recently
in a small prospective Japanese trial.
68
Recommendations
16. For the rare patient with very limited-stage
disease (ie, T12, N0 tumors), surgical resec-
tion followed by a platinum-based chemother-
apy could be offered. Level of evidence, fair;
benefit, small; grade of recommendation, C
17. Mediastinoscopy should be performed in all
patients undergoing surgical resection. Level
recommendation, C
18. PCI should be offered to patients achieving a
CR. Level of evidence, fair; benefit, small;
Conclusion
The incidence of SCLC has been decreasing, and
in 1998 it was reported to be 13.8% of all lung
cancers. A two-stage staging system is generally
utilized. Limited-stage SCLC is optimally treated
with a concurrent chemotherapy and radiation ther-
apy approach, and approximately 20% of patients are
cured. A platinum-based chemotherapy is the stan-
dard for treating extensive-stage SCLC. PCI pro-
vides a small absolute benefit in survival in patients
achieving CRs. Future research should be focused
on optimizing chemotherapy regimens and radiation
therapy schedules.
Staging of SCLC
1. In all patients, routine staging of SCLC should
include history and physical examinations, com-
plete blood counts, a comprehensive chemistry
panel, a CT scan of the chest and abdomen, a
CT or MRI scan of the brain, and a bone scan.
2. For the routine staging of patients with SCLC,
PET scanning is not recommended outside of a
clinical trial. Level of evidence, fair; benefit,
Treatment of Extensive-Stage SCLC
First-Line Treatment:
3. Patients with extensive stage disease should
receive platinum-based chemotherapy. Level
4. Patients achieving CRs should be offered PCI.
Level of evidence, fair; benefit, small; grade of
recommendation, C
Maintenance Treatment:
5. For patients with extensive-stage or limited-
stage SCLC achieving a partial response or a
CR, there is no evidence, outside of a clinical
trial, for the use of maintenance treatment.
Treatment of Relapsed or Refractory SCLC
6. Patients with SCLC who have relapsed follow-
ing an initial response to treatment or who are
refractory to the initial treatment should be
offered further chemotherapy. The chemo-
therapy offered will depend on the duration of
the response after receiving first-line chemo-
therapy or the lack of response to first-line
chemotherapy (ie, sensitive relapses vs refrac-
tory patients). Level of evidence, fair; benefit,
small/weak; grade of recommendation, C
Treatment of Elderly ( 70 years of age) Patients
With Extensive-Stage SCLC
7. Elderly patients with good performance status
and with intact organ function should be
treated with platinum-based chemotherapy.
Level of evidence, good; benefit, moderate;
8. Elderly patients with poor prognostic factors
such as poor performance status or severe con-
comitant comorbid disease may still be consid-
ered for chemotherapy. Level of evidence, poor;
9. Elderly patients achieving CRs should be of-
fered PCI. Level of evidence, fair; benefit,
Dose Intensity in SCLC
10. For patients with either extensive-stage or lim-
ited-stage SCLC, there is no role for the admin-
istration of dose-dense/intense, initial/induction,
or maintenance treatment outside of a clinical
trial. Level of evidence, good; benefit, none/
The Role of Growth Factor and the Use of Stem
Cell Support in SCLC
11. In patients with SCLC who are receiving
chemotherapy, the routine use of growth fac-
tor is not recommended. Level of evidence,
good; benefit, none/negative; grade of recom-
mendation, D
Treatment of Limited-Stage SCLC
12. Patients with limited-stage SCLC should be
referred to a radiation oncologist and a med-
ical oncologist for chemotherapy and radiation
therapy. Level of evidence, good; benefit,
13. Patients with limited-stage SCLC achieving
CRs should be offered PCI. Level of evi-
recommendation, A
PCI
14. Patients with limited-stage SCLC achieving
CRs or patients who have undergone resec-
tion who have stage I disease should be
offered PCI. Level of evidence, good; ben-
efit, substantial; grade of recommendation,
A
15. Patients with extensive-stage SCLC achieving
CRs should be offered PCI. Level of evi-
dence, fair; benefit, small; grade of recom-
mendation, C
Role of Surgery in Early-Stage SCLC
16. For the rare patient with very limited-stage
disease (ie, T12,N0 tumors), surgical resec-
tion followed by platinum-based chemother-
apy could be offered. Level of evidence, fair;
17. Mediastinoscopy should be performed in all
patients undergoing surgical resection. Level
recommendation, C
18. PCI should be offered to patients achieving
CRs. Level of evidence, fair; benefit, small;
References
1 Page NC, Read WL, Tierney RM, et al. The epidemiology of
small cell lung cancer [abstract]. Proc Am Soc Clin Oncol
2002; 21:305a
2 Hauber HP, Bohuslavizki KH, Lund CH, et al. Positron
emission tomography in the staging of small-cell lung can-
cer: a preliminary study. Chest 2001; 119:950954
3 Schumacher T, Brink I, Mix M, et al. FDG-PET imaging for
the staging and follow-up of small cell lung cancer. Eur
J Nucl Med 2001; 28:483488
4 Pujol JL, Carestia L, Daures JP. Is there a case for cisplatin
in the treatment of small-cell lung cancer?: a meta-analysis
of randomized trials of a cisplatin-containing regimen vs a
regimen without this alkylating agent. Br J Cancer 2000;
83:815
5 Berghmans T, Paesmans M, Mascaux C, et al. A meta-
analysis of the role of etoposide (VP-16) and cisplatin
(CDDP) in small cell lung cancer (SCLC) with a method-
ology assessment [abstract]. Eur J Cancer 1999; 35(suppl):
s248
6 Chute JP, Chen T, Feigal E, et al. Twenty years of phase III
trials for patients with extensive-stage small-cell lung cancer:
perceptible progress. J Clin Oncol 1999; 17:17941801
7 Brahmer JR, Ettinger DS. Carboplatin in the treatment of
lung cancer. Oncologist 1998; 3:143154
8 Noda K, Nishiwaki Y, Kawahara M, et al. Irinotecan plus
cisplatin compared with etoposide plus cisplatin for exten-
sive small cell lung cancer. N Engl J Med 2002; 346:8591
9 Loehrer PJ Sr, Ansari R, Gonin R, et al. Cisplatin plus
etoposide with and without ifosfamide in extensive small-cell
lung cancer: a Hoosier Oncology Group study. J Clin Oncol
1995; 13:25942599
10 Mavroudis D, Papadakis E, Veslemes M, et al. A multicenter
randomized clinical trial comparing paclitaxel-cisplatin-
etoposide vs cisplatin-etoposide as first-line treatment in
patients with small-cell lung cancer. Ann Oncol 2001;
12:463470
11 Niell HB, Herndon JE, Miller AA, et al. Randomized phase
III intergroup trial (CALGB 9732) of etoposide (VP-16) and
cisplatin (DDP) with or without paclitaxel (TAX) and G-CSF
in patients with extensive stage small cell lung cancer (ED-
SCLC) [abstract]. Proc Am Soc Clin Oncol 2002; 21:293a
12 Sculier JP, Berghmans T, Castaigne C, et al. Maintenance
chemotherapy for small cell lung cancer: a critical review of
the literature. Lung Cancer 1998; 19:141151
13 Schiller JH, Adak S, Cella D, et al. Topotecan vs observation
after cisplatin plus etoposide in extensive-stage small-cell
lung cancer: E7593; a phase III trial of the Eastern Coop-
erative Oncology Group. J Clin Oncol 2001; 19:21142122
14 Yuen AR, Zou G, Turrisi AT, et al. Similar outcome of
elderly patients in intergroup trial 0096: cisplatin, etoposide,
and thoracic radiotherapy administered once or twice daily
in limited stage small cell lung carcinoma. Cancer 2000;
89:19531960
15 Quon H, Shepherd FA, Payne DG, et al. The influence of
age on the delivery, tolerance, and efficacy of thoracic
irradiation in the combined modality treatment of limited
stage small cell lung cancer. Int J Radiat Oncol Biol Phys
1999; 43:3945
16 Ando M, Minami H, Ando Y, et al. Pharmacological analysis
of etoposide in elderly patients with lung cancer. Clin
Cancer Res 1999; 5:16901695
17 Girling DJ. Comparison of oral etoposide and standard
intravenous multidrug chemotherapy for small-cell lung
cancer: a stopped multicentre randomised trial; Medical
Research Council Lung Cancer Working Party. Lancet
1996; 348:563566
18 Souhami RL, Spiro SG, Rudd RM, et al. Five-day oral
etoposide treatment for advanced small-cell lung cancer:
randomized comparison with intravenous chemotherapy.
19 Matsui K, Masuda N, Fukuoka M, et al. Phase II trial of
carboplatin plus oral etoposide for elderly patients with
small-cell lung cancer. Br J Cancer 1998; 77:19611965
20 Murray N, Grafton C, Shah A, et al. Abbreviated treatment
for elderly, infirm, or noncompliant patients with limited-
stage small-cell lung cancer. J Clin Oncol 1998; 16:3323
3328
21 Westeel V, Murray N, Gelmon K, et al. New combination of
the old drugs for elderly patients with small-cell lung cancer:
a phase II study of the PAVE regimen. J Clin Oncol 1998;
16:19401947
22 Steele JP. Gemcitabine/carboplatin vs cisplatin/etoposide for
patients with poor-prognosis small cell lung cancer: a phase
III randomized trial with quality-of-life evaluation. Semin
Oncol 2001; 28(suppl):1518
23 White SC, Lorigan P, Middleton MR, et al. Randomized phase
II study of cyclophosphamide, doxorubicin, and vincristine
compared with single-agent carboplatin in patients with poor
prognosis small cell lung carcinoma. Cancer 2001; 92:601608
24 Sculier JP, Paesmans M, Bureau G, et al. Multiple-drug
weekly chemotherapy vs standard combination regimen in
small-cell lung cancer: a phase III randomized study con-
ducted by the European Lung Cancer Working Party. J Clin
Oncol 1993; 11:18581865
25 Souhami RL, Rudd R, Ruiz de Elvira MC, et al. Random-
ized trial comparing weekly vs 3-week chemotherapy in
small- cell lung cancer: a Cancer Research Campaign trial.
J Clin Oncol 1994; 12:18061813
26 Fukuoka M, Masuda N, Negoro S, et al. CODE chemother-
apy with and without granulocyte colony-stimulating factor
in small-cell lung cancer. Br J Cancer 1997; 75:306309
27 Murray N, Shah A, Osoba D, et al. Intensive weekly
chemotherapy for the treatment of extensive-stage small-cell
28 Furuse K, Fukuoka M, Nishiwaki Y, et al. Phase III study of
intensive weekly chemotherapy with recombinant human
granulocyte colony-stimulating factor vs standard chemo-
therapy in extensive-disease small-cell lung cancer: the
Japan Clinical Oncology Group. J Clin Oncol 1998; 16:
21262132
29 Skarlos DV, Samantas E, Pectasides D, et al. Weekly
alternating non-cross-resistant chemotherapy for small cell
lung cancer with a good prognosis: a study of the Hellenic
Cooperative Oncology Group. Am J Clin Oncol 1999;
22:8793
30 Pujol JL, Douillard JY, Riviere A, et al. Dose-intensity of a
four-drug chemotherapy regimen with or without recombi-
nant human granulocyte-macrophage colony-stimulating
factor in extensive-stage small-cell lung cancer: a multi-
center randomized phase III study. J Clin Oncol 1997;
15:20822089
31 Steward WP, von Pawel J, Gatzemeier U, et al. Effects of
granulocyte-macrophage colony-stimulating factor and dose
intensification of V-ICE chemotherapy in small-cell lung
cancer: a prospective randomized study of 300 patients.
J Clin Oncol 1998; 16:642650
32 Thatcher N, Girling DJ, Hopwood P, et al. Improving
survival without reducing quality of life in small-cell lung
cancer patients by increasing the dose-intensity of chemo-
therapy with granulocyte colony-stimulating factor support:
results of a British Medical Research Council Multicenter
Randomized Trial; Medical Research Council Lung Cancer
Working Party. J Clin Oncol 2000; 18:395404
33 Arriagada R, Le Chevalier T, Pignon JP, et al. Initial
chemotherapeutic doses and survival in patients with limited
small-cell lung cancer. N Engl J Med 1993; 329:18481852
34 Chouaid C, Bassinet L, Fuhrman C, et al. Routine use of
granulocyte colony-stimulating factor is not cost-effective
and does not increase patient comfort in the treatment of
small-cell lung cancer: an analysis using a Markov model.
J Clin Oncol 1998; 16:27002707
35 Warde P, Payne D. Does thoracic irradiation improve survival
and local control in limited-stage small-cell carcinoma of the
lung? A meta-analysis. J Clin Oncol 1992; 10:890895
36 Pignon JP, Arriagada R, Ihde DC, et al. A meta-analysis of
thoracic radiotherapy for small-cell lung cancer. N Engl
J Med 1992; 327:16181624
37 Murray N, Coldman A. The relationship between the tho-
racic irradiation timing and long-term survival in combined
modality therapy of limited stage small cell lung can-
cer(LSCLC) [abstract]. Proc Am Soc Clin Oncol 1995; 14
38 Choi NC, Carey RW. Importance of radiation dose in
achieving improved loco-regional tumor control in limited
stage small-cell lung carcinoma: an update. Int J Radiat
Oncol Biol Phys 1989; 17:307310
39 Choi NC, Herndon JE 2nd, Rosenman J, et al. Phase I study
to determine the maximum-tolerated dose of radiation in
standard daily and hyperfractionated-accelerated twice-daily
radiation schedules with concurrent chemotherapy for limited-
stage small-cell lung cancer. J Clin Oncol 1998; 16:35283536
40 Turrisi AT 3rd, Kim K, Blum R, et al. Twice-daily compared
with once-daily thoracic radiotherapy in limited small-cell
lung cancer treated concurrently with cisplatin and etopo-
side. N Engl J Med 1999; 340:265271
41 Maksymiuk AW, Jett JR, Earle JD, et al. Sequencing and
schedule effects of cisplatin plus etoposide in small-cell lung
cancer: results of a North Central Cancer Treatment Group
randomized clinical trial. J Clin Oncol 1994; 12:7076
42 Turrisi AT 3rd, Glover DJ, Mason BA. A preliminary report:
concurrent twice-daily radiotherapy plus platinum- etopo-
side chemotherapy for limited small cell lung cancer. Int J
Radiat Oncol Biol Phys 1988; 15:183187
43 Bonner JA, Sloan JA, Shanahan TG, et al. Phase III
comparison of twice-daily split-course irradiation vs once-
daily irradiation for patients with limited stage small-cell
lung carcinoma. J Clin Oncol 1999; 17:26812691
44 Liengswangwong V, Bonner JA, Shaw EG, et al. Limited-
stage small-cell lung cancer: patterns of intrathoracic recur-
rence and the implications for thoracic radiotherapy. J Clin
Oncol 1994; 12:496502
45 Kies MS, Mira JG, Crowley JJ, et al. Multimodal therapy for
limited small-cell lung cancer: a randomized study of induc-
tion combination chemotherapy with or without thoracic
radiation in complete responders; and with wide-field vs
reduced- field radiation in partial respondersa Southwest
Oncology Group Study. J Clin Oncol 1987; 5:592600
46 Arriagada R, Le Chevalier T, BF, et al. Randomized trial of
on prophylactic cranial irradiation (PCI) for patients with
small cell lung cancer (SCLC) in complete remission [ab-
stract]. Proc Am Soc Clin Oncol 1994; 13
47 Hansen HH, Dombernowsky P, Hirsch FR, et al. Prophylactic
irradiation in bronchogenic small cell anaplastic carcinoma: a
comparative trial of localized vs extensive radiotherapy includ-
ing prophylactic brain irradiation in patients receiving combi-
nation chemotherapy. Cancer 1980; 46:279284
48 Auperin A, Arriagada R, Pignon JP, et al. Prophylactic cranial
irradiation for patients with small-cell lung cancer in complete
remission: Prophylactic Cranial Irradiation Overview Collabo-
rative Group. N Engl J Med 1999; 341:476484
49 Kotalik J, Yu E, Markman BR, et al. Practice guideline on
prophylactic cranial irradiation in small-cell lung cancer. Int
50 PCIOC G. Cranial Irradiation for preventing brain metas-
tases of small cell lung cancer in complete remission.
Cochrane Database Syst Rev 2000; 4:CD002805
51 Cull A, Gregor A, Hopwood P, et al. Neurological and
cognitive impairment in long-term survivors of small cell
lung cancer. Eur J Cancer 1994; 8:10671074
52 Gregor A, Cull A, Stephens RJ, et al. Prophylactic cranial
irradiation is indicated following complete response to in-
duction therapy in small cell lung cancer: results of a
multicentre randomised trial; United Kingdom Coordinating
Committee for Cancer Research (UKCCCR) and the Euro-
pean Organization for Research and Treatment of Cancer
(EORTC). Eur J Cancer 1997; 33:17521758
53 Van Oosterhout AG, Ganzevles PG, Wilmink JT, et al.
Sequelae in long-term survivors of small cell lung cancer. Int
54 Simon GR, Ginsberg RJ, Ruckdeschel J. Small-cell lung
cancer. Chest Surg Clin N Am 2001; 11:165188
55 Fox W, Scadding JG. Medical Research Council compara-
tive trial of surgery and radiotherapy for primary treatment
of small-celled or oat-celled carcinoma of bronchus: ten-year
follow-up. Lancet 1973; 2:6365
56 Higgins GA, Shields TW, Keehn RJ. The solitary pulmonary
nodule: ten-year follow-up of veterans administration-armed
forces cooperative study. Arch Surg 1975; 110:570575
57 Lennox SC, Flavell G, Pollock DJ, et al. Results of resection
for oat-cell carcinoma of the lung. Lancet 1968; 2:925927
58 Miyazawa N, Tsuchiya R, Naruke T, et al. A clinico-
pathological study of surgical treatment for small cell carci-
noma of the lung. Jpn J Clin Oncol 1986; 16:297307
59 Shields TW, Higgins GA Jr, Matthews MJ, et al. Surgical
resection in the management of small cell carcinoma of the
lung. Thorac Cardiovasc Surg 1982; 84:481488
60 Lucchi M, Mussi A, Chella A, et al. Surgery in the manage-
ment of small cell lung cancer. Eur J Cardiothorac Surg
1997; 12:689693
61 Merkle NM, Mickisch GH, Kayser K, et al. Surgical resec-
tion and adjuvant chemotherapy for small cell carcinoma.
62 Maassen W, Greschuchna D. Small cell carcinoma of the
lung: to operate or not?; surgical experience and results.
63 Karrer K, Shields TW, Denck H, et al. The importance of
surgical and multimodality treatment for small cell bronchial
carcinoma. Thorac Cardiovasc Surg 1989; 97:168176
64 Osterlind K, Hansen M, Hansen HH, et al. Influence of
surgical resection prior to chemotherapy on the long-term
results in small cell lung cancer: a study of 150 operable
patients. Eur J Cancer Clin Oncol 1986; 22:589593
65 Shepherd FA, Ginsberg RJ, Feld R, et al. Surgical treatment
for limited small-cell lung cancer: the University of Toronto
Lung Oncology Group experience. Thorac Cardiovasc Surg
1991; 101:385393
66 Friess GG, McCracken JD, Troxell ML, et al. Effect of
initial resection of small-cell carcinoma of the lung: a review
of Southwest Oncology Group Study 7628. J Clin Oncol
1985; 3:964968
67 Lad T, Piantadosi S, Thomas P, et al. A prospective random-
ized trial to determine the benefit of surgical resection of
residual disease following response of small cell lung cancer
to combination chemotherapy. Chest 1994; 106(suppl):
320S323S
68 Inoue M, Nakagawa K, Fujiwara K, et al. Results of
preoperative mediastinoscopy for small cell lung cancer.
Ann Thorac Surg 2000; 70:16201623
69 Nakanishi Y, Kawarada Y, Hirose N, et al. Phase II trial of
combination chemotherapy with cisplatin, carboplatin and
etoposide in stage IIIB and IV small-cell lung cancer:
Fukuoka Lung Cancer Study Group. Cancer Chemother
Pharmacol 1998; 41:453456
70 Glisson BS, Kurie JM, Perez-Soler R, et al. Cisplatin,
etoposide, and paclitaxel in the treatment of patients with
extensive small-cell lung carcinoma. J Clin Oncol 1999;
17:23092315
71 Kelly K, Crowley J, Bunn PA, et al. Arandomized phase III trial
of paclitaxel plus carboplatin (PC) versus vinorelbine plus
cisplatin (VC) in untreated advanced non-small cell lung cancer
(NSCLC): a South West Oncology Group (SWOG) Trial
72 Kudoh S, Fujiwara Y, Takada Y, et al. Phase II study of
irinotecan combined with cisplatin in patients with previ-
ously untreated small-cell lung cancer. J Clin Oncol 1998;
16:10681074
73 Hickish TF SE, Nicolson MC, et al. A pilot study of MVP
(mitomycin-C, vinblastine and cisplatin) chemotherapy in
small-cell lung cancer. Br J Cancer 1998; 77:19661970
74 Kalemkerian GP, Jiroutek M, Ettigner DS, et al. A phase II
study of all-trans-retinoic acid plus cisplatin and etoposide in
patients with extensive stage lung carcinoma. Cancer 1998;
83:11021108
75 Lyss AP, Herndon JE, Lynch TC, et al. Paclitaxel cisplatin
G-CSF in patients with previously untreated extensive
stage small cell lung cancer (E-SCLC): preliminary analysis
of Cancer and Leukemia Group B (CALBG) 9430 [ab-
stract]. Proc Am Soc Clin Oncol 1999; 18:468a
76 Jacobs SA, Jett JR, Belani CP, et al. Topotecan and paclitaxel,
and active couplet, in untreated extensive disease small cell
lung cancer [abstract]. Proc Am Soc Clin Oncol 1999; 18:470a
77 Nakamura S, Kudoh S, Komuta K, et al Phase II. study of
irinotecan (CPT-11) combined with etoposide (VP-16) for
previously untreated extensive-disease small-cell lung cancer
(ED-SCLC) [abstract]. Proc Am Soc Clin Oncol 1999; 18:470S
78 Deppermanan KM, Serke M, Oehm C, et al. Paclitaxel
(TAX), and carboplatin (CBDA) in advaned SCLC: a phase
II study [abstract]. Proc Am Soc Clin Oncol 1999; 48:482a
79 Rushing DA, Tipping SJ, Clouse LH, et al. Phase 1 study of the
combination of irinotecan (CPT-11) and paclitaxel (Taxol) in
previously treated patients with small cell carcinoma of the lung
(SCCL) [abstract]. Proc Am Soc Clin Oncol 1999; 18:484a
80 Lad T, Mauer A, Hoffman P, et al. Phase II trial of paclitaxel
and doxorubicin in small cell lung cancer [abstract]. Proc
Am Soc Clin Oncol 1999; 18:513a
81 Lianes P, Moreno JA, Sevilla I, et al. Phase II study of
docetaxel and cisplatin in first line treatment of disseminated
small cell lung cancer (E-SCLC) [abstract]. Proc Am Soc
82 Lynch TJ, Herndon JE, Lilenbaum RD, et al. Toxicity of
paclitaxel (P) and topotecan (T) in patients with previously
untreated extensive small cell lung cancer [abstract]. Proc
83 Tweedy CR, Andrews DR, Ball T. Topotecan and paclitaxel
in extensive stage small cell lung cancer as initial therapy
84 Panza N, Frasci G, Comella P, et al. Cisplatin-paclitaxel-
topotecan (CPT) weekly administration in chemonaive or
pretreated extensive disease small cell lung cancer (ED-
SCLC). Eur J Cancer 1999; 35:S253
85 Baldini E, Chella A, del Freo A, et al. Paclitaxel/epirubicin/
etoposide in patients with extensive-disease small-cell lung
cancer (SCLC) [abstract]. Eur J Cancer 1999; 35:S257
86 Masuda N, Matsui K, Negoro S, et al. Combination of irinote-
can and etoposide for treatment of refractory or relapsed
small-cell lung cancer. J Clin Oncol 1998; 16:33293334
87 Ardizzoni A, Manegold G, Gaffaar R, et al. Combination
chemotherapy with cisplatin and topotecan as second-line
treatment of sensitive and refractory small cell lung cancer-
(SCLC): an EORTC LCCG phase II study [abstract]. Proc
88 Lusch G, Rausch G, and Kosierowski R. Phase II study of
daily oral etoposide and hexamethylmelamine (altretamine)
in relapsed small cell lung cancer (SCLC) [abstract]. Proc
89 Rushing DA, Tipping SJ, Close LH, et al. Phase I study of
combination of irinotecan (CPT-11) and paclitaxel (Taxol) in
previously treated patients with small cell lung cancer
(SCLC) [abstract]. Proc Am Soc Clin Oncol 1999; 18:484a
90 Agelaki S, Agelidou A, Blazogiannakis G, et al. A phase II
study of paclitaxel (P) and carboplatin (C) as second-line
treatment in patients with small cell lung cancer. Eur J
Cancer 1999; 35:S258
91 Perry MC, Herndon JE 3rd, Eaton WL, et al. Thoracic
radiation therapy added to chemotherapy for small-cell lung
cancer: an update of Cancer and Leukemia Group B Study
8083. J Clin Oncol 1998; 16:24662467
92 Murray N, Coy P, Pater JL, et al. Importance of timing for
thoracic irradiation in the combined modality treatment of
limited-stage small-cell lung cancer: the National Cancer
Institute of Canada Clinical Trials Group. J Clin Oncol 1993;
11(2):336344
93 Shultz HP, Nielsen OS, Sell A, et al. Timing of chest
irradiation with respect to combination chemotherapy in
small cell lung cancer, limited disease. [abstract]. Lung
Cancer 1988; 4:153
94 Tsukada H, Yokoyama A, Goto K, et al. Concurrent vs
sequential radiotherapy for small cell lung cancer. Semin
Oncol 2001; 28(suppl):2326
95 Gregor A, Drings P, Burghouts J, et al. Randomized trial of
alternating vs sequential radiotherapy/chemotherapy in limited-
disease patients with small-cell lung cancer: a European Orga-
nization for Research and Treatment of Cancer Lung Cancer
Cooperative Group Study. J Clin Oncol 1997; 15:28402849
96 LeBeau B, Chastang C, Urban T, et al. A randomized
clinical trial comparing concurrent and alternated thoracic
irradiation in limited stage small cell lung cancer [abstract].
Proc Am Soc Clin Oncol 1996; 15:383
97 Work E, Nielsen OS, Bentzen SM, et al. Randomized study
of initial vs late chest irradiation combined with chemother-
apy in limited-stage small-cell lung cancer: Aarhus Lung
Cancer Group. J Clin Oncol 1997; 15:30303037
98 Samantas E, Fountzilas G, Briasoulis T, et al. Randomized
comparison of early vs late twice-daily chest irradiation therapy
concurrently with chemotherapy in limited disease small cell
lung cancer: final results [abstract]. Proc Am Soc Clin Oncol
2000; 491A
99 Jeremic B, Shibamoto Y, Acimovic L et al. Initial vs delayed
accelerated hyperfractionated radiation therapy and concur-
rent chemotherapy in limited small-cell lung cancer: a
randomized study. J Clin Oncol 1997; 15:893900
100 Wagner H Jr. Radiation therapy in the management of
limited small cell lung cancer: when, where, and how much?
Chest 1998; 113(suppl):92S100S
101 von Pawel J, Schiller JH, Shepherd FA, et al. Topotecan vs
cyclophosphamide, doxorubicin, and vincristine for the
treatment of recurrent small-cell lung cancer. J Clin Oncol
1999; 17:658667
102 Schirrmeister H, Glatting G, Hetzel J, et al. Prospective
evaluation of the clinical value of planar bone scans, SPECT,
and (18)F-labeled NaF PET in newly diagnosed lung can-
cer. J Nucl Med 2001; 42:18001804
103 Kosmidis PA, Samantas E, Fountzilas G, et al. Cisplatin/
etoposide vs carboplatin/etoposide chemotherapy and irra-
diation in small cell lung cancer: a randomized phase III
study. Semin Oncol 1994; 21(suppl):2330
2003;123;259-271 Chest
George R. Simon and Henry Wagner
Small Cell Lung Cancer
& Services
Updated Information
S
References
S#BIBL
free at:
Citations
S#otherarticles
Reprints
2003;123;272-283 Chest
Gene L. Colice, Jeffrey Rubins and Michael Unger
Curative-Intent Therapy
Follow-up and Surveillance of the Lung Cancer Patient Following
ISSN: 0012-3692.
Follow-up and Surveillance of the Lung
Cancer Patient Following Curative-
Intent Therapy*
Gene L. Colice, MD, FCCP; Jeffrey Rubins, MD, FCCP; and
Michael Unger, MD, FCCP
The following two distinctly different issues should be taken into account when planning patient
care following curative-intent therapy for lung cancer: adequate follow-up to manage complica-
tions related to the curative-intent therapy; and surveillance to detect recurrences of the primary
lung cancer and/or development of a new primary lung cancer early enough to allow potentially
curative retreatment. Follow-up for complications should be performed by the specialist
responsible for the curative-intent therapy and should last 3 to 6 months. Recurrences of the
original lung cancer will be more likely during the first 2 years after curative-intent therapy, but
there will be an increased lifelong risk of approximately 1 to 2% per year of developing a
metachronous, or new primary, lung cancer. A standard surveillance program for these patients
is recommended based on periodic visits, with chest-imaging studies and counseling patients on
symptom recognition. Whether subgroups of patients with a higher risk of developing a
metachronous lung cancer (eg, those patients whose primary lung cancer was radiographically
occult or central and those patients surviving for > 2 years after treatment for small cell lung
cancer) should have a more intensive surveillance program is presently unclear. The surveillance
program should be coordinated by a multidisciplinary tumor board and overseen by the physician
who diagnosed and initiated therapy for the original lung cancer. Smoking cessation is recom-
mended for all patients following curative-intent therapy for lung cancer.
(CHEST 2003; 123:272S283S)
Key words: lung cancer; metachronous tumors; recurrence; surveillance
Abbreviations: NSCLC non-small cell lung cancer; PET positron emission tomography
A
pproximately 170,000 new cases of lung cancer
are diagnosed annually in the United States.
1
Unfortunately, only about 50 to 55% of patients with
newly diagnosed lung cancer will have localized
disease and will be candidates for potentially curative
treatment.
2
Furthermore, some patients with local-
ized non-small cell lung cancer (NSCLC) either may
refuse potentially curative surgical therapy or may be
unable to tolerate surgery because of limiting comor-
bid cardiopulmonary disease or other diseases. Con-
sequently, it has been estimated that only 35,000
patients underwent curative-intent surgical resection
for NSCLC in 1998.
3
Smaller numbers of patients
will receive curative-intent radiation therapy for
localized NSCLC and some combination of curative-
intent chemotherapy and radiation therapy for the
treatment of localized small cell carcinoma.
Two distinctly different issues should be taken into
account when planning patient care following cura-
tive-intent therapy for lung cancer. First, adequate
follow-up should be ensured to manage complica-
tions related to the curative-intent therapy itself.
This should be a specialist-directed process. The
thoracic surgeon should be responsible for managing
complications related to any surgical procedures that
have been performed, as should the radiation oncol-
ogist and the oncologist be responsible for managing
complications related to radiation therapy and che-
motherapy, respectively. In most cases, this specialist-
directed follow-up should be transient.
Second, a surveillance program should be consid-
ered to detect recurrences of the primary lung
cancer and/or the development of a new primary
lung cancer early enough to allow potentially cura-
tive retreatment. Numerous guidelines have been
published regarding the management of lung cancer.
*From Critical Care and Respiratory Services (Dr. Colice),
Washington Hospital Center, Washington, DC; Pulmonary Divi-
sion (Dr. Rubins), Minneapolis Veterans Affairs Medical Center,
Minneapolis, MN; and Pulmonary Cancer Detection and Preven-
tion Program (Dr. Unger), Fox Chase Cancer Center, Philadel-
phia, PA.
Correspondence to: Gene L. Colice, MD, FCCP, Washington
Hospital Center, 110 Irving St NW, Washington, DC 20010;
e-mail: gxc@mhg.edu
Several of these guidelines include recommenda-
tions for a posttreatment surveillance program.
These recommendations will be summarized and
compared. Available data on rates, patterns, and
diagnostic tools for identifying a recurrence of the
primary lung cancer and/or the development of a
second primary lung cancer will be reviewed as the
basis for recommendations on an ongoing surveil-
lance program following curative-intent therapy for
lung cancer. Issues related to follow-up for palliative
therapy of lung cancer will not be discussed (see
article on palliative treatment in this supplement).
Methods and Definitions
Methods
Published guidelines on lung cancer diagnosis and manage-
ment were identified by a systematic review of the literature and
were evaluated (see article on methods and grading in this
supplement). Those guidelines, including recommendations that
are specific to the follow-up and surveillance of patients with lung
cancer after receiving curative-intent therapy, were identified for
inclusion in this section. Supplemental material appropriate to
this topic was obtained by a literature search of a computerized
database (MEDLINE) and a review of the reference lists of
relevant articles. Recommendations were developed by the sec-
tion editor and writing committee, were graded by a standardized
method (see article on methods and grading in this supplement)
and then reviewed by all section editors, the Chair of the lung
cancer panel, and the Co-Chair of the lung cancer panel.
Definitions
A difficult issue in the surveillance of the lung cancer patient
following curative-intent therapy is distinguishing between a
recurrence of the original lung cancer and the identification of a
new primary, or metachronous, lung cancer. Martini and Mel-
amed
4
proposed criteria for making this distinction in 1975. One
confusing aspect of these criteria was the inclusion of synchro-
nous lung cancers, which were described as physically distinct
and separate from the primary tumor. In the original series by
Martini and Melamed,
4
15 of the 18 patients with synchronous
lung cancers were identified at the time of the initial treatment.
In current lung cancer staging terminology, these cancers would
have been described either as satellite tumor nodules, if they were
within the same lobe as the primary tumor, or as intrapulmonary
metastases, if they were not within the same lobe.
2
With current
imaging capability, synchronous lung cancer usually would be
discovered prior to the performance of curative-intent surgical
resection of the primary lung cancer. Hence, synchronous lung
cancers should not be considered an issue in the surveillance of
the lung cancer patient following curative-intent therapy.
There are also difficulties with the criteria that Martini and
Melamed
4
used for diagnosing metachronous tumors (Table 1). If
the primary lung cancer had a mixed histology, the histologic
pattern of a second cancer might not adequately distinguish a
recurrence from a metachronous tumor. After curative-intent
surgical resection, it would not be possible for a newly recognized
cancer to have intrapulmonary lymphatics in common with the
original lung cancer. Because systematic mediastinal lymph node
sampling is included in the procedure for curative-intent lung
cancer surgery, identifying mediastinal nodes in common be-
tween the new and old lung cancer also would be problematic.
The choice of a tumor-free interval of 2 years for distinguishing
a metachronous lung cancer from a recurrence of lung cancer
with similar histology was arbitrary. Although the most appropri-
ate tumor-free interval for making this distinction has not been
defined (nor has it even been determined whether such an
interval is possible to define), Detterbeck and colleagues have
suggested (see article on special treatment issues in this supple-
ment) that a 4-year interval might be more appropriate. Based on
these considerations, it might be appropriate to revise the criteria
of Martini and Melamed
4
for identifying metachronous tumors
(Table 1). Whichever criteria are used, Martini and Melamed
remind us that the distinction between a new primary lung cancer
and a recurrence of the original lung cancer is not as important as
determining whether the tumor can be treated with curative
intent.
4
Current Guidelines
Four guidelines
58
were identified that included
specific recommendations for surveillance methods
in patients with NSCLC following curative-intent
therapy. Two guidelines
8,9
provided specific recom-
mendations for patients with small cell lung cancer.
These guidelines were developed by a consensus of
expert panels. In addition, published information is
available on surveillance methods for patients follow-
ing curative-intent therapy for NSCLC that have
been used by two leading cancer institutes in the
United States and one in Japan.
1012
The specific
recommendations from the guidelines and institu-
tional practices are summarized in Tables 2 and 3.
One other guideline
13
provided only the general
recommendation that respiratory physicians should
develop with their colleagues an explicit follow-up
Table 1Distinguishing Between Recurrence of the
Original Lung Cancer and Development of a New
Lung Cancer During Surveillance*
Metachronous Tumors
Criteria of Martini and Melamed
4
Proposed Revision
Histology different Histology different
Histology the same, if: Histology the same, if:
Free interval between cancers
was at least 2 years or origin
is from carcinoma in situ or,
second cancer is in different
lobe or lung,
Free interval between
cancers was at least 4
years, or origin is from
carcinoma in situ, and
there were no
extrapulmonary
metastases at the time of
diagnosis
But, no carcinoma in
lymphatics was common to
both and there were no
extrapulmonary metastases at
the time of diagnosis
*This table was adapted from the American Society of Clinical
Oncology.
5
policy that would be appropriate to the needs and
resources of the patient and health-care providers.
The guidelines and institutional practices have
remarkable similarities. Each recommends more fre-
quent visits during the first 2 years following cura-
tive-intent therapy. Visits are less frequent for years
3 through 5 and decrease to a minimal level of
annual visits after year 5. This pattern of visits is
based on the expectation that recurrences of the
original lung cancer will be more likely during the
first 2 years after curative-intent therapy but that
there will be an increased lifelong risk of developing
a new primary lung cancer. Each of the guidelines
and the institutional practices emphasizes symptoms
that are elicited in the patients medical history as an
extremely important indication of recurrence. The
physical examination is included as an adjunctive,
but less valuable, tool for identifying recurrences or
new primaries. All the guidelines and institutional
practices include the chest radiograph as a surveil-
lance technique with slight variations as to how often
the chest radiograph should be performed.
Most guidelines and institutional practices agree
that chest CT scans should not be routinely per-
formed during these surveillance visits. One guide-
line
6
suggested that a chest CT scan should be
performed at baseline, soon after curative-intent
therapy, for comparison purposes if suspicion should
arise either of a recurrence or a new primary tumor.
Another guideline
7
recommended an annual spiral
chest CT scan as a surveillance method, although
disagreement was noted within the panel about the
added value of this approach. One guideline
8
and
one institutional practice
11
incorporated regular
complete blood counts and serum chemistry mea-
surements into surveillance monitoring, but other
groups found little value in performing these tests
routinely. The National Kyushu Cancer Center
11
Table 3Specific Recommendations for Surveillance Methods in Patients with Small Cell Lung Cancer Following
Curative-Intent Therapy*
Source
Surveillance
First 2 Years Years 35 After Year 5
NCCN
9
Hx, PE, CXR Q 23 mo Hx, PE, CXR Q 6 mo Hx, PE, CXR Q 12 mo
ACCC
8
Hx, PE, CXR, CBC, blood
chemistry Q 3 mo
chemistry Q 6 mo
chemistry Q 12 mo
Table 2Specific Recommendations for Surveillance Methods in Patients with NSCLC Following Curative-Intent
Therapy*
Source Baseline
Surveillance
First 2 Years Years 35 After Year 5
Guidelines
ASCO
5
Hx, PE Q 3 mo; CXR Q
12 mo
Hx, PE Q 6 mo; CXR
Q 12 mo
Hx, PE Q 12 mo; CXR
Q 12 mo
ACR
6
Chest CT at 3 mo
posttherapy
NCCN
7
Hx, PE, CXR Q 34 mo;
spiral CT Q 12 mo
Hx, PE, CXR Q 6 mo;
spiral CT Q 12 mo
Hx, PE, CXR Q 12 mo;
spiral CT Q 12 mo
ACCC
8
Hx, PE, CXR, CBC,
blood chemistry Q 3
mo
Hx, PE, CXR, CBC,
blood chemistry Q 6
mo
Hx, PE, CXR, CBC,
blood chemistry Q 12
mo
Institutional practice
Memorial Sloan-
Kettering
10
CXR at 2 wk
posttherapy
Hx, PE CXR Q 34 mo Hx, PE, CXR Q 6 mo Hx, PE, CXR Q 12 mo
National Kyushu Cancer
Center
11
Hx, PE, CBC, blood
chemistry, tumor
markers Q 2 mo; CXR
Q 3 mo
Hx, PE, CBC, blood
chemistry, tumor
markers Q 4 mo;
CXR Q 6 mo
University of
Washington
12
*Hx history; PE physical examination; CXR chest radiograph; Q every; ASCO American Society of Clinical Oncology;
ACR American College of Radiology; NCCN National Comprehensive Cancer Network; ACCC Association of Community Cancer
Centers.
also recommended following serial levels of tumor
markers (eg, carcinoembryonic antigen, Sialyl Lewis
X, and squamous cell carcinoma antigen) as an
indicator of cancer recurrence, but other investiga-
tors did not find this practice helpful.
12
Sputum
cytology and bronchoscopy were specifically not
incorporated into either guidelines or institutional
surveillance practices.
Follow-up of Complications Related to
the Original Mode of Curative-Intent
Therapy
Hospital admissions following pulmonary resec-
tion are common. A recent study
14
found that 19% of
patients discharged from the hospital after undergo-
ing pulmonary resection, most often for lung cancer,
were readmitted within 90 days. The reasons for
hospital readmission were not specifically described
but were most often related to pulmonary problems,
postsurgical infections, and cardiac issues.
14
This
high hospital readmission rate may reflect the nu-
merous comorbid conditions that are often found in
The most common long-term morbidities from
pulmonary resection result from the loss of lung
function and from chronic pain. The degree of
reduction in pulmonary function after surgery is
directly related to the extent of the resection per-
formed. Six months after a patient undergoes lobec-
tomy, FEV
1
is approximately 10 to 15% lower than
preoperative values, and after pneumonectomy, ap-
proximately 25 to 35% lower than preoperative
values.
15
Similarly, maximal exercise capacity stabi-
lizes at 6 months after lobectomy at a 10% reduction,
and after pneumonectomy at a 20% reduction, com-
pared with preoperative values.
15
Estimates of the
frequency of chronic postthoracotomy pain vary
widely, but pain may persist in up to 50% of patients
at 18 to 24 months after resection.
16
Landreneau et
al
17
reported that 16% of their patients required
narcotic analgesia to control pain 1 year after they
underwent thoracotomy. Small numbers of patients
will require more aggressive therapy, such as inter-
costal nerve blocks, for pain control.
18
The degree of
early postoperative pain may predict the occurrence
of chronic pain.
16
Of note, measures of quality of life
return to preoperative values by 6 months after
surgery.
19
Unusual complications related to pulmonary re-
section may occur after hospital discharge. Case
series from the 1960s
20,21
reported that persistent air
in the pleural space was noted for weeks to months
following lobectomy and pneumonectomy but usu-
ally resolved without complications. An autopsy se-
ries from the same time period confirmed residual
air in the pleural space after pneumonectomy in 27
of 37 cases, even though surgery had been per-
formed years before.
22
In very rare situations, empy-
ema may develop in these spaces.
20
Torsion of the
mediastinum developing after pneumonectomy may
lead to mainstem bronchus obstruction.
23
Complications of radiation therapy with curative
intent for lung cancer include early and late injury
predominantly to the lungs and skin, and, much less
commonly, injury to the heart, pericardium, and
esophagus. Pulmonary radiation toxicity is related to
the volume of the lung that was irradiated, the
cumulative dose, and undefined factors determining
the biological predisposition of the patient. In a large
study
24
using high-dose radiation therapy, acute tox-
icity was seen in 11% of the patients, with most
injuries relating to esophageal problems and only a
third to lung toxicity. Acute radiation pneumonitis
may be responsive to corticosteroid therapy. In
contrast, late radiation pneumonitis represents irre-
versible tissue damage, occurs in approximately 8%
of patients treated with curative intent, and may
present as early as 3 months after treatment and as
late as 24 months.
24
Even without producing overt
pneumonitis, effective radiation therapy may result
in a loss of pulmonary function. Curran et al
25
described an average decrease in FEV
1
of 10% after
irradiation therapy, which is similar to that reported
after lobectomy. However, Choi and Kanarek
26
found that patients with poor lung function before
treatment had little decrease in FEV
1
after under-
going irradiation therapy.
Most of the complications related to the chemo-
therapeutic agents used for the treatment of NSCLC
and small cell lung cancer are detected during the
course of therapy. A long-term morbidity that is of
concern in patients who have completed chemother-
apy is a mild-to-moderate peripheral neuropathy,
which results from multiple treatments with the
commonly used platin vinca alkaloid and taxane
compounds.
Recommendation 1
In lung cancer patients who have been treated
with curative-intent therapy, the follow-up for com-
plications related to the curative-intent therapy
should be managed by the appropriate specialist and
should probably last 3 to 6 months. At that point, the
patient should be reevaluated by the multidisci-
plinary tumor board for entry into an appropriate
surveillance program for detecting recurrences and/or
metachronous tumors. Level of evidence, poor; benefit,
Recurrence of the Original Lung Cancer
and Development of New Primary
Lung Cancers
In 1973, Matthews et al
27
reported the results of
autopsies performed on patients who had died within
1 month of undergoing curative-intent surgical ther-
apy for lung cancer. Of the 202 patients autopsied,
73 (35%) had either residual local disease or systemic
metastases. Certainly, our ability to image extrapul-
monary metastases is much improved since the early
1970s. However, within the last decade, occult mi-
crometastases have been found in histologically neg-
ative bone marrow and thoracic lymph nodes using
immunohistochemical staining and reverse transcrip-
tase-polymerase chain reaction techniques in lung
cancer patients who were presumed to be eligible for
curative-intent surgical resection.
2831
Given this in-
formation, it should not be surprising that lung
cancer recurs distressingly far too often following
curative-intent therapy.
Numerous studies
11,3240
have reported on recur-
rence rates and patterns in patients with NSCLC
who have been treated with curative-intent surgical
resection. In patients with stage I disease that was
confirmed at surgery, 5-year recurrence rates of up
to 39% have been reported.
33,36,37
Most of these
recurrences were distant metastases.
33,36,38
Although
most recurrences were detected within the first 4 years
following curative-intent surgery,
36,37
up to 10% of
recurrences may be discovered 5 years following
curative-intent therapy.
33,36,38,39
In patients with nodal
involvement, recurrence rates increase
34,35,40
and re-
currences probably occur earlier.
32,34,40
It has been estimated from published studies on
treatment outcomes that the approximate rate of
developing a new primary lung cancer after under-
going curative-intent therapy for a NSCLC is 1 to 2%
per patient per year.
41,42
Metachronous tumors are
usually of the same histology as the original lung
cancer.
42
A review of a regional cancer registry in
Switzerland suggests that the rate may actually be
slightly less than this estimate (about 0.5% per
patient per year).
43
Experience with long-term sur-
vivors of lung cancer has indicated that new primary
lung cancers may develop up to 20 years after the
original cancer had been treated.
44
Although John-
son
42
has suggested that the risk of developing a new
lung cancer following curative-intent therapy in-
creases with time, the available data are unclear on
whether the rate of development of metachronous
tumors increases or decreases over time. Pairolero et
al
33
noted a lower rate of development of metachro-
nous tumors 5 years after curative-intent therapy
for the original lung cancer, but the Lung Cancer
Study Group
38
found an increased rate after 5 years.
An important point is that following curative-intent
therapy for NSCLC, patients are also at increased
risk for developing other aerodigestive cancers (eg,
carcinoma of the oropharynx and esophagus).
43
Roentgenographically occult lung cancers de-
tected by sputum cytology have been reported to
have an especially high rate of metachronous tumors.
Saito et al
45
described 13 metachronous tumors
occurring in a group of 127 patients who underwent
surgical resection for roentgenographically occult
NSCLC. The cumulative rate at 5 years for meta-
chronous tumors was 11%, and the incidence per
patient-year of surveillance was 2.2%. Bechtel and
colleagues
46
reported that seven metachronous tu-
mors were identified in a group of 27 patients
following surgical resection of roentgenographically
occult NSCLC. Consistent with these findings has
been the observation that central lung cancers that
have been treated with sleeve resection also may
have a high rate of metachronous tumors. Van Schil
et al
47
found that metachronous tumors developed in
11 of 145 patients undergoing sleeve resection.
Patients who have been treated for small cell lung
cancer and have survived for 2 years also have been
reported to have an especially high rate of develop-
ing metachronous NSCLCs. In two separate obser-
vational studies,
48,49
NSCLC was diagnosed in 12 to
15% of patients who had survived for at least 2 years
after undergoing therapy for small cell lung cancer
(six cases in one group of 40 patients and six cases in
another group of 47 patients). It has been estimated
that the rate of developing NSCLC 2 years after
undergoing effective therapy for small lung cancer is
2 to 13% per patient per year.
42
Another study
confirmed that the rate of developing NSCLC fol-
lowing therapy for small cell lung cancer was signif-
icantly greater than that expected from the popula-
tion data.
50
A more recent study
51
estimated that
10% of 2-year survivors of small cell lung cancer will
eventually develop NSCLC.
Curative-Intent Therapy for a Recurrence
and/or for a New Primary Tumor
Most recurrences of lung cancer are found outside
the thorax.
11,3236,48,50
Effective treatment of isolated
metastases may be possible (see article on special
treatment issues in this supplement). Locoregional
intrathoracic recurrences are treated only infre-
quently with curative-intent surgical therapy,
36,39,52
and more often are treated with radiation thera-
py.
53,54
Regardless of the therapy, survival with
locoregional recurrence of lung cancer appears to be
poor. Accumulating data suggest that curative-intent
surgical therapy is more likely to be possible with
metachronous tumors than with locoregional recur-
rences of the primary lung cancer.
44,52,5561
How-
ever, survival rates for patients with metachronous
lung cancers following curative-intent surgical resec-
tion are generally not as good as for primary lung
cancer (Table 4). Curative-intent surgical therapy
may not be possible because patients with metachro-
nous tumors may present with advanced-stage dis-
ease or may be unable to tolerate surgical resection
due to pulmonary insufficiency.
42
The limited data
suggest that, even controlling for stage of disease,
survival following curative-intent surgical resection
of metachronous tumors may not be as favorable as
that for the original lung cancer. Despite limitations
in the approach to curative-intent therapy of meta-
chronous lung cancers, 5-year survival rates of 25 to
53% (Table 4) have been reported when surgical
resection is possible.
Issues in Surveillance for Recurrence and
New Primary Tumors
Intensity of the Surveillance Program
There may be differences in how recurrences and
metachronous tumors are identified. Pairolero et al
33
scheduled visits for their stage I NSCLC patients
every 4 months for the first 2 years and then every 4
to 6 months thereafter following curative-intent sur-
gery. A medical history, physical examination, chest
radiograph, blood tests, urine analysis, and pooled
sputum cytology were performed at each visit. Most
recurrences were detected at scheduled visits (59%),
but a substantial number of recurrences were de-
tected at unscheduled visits. Most patients with
recurrences were symptomatic (53%), and symptom
assessment was the most sensitive method for de-
tecting recurrences. The blood tests, urine analysis,
physical examination, and sputum cytology added
little to detecting recurrences. Ichinose
11
described a
similarly intensive surveillance program and also
reported that most recurrences were recognized by
symptoms. Neither CT scans nor standard blood
tests provided appreciable additional benefit in iden-
tifying recurrences.
11
In contrast, some case se-
ries
59,60,61
have reported that 68 to 100% of patients
with metachronous lung cancers were asymptomatic
and had the new primary lung cancer detected by
radiographic methods.
More recent studies have provided an expanded
view of the methods used for detecting recurrences
and/or metachronous tumors by considering the
costs involved in a surveillance program. Walsh et
al
62
retrospectively evaluated the courses of 358
patients following curative-intent surgical resection
for NSCLC. There were 135 recurrences, and most
(76%) were recognized through symptoms. Although
the asymptomatic patients had a longer survival time
following detection of the recurrence, the authors
thought that this reflected lead-time bias and was not
a true survival benefit. Similar percentages of symp-
tomatic patients (29%) and asymptomatic patients
(30%) could be treated with curative intent. Seven
metachronous lung cancers were recognized in this
study, but information on therapy and survival for
these patients was not provided. The authors con-
cluded that intensive surveillance was not cost-
effective and suggested a reduced surveillance ap-
proach consisting of a medical history, physical
examination, and chest radiograph every 6 months
for the first year following curative-intent surgery,
and annually thereafter.
Virgo and colleagues
63
compared two groups ret-
Table 4Survival after Surgical Resection for Metachronous Lung Cancers*
Study
Patients With
Metachronous Tumors
Patients Undergoing
Surgical Resection
Patients With
Stage I Disease
5-Year Survival Rate After Surgical
Resection of Metachronous
Cancer (5-Year Survival after
Surgical Resection of Primary
Lung Cancer)
Rosengart et al
44
78 54 (69) 60 (77) 23% (70%)
Watanabe et al
52
8 8 (100) 6 (75) 53%
Wu et al
55
20 20 (100) Not stated 42%
Van Bodegom et al
56
89 45 (51) 35 (39) Not stated
Deschamps et al
57
44 44 (100) 34 (77) 34% (55%)
Westermann et al
58
8 8 (100) 7 (88) Not stated
Antakli et al
59
39 21 (54) Not stated 23%
Adebonojo et al
60
37 36 (97) 29 (78) 37%
Asaph et al
61
37 37 (100) 25 (68) 33%
*Values given as No. (%), unless otherwise indicated.
Values in parentheses indicate 5-year survival rate after surgical resection of primary lung cancer.
Five-year survival comparative data following surgical resection of primary lung cancer not provided.
rospectively following surgery for NSCLC. One
group of 120 patients had intensive surveillance,
consisting of at least four visits per year, with testing
of serum chemistry and a chest radiograph, and
annual bronchoscopy and/or sputum cytology with
CT scan.
63
The other group of 62 patients underwent
nonintensive surveillance with, on average, only two
visits per year, with testing of serum chemistry and a
chest radiograph. No differences were found be-
tween the groups in either the time to detection of
recurrences or metachronous tumors or in survival
time. Virgo et al
63
agreed that intensive surveillance
was not cost-effective and supported the surveillance
schedule suggested by Walsh et al.
62
Two other
retrospective analyses of intensive surveillance meth-
ods provided similar results. Younes and colleagues
64
found that intensive surveillance yielded no survival
advantage and was more expensive than a symptom-
based approach, although more patients in the symp-
tom-based group had disease identified through
emergency department visits. Gilbert and cowork-
ers
65
showed that more recurrences were found by
family physicians based on a symptomatic presenta-
tion than were identified through regularly sched-
uled surveillance visits to the surgical clinic. These
investigators also found that the costs of identifying
recurrences would be much lower using family
physicians than using intensive surveillance through
the surgical clinic. Reviews of this topic
66,67
have
endorsed the concept of less intense surveillance
because more intensive diagnostic testing has yet to
demonstrate survival and quality of life benefits.
66
The concept of less intensive surveillance has been
challenged by the work of Westeel et al.
68
They
instituted a very intensive surveillance program in
192 patients following curative-intent surgical re-
section for NSCLC. Visits were scheduled every 3
months for 3 years, with medical history, physical
examination, and chest radiographs at each visit.
Bronchoscopy and CT scans were performed at
6-month intervals. From the fourth year postopera-
tively, visits with chest radiographs occurred at
6-month intervals, and CT scans and bronchoscopy
were performed annually. At year 8, surveillance was
reduced to a visit and a chest radiograph annually.
Westeel et al
68
claimed good compliance with this
surveillance regimen in a subset of the entire group.
There were two remarkable findings in this study.
Survival for the 36 patients with asymptomatic
recurrences was significantly better than for the 100
patients with symptomatic recurrences. A subset of
10 patients was treated with curative-intent therapy
after asymptomatic recurrences were recognized
through either bronchoscopy (5 patients) or CT
scanning (5 patients). In their economic analysis,
they suggested that this very intensive surveillance
regimen provided an acceptable cost per additional
year of life gained.
Reconciling the conflicting findings from these
various studies is difficult. The panel recognizes that
periodic patient encounters following curative-intent
therapy for lung cancer are essential and strongly
feels that imaging studies of the chest should be
included in these visits. A CT scan is accepted as
being more sensitive for detecting pulmonary nod-
ules than is a chest radiograph and has been shown
to be more accurate for evaluating lung cancer
response during chemotherapy.
69
Small series
70,71
have shown that a CT can detect changes that are
consistent with cancer recurrence earlier than can a
chest radiograph. CT scanning is also being widely
studied as a method for the early detection of lung
cancer (see article on screening for lung cancer in
this supplement). Unfortunately, the performance
characteristics of CT scanning (ie, sensitivity and
specificity) for distinguishing nonspecific posttreat-
ment changes related to surgery, radiation therapy,
and/or chemotherapy from a recurrence, and/or
metachronous lung cancer have not been defined.
Consequently, the panel was evenly divided between
recommending a chest radiograph or a CT scan as
the imaging procedure of choice.
Recommendation 2
In lung cancer patients who have been treated
with curative-intent therapy, surveillance with a
medical history, physical examination, and imaging
study (ie, either a chest radiograph or a chest CT
scan) is recommended every 6 months for 2 years
and then annually. Patients should be counseled on
symptom recognition and should be advised to con-
tact their physician if worrisome symptoms are rec-
ognized. Level of evidence, poor; benefit, moderate;
Physician Factors Influencing Current Surveillance
Methods
Numerous reports have evaluated individual fac-
tors that might influence the surveillance methods
used by thoracic surgeons. Many thoracic surgeons
perform regular surveillance for detecting recur-
rences and/or metachronous lung cancers following
curative-intent surgical therapy.
72
The most com-
monly used methods were the history, physical ex-
amination, chest radiograph, CBC count, and serum
chemistry measurement. Infrequently used surveil-
lance methods were CT scanning, bronchoscopy,
sputum cytology, bone scan, and head CT scan.
There was wide variation in the frequency with
which these methods were used. This wide variation
was probably due to the common belief that the
clinical benefits of a surveillance program, particu-
larly in terms of improving survival, had not been
demonstrated. Interestingly, the age of the surgeon,
the geographic region of the practice, and the stage
of the original lung cancer did not seem to influence
the surveillance methods used by individual thoracic
surgeons.
7375
Motivating factors for continued sur-
veillance seemed to be pleasing the patient, avoiding
malpractice litigation, and potentially improving the
patients quality of life.
76
A more important issue, not
specifically addressed in the surveys, was articulated
as follows by Shields
77
: The least desirable course of
action (in regard to care of the lung cancer patient
following curative-intent surgical therapy) is to pass
the patient from one team member to another
without continued surveillance by the primary re-
sponsible physician.
Recommendation 3
Ideally, surveillance for the recognition of a recur-
rence of the original lung cancer and/or the devel-
opment of a metachronous tumor should be coordi-
nated through a multidisciplinary team approach.
This team should develop a lifelong surveillance plan
that is appropriate for the individual circumstances
of each patient immediately following the initial
curative-intent therapy. If possible, the physician
who diagnosed the primary lung cancer and initiated
the curative-intent therapy should remain as the
health-care provider who is overseeing the surveil-
lance process. Level of evidence, poor; benefit,
Alternative Surveillance Techniques
There is considerable interest in developing non-
invasive, easily performed, safe, and accurate tech-
niques for detecting recurrences and/or metachro-
nous tumors at the earliest possible time. An
additional radiographic approach to early detection is
positron emission tomography (PET) scanning. PET
scanning appears to have improved performance
characteristics compared to CT scanning for identi-
fying malignant pulmonary nodules and mediastinal
nodal involvement in confirmed cases of lung can-
cer.
78,79
Preliminary studies with PET scanning also
have suggested that this technique may prove valu-
able in detecting recurrences.
7883
A 1999 study
84
compared PET scanning to CT scanning for detect-
ing recurrences in a group of 58 patients following
curative-intent therapy for NSCLC. Both PET scan-
ning and CT scanning were initially performed 3
months after the completion of therapy and subse-
quently at 6-month intervals. PET scanning correctly
identified 13 recurrences, but CT scanning identi-
fied only 9 recurrences. PET scanning and CT
scanning each incorrectly suggested a recurrence in
one patient. The impact of these imaging results on
patient survival and quality of life was not described.
Patz and colleagues
85
found that patients with posi-
tive findings on PET scans 8 months after undergo-
ing curative-intent therapy for NSCLC had signifi-
cantly shorter survival times than did those with
negative findings on PET scans.
Another approach to the early identification of
recurrences of lung cancer is based on measuring
serum levels of tumor markers. Ichinose
11
has rec-
ommended using serum carcinoembryonic antigen
levels as a marker of tumor recurrence. Another
marker used for detecting tumor recurrence has
been serum levels of cytokeratin-19 fragments.
86
More recently, levels of pro-gastrin-releasing pep-
tide have been suggested as a useful marker of tumor
recurrence in patients with small cell lung cancer.
87
Further studies will be needed to confirm the per-
formance characteristics of tumor markers for iden-
tifying tumor recurrence.
A pilot study
88
has been performed using fluores-
cence bronchoscopy to detect metachronous tumors
after curative-intent surgical resection of NSCLC. In
a group of 73 patients who underwent fluorescence
bronchoscopy at a median of 13 months following
surgical resection, one invasive carcinoma and three
cases of intraepithelial neoplasia were identified.
The carcinoma was identified on routine white-light
bronchoscopy, but fluorescence bronchoscopy was
useful in identifying two of the three cases of
intraepithelial neoplasia. The impact of the early
detection of intraepithelial neoplasia on survival
should be confirmed in larger studies before fluores-
cence bronchoscopy is incorporated into surveillance
programs.
Recommendation 4
In lung cancer patients following curative-intent
therapy, the use of blood tests, PET scanning,
sputum cytology, tumor markers, and fluorescence
bronchoscopy is not currently recommended for
surveillance. Level of evidence, poor; benefit, nega-
Smoking Cessation
Smoking is common in patients with lung cancer.
Gritz and colleagues
89
studied smoking behavior in
840 adults with stage I NSCLC who had participated
in clinical trials. At the time of diagnosis, 60% of the
patients were smokers. By 2 years after diagnosis,
40% of the smokers had quit smoking. Smoking
cessation at the time of diagnosis of lung cancer may
reduce the rate of development of metachronous
tumors. Richardson et al
90
found that the relative risk
of developing a second lung cancer following cura-
tive-intent therapy for small cell lung cancer was
lower for those who had stopped smoking. Tucker
and coworkers
91
found that continuing to smoke
increased the risk of metachronous lung cancers in
small cell lung cancer survivors.
Recommendation 5
Lung cancer patients who smoke should be
strongly encouraged to stop smoking. Level of evi-
dence, fair; benefit, moderate; grade of recommen-
dation, B
Summary
Following curative-intent therapy for lung cancer,
patients should be followed for 3 to 6 months by the
appropriate specialist for potential complications. In
addition to this follow-up, a recurrence of the origi-
nal lung cancer and/or the development of a second
primary lung cancer should be expected possibilities.
Most recurrences of the original lung cancer will
occur within 4 years of undergoing curative-intent
therapy, but 10% of recurrences may occur 5
years after surgery. Following curative-intent ther-
apy for lung cancer, the lifelong risk of developing a
second primary, or metachronous, lung cancer may
be 1 to 2% per patient per year. The risk of
developing a metachronous lung cancer may be even
higher when the original primary cancer was roent-
genographically occult, central, or a small cell carci-
noma.
Curative-intent therapy is less likely to be possible
with locoregional recurrences of the original lung
cancer than with metachronous tumors. Although
survival is not as good with the treatment of meta-
chronous tumors as for the original primary cancer,
reasonable 5-year survival rates should be expected
with surgical resection of metachronous lung cancers.
Benefits in terms of survival advantages or im-
provements in quality of life have not been demon-
strated with intensive surveillance programs com-
pared either with a symptom-based approach or with
a less intensive regimen. In addition, the intensive
surveillance programs seem to be more expensive. A
clinically reasonable and cost-effective surveillance
approach would include a medical history, a physical
examination, and an imaging study (either a chest
radiograph or a chest CT scan) every 6 months for 2
years and then annually. In addition, patients would
be counseled on symptom recognition and should be
advised to contact the appropriate physician on
symptom recognition. Further studies are needed to
determine whether very intensive surveillance pro-
grams might be warranted in selected subsets of lung
cancer patients (eg, patients with roentgenographi-
cally occult primary lung cancers and patients sur-
viving 2 years with small cell lung cancer and a
complete response to original therapy) who have a
very high expected rate of developing a metachro-
nous lung cancer.
Ideally, surveillance programs for the recognition
of a recurrence of the original lung cancer and/or the
development of a metachronous tumor following
curative-intent therapy should be coordinated
through a multidisciplinary team approach. This
team should develop a lifelong surveillance plan that
is appropriate for the individual circumstances of
each patient immediately following the initial cura-
tive-intent therapy. If possible, the physician who
diagnosed the primary lung cancer and initiated the
curative-intent therapy should remain as the health-
care provider, overseeing the surveillance process.
Patients either with a recurrence of their original
cancer or with a new primary lung cancer that has
been identified through the surveillance process
should be reevaluated by the entire multidisciplinary
team for potentially curative retreatment.
Although advanced imaging techniques, such as
PET scanning, appear to be more sensitive than a
chest radiograph for identifying recurrences and/or
metachronous tumors, their value in improving ei-
ther survival or quality of life following curative-
intent therapy for NSCLC is as yet unproven. Incor-
porating PET scanning into a surveillance program
should await the results of adequately designed and
controlled prospective trials. Similarly, serum levels
of various tumor markers and fluorescence bron-
choscopy should be demonstrated to be sensitive and
specific predictors of tumor recurrence in ade-
quately designed and controlled prospective trials
before being incorporated into surveillance pro-
grams.
1. In lung cancer patients who have been treated
with curative-intent therapy, the follow-up for
complications related to the curative-intent
therapy should be managed by the appropriate
specialist and should probably last 3 to 6
months. At that point, the patient should be
reevaluated by the multidisciplinary tumor
board for entry into an appropriate surveillance
program for detecting recurrences and/or
metachronous tumors. Level of evidence, poor;
2. In lung cancer patients who have been treated
with curative-intent therapy, surveillance with a
medical history, physical examination, and imag-
ing study (either chest radiograph or chest CT
scan) is recommended every 6 months for 2 years
and then annually. Patients should be counseled
on symptom recognition and should be advised to
contact their physician if worrisome symptoms
are recognized. Level of evidence, poor; benefit,
3. Ideally, surveillance for the recognition of a re-
currence of the original lung cancer and/or the
development of a metachronous tumor should be
coordinated through a multidisciplinary team ap-
proach. This team should develop a lifelong sur-
veillance plan appropriate for the individual cir-
cumstances of each patient immediately following
initial curative-intent therapy. If possible, the
physician who diagnosed the primary lung cancer
and initiated the curative-intent therapy should
remain as the health-care provider overseeing the
surveillance process. Level of evidence, poor;
4. In lung cancer patients following curative-intent
therapy, the use of blood tests, PET scanning,
sputum cytology, tumor markers, and fluores-
cence bronchoscopy is not currently recom-
mended for surveillance. Level of evidence, poor;
benefit, negative; grade of recommendation, D
5. Lung cancer patients who smoke should be
strongly encouraged to stop smoking. Level of
evidence, fair; benefit, moderate; grade of recom-
mendation, B
References
1 American Cancer Society. Cancer facts and figures 2001.
Atlanta, GA: American Cancer Society
3 Virgo KS, Johnson FE, Naunheim KS. Follow-up of patients
with thoracic malignancies. Surg Oncol Clin N Am 1999;
8:355369
4 Martini N, Melamed MR. Multiple primary lung cancers.
guidelines for the treatment of unresectable non-small-cell
lung cancer: adopted on May 16, 1997 by the American
Society of Clinical Oncology. J Clin Oncol 1997; 15:2996
3018
6 American College of Radiology. Follow-up of non-small cell
lung cancer: appropriateness criteria. Reston, VA: American
College of Radiology, 1999
7 National Comprehensive Cancer Network. Practice guide-
lines for non-small-cell lung cancer. Rockledge, PA: National
Comprehensive Cancer Network, 2000
8 Association of Community Cancer Centers. Oncology patient
management guidelines, version 3.0. Rockville, MD: Associ-
ation of Community Cancer Centers, 2000
9 National Comprehensive Cancer Network. Practice guide-
lines for small-cell lung cancer. Rockledge, PA: National
Comprehensive Cancer Network, 2000
10 Downey RJ, Martini N, Ginsberg RJ. Bronchogenic carci-
noma. In: Johnson FE, Virgon KS, eds. Cancer patient
follow-up. St. Louis, MO: Mosby, 1997; 226230
11 Ichinose Y. Counterpoint. In: Johnson FE, Virgon KS, eds.
Cancer patient follow-up. St. Louis, MO: Mosby, 1997;
230232
12 Johnkoski JA, Wood DE. Counterpoint. In: Johnson FE,
Virgon KS, eds. Cancer patient follow-up. St. Louis, MO:
Mosby, 1997; 233236
13 British Thoracic Society. BTS recommendations to respira-
tory physicians for organising the care of patients with lung
cancer. Thorax 1998; 53(suppl):S1S8
14 Handsy JR, Child AI, Grunkmeier GL, et al. Hospital read-
missions after pulmonary resection: prevalence, patterns and
predisposing characteristics. Ann Thoracic Surg 2001; 72:
18551860
15 Nezu K, Kushibe K, Tojo T, et al. Recovery and limitation of
exercise capacity after lung resection for lung cancer. Chest
1998; 113:15111516
16 Katz J, Jackson M, Kavanagh BP, et al. Acute pain after
thoracic surgery predicts long-term post-thoracotomy pain.
Clin J Pain 1996; 12:5560
17 Landreneau RJ, Mack MJ, Hazelrigg SR, et al. Prevalence of
chronic pain after pulmonary resection by thoracotomy or
video-assisted thoracic surgery. Thorac Cardiovasc Surg 1994;
107:10791086
18 Dajczman E, Gordon A, Kreisman H, et al. Long-term
postthoracotomy pain Chest 1991; 99:270274
19 Dales RE, Belanger R, Shamji FM, et al. Quality-of-life
following thoracotomy for lung cancer. J Clin Epidemiol
1994; 47:14431449
20 Barker Wl, Langston HT, Naffah P. Postresectional spaces.
Ann Thoracic Surg 1966; 2:299310
21 Silver AW, Espinas EE, Byron FX. The fate of the postresec-
tion space. Ann Thoracic Surg 1966; 2:311326
22 Suarez J, Clagett OT, Brown AL Jr. The postpneumonectomy
space: factors influencing obliteration. J Thorac Cardiovasc
Surg 1969; 57:539542
23 Grillo HC, Shepard J, Mathisen DJ, et al. Postpneumonec-
tomy syndrome: diagnosis, management, and results. Ann
Thorac Surg 1992; 54:638651
24 Cox JD, Azarnia N, Byhardt RW, et al. A randomized phase
I/II trial of hyperfractionated radiation therapy with total
doses of 60.0 Gy to 79.2 Gy. J Clin Oncol 1990; 8:15431555
25 Curran WJ, Moldofsky PJ, Solin LJ. Quantitative analysis of
the impact of radiation therapy field selection on post-RT
pulmonary function. Int J Radiat Oncol Biol Phys 1988;
15:121
26 Choi NC, Kanarek DJ. Toxicity of thoracic radiotherapy on
pulmonary function in lung cancer. Lung Cancer 1994; 10
:S219S230
27 Matthews MJ, Kanhouwa S, Pickren J, et al. Frequency of
residual and metastatic tumor in patients undergoing surgical
resection for lung cancer. Cancer Chemother Rep 1973;
4:6367
28 Salerno CT, Frizell S, Nichans GA, et al. Detection of occult
micrometastases in non-small cell lung carcinoma by reverse
transcriptase-polymerase chain reaction. Chest 1998; 113:
15261532
29 Pantel K, Izibicki J, Passlick B, et al. Frequency and prog-
nostic significance of isolated tumor cells in bone marrow of
patients with non-small-cell lung cancer without overt metas-
tases. Lancet 1996; 347:649653
30 Passlick B, Izibicki J, Kubuschok B, et al. Detection of
disseminated lung cancer cells in lymph nodes. Ann Thorac
Surg 1996; 61:177183
31 Chen Z, Perez S, Holmes EC, et al. Frequency and distribu-
tion of occult micrometastases in lymph nodes of patients
with non-small cell lung carcinoma. J Natl Cancer Inst 1993;
85:493498
32 Immerman SC, Vanecko RM, Fry WA, et al. Site of recur-
rence in patients with stages I and II carcinoma of the lung
resected for cure. Ann Thorac Surg 1981; 32:2326
33 Pairolero PC, Williams DE, Bergstralh EJ, et al. Postsurgical
stage I bronchogenic carcinoma. Ann Thorac Surg 1984;
38:331336
34 Iascone C, DeMeester TR, Albertucci M, et al. Local recur-
rence of resectable non-oat cell carcinoma of the lung.
Cancer 1986; 57:471476
35 The Ludwig Lung Cancer Study Group. Patterns of failure in
patients with resected stage I and II non-small-cell carcinoma
of the lung. Ann Surg 1986; 205:6771
36 Martini N, Bains MS, Burt ME, et al. Incidence of local
recurrence and second primary tumors in resected stage I
lung cancer. Thorac Cardiovasc Surg 1995; 109:120129
37 Harpole DH, Herndon JE, Wolfe WG, et al. A prognostic
model of recurrence and death in stage I non-small cell lung
cancer utilizing presentation, histopathology, and oncoprotein
expression. J Cancer Res 1995; 55:5056
38 Thomas P, Rubinstein L. Cancer recurrence after resection:
T1 N0 non-small cell lung cancer; Lung Cancer Study Group.
39 Thomas P Jr, Rubinstein L. Malignant disease appearing late
after operation for T1 N0 non-small-cell lung cancer: Lung
Cancer Study Group. J Thorac Cardiovasc Surg 1993; 106:
10531058
40 Baldini EH, DeCamp MM, Katz MS, et al. Patterns of
recurrence and outcome for patients with clinical stage II
non-small-cell lung cancer. Am J Clin Oncol 1999; 22:814
41 Poon RB. Lightening can strike twice. Chest 2000; 118:1526
1529
42 Johnson BE. Second lung cancers in patients after treatment
for an initial lung cancer. J Natl Cancer Inst 1998; 90:1335
1345
43 Levi F, Randimbison L, Te VC, et al. Second primary cancers
in patients with lung carcinoma. Cancer 1999; 86:186190
44 Rosengart TK, Martini N, Ghosn P, et al. Multiple primary
lung carcinomas. Ann Thorac Surg 1991; 52:773779
45 Saito Y, Sato M, Sagawa M, et al. Multicentricity in resected
occult bronchogenic squamous cell carcinoma. Ann Thorac
Surg 1994; 57:12001205
46 Bechtel JJ, Petty TL, Saccomanno G. Five year survival and
later outcome of patients with x-ray occult lung cancer
detected by sputum cytology. Lung Cancer 2000; 30:17
47 Van Schil PEY, Riviere AB, Knaapen PJ, et al. Second
primary lung cancer after bronchial sleeve resection. Thorac
48 Johnson BE, Ihde DC, Matthews MJ, et al. Non-small-cell
lung cancer. Am J Med 1986; 80:11031110
49 Heyne KH, Lippman SM, Lee JJ, et al. The incidence of
second primary tumors in long-term survivors of small-cell
50 Sagman U, Lishner M, Maki E, et al. Second primary
malignancies following diagnosis of small-cell lung cancer.
J Clin Oncol 1992; 10:15251533
51 Smythe WR, Estrera AL, Swisher SG, et al. Surgical resection
of non-small cell carcinoma after treatment for small cell
52 Watanabe Y, Shimuzu J, Oda M, et al. Second surgical
intervention for recurrent and second primary bronchogenic
carcinomas. Scand J Thorac Cardiovasc Surg 1992; 26:7378
53 Curran WJ, Herbert SH, Stafford PM, et al. Should patients
with post-resection locoregional recurrence of lung cancer
receive aggressive therapy? Int J Radiat Oncol Biol Phys
1992; 24:2530
54 Green N, Kern W. The clinical course and treatment results
of patients with postresection locally recurrent lung cancer.
Cancer 1978; 42:24782482
55 Wu S, Lin Z, Xu C, et al. Multiple primary lung cancers.
Chest 1987; 92:892896
56 Van Bodegom PC, Wagenaar SS, Corrin B, et al. Second
primary lung cancer: importance of long term follow up.
Thorax 1989; 44:788793
57 Deschamps C, Pairolero PC, Trastek VF, et al. Multiple
primary lung cancers. Thorac Cardiovasc Surg 1990; 99:769
778
58 Westermann CJJ, van Swieten HA, Riviere AB, et al. Pulmo-
nary resection after pneumonectomy in patients with bron-
chogenic carcinoma. Thorac Cardiovasc Surg 1993; 106:868
874
59 Antakli T, Schaefer RF, Rutherford JE, et al. Second primary
60 Adebonojo SA, Moritz DM, Danby CA. The results of
modern surgical therapy for multiple primary lung cancers.
Chest 1997; 112:693701
61 Asaph JW, Keppel JF, Handy JR, et al. Surgery for second
lung cancers. Chest 2000; 118:16211625
62 Walsh GL, OConnor M, Willis KM, et al. Is follow-up of lung
cancer patients after resection medically indicated and cost-
effective? Ann Thorac Surg 1995; 60:15631572
63 Virgo KS, McKirgan LW, Caputo MCA, et al. Post-treatment
management options for patients with lung cancer. Ann Surg
1995; 222:700710
64 Younes RN, Gross JI, Deheinzelin D. Follow-up in lung
cancer. Chest 1999; 115:14941499
65 Gilbert S, Reid KR, Lam MY, et al. Who should follow up
lung cancer patients after operation? Ann Thorac Surg 2000;
69:16961700
66 Virgo KS, Naunheim KS, McKirgan LW, et al. Cost of patient
follow-up after potentially curative lung cancer treatment.
67 Edelman MJ, Meyers FJ, Siegel D. The utility of follow-up
testing after curative cancer therapy. J Gen Intern Med 1997;
12:318331
68 Westeel V, Choma D, Clement F, et al. Relevance of an
intensive postoperative follow-up after surgery for non-small
69 Pujol J, Demoly P, Daures J, et al. Chest tumor response
measurement during lung cancer chemotherapy. Am Rev
Respir Dis 1992; 148:11491154
70 Gorich J, Beyer-Enke SA, Flentje M, et al. Evaluation of
recurrent bronchogenic carcinoma by computed tomography.
Clin Imaging 1990; 14:131137
71 Libshitz HI, Sheppard DG. Filling in of radiation therapy-
induced bronchiectatic change. Radiology 1999; 210:2527
72 Naunheim KS, Virgo KS, Coplin MA, et al. Clinical surveil-
lance testing after lung cancer operations. Ann Thorac Surg
1995; 60:16121616
73 Johnson FE, Naunheim KS, Coplin MA, et al. How practice
patterns in lung cancer patient follow-up are affected by
surgeon age. Oncol Rep 1996; 3:851855
74 Johnson FE, Naunheim KS, Coplin MA, et al. Geographic
variation in the conduct of patient surveillance after lung
cancer surgery. J Clin Oncol 1996; 14:29402949
75 Johnson FE, Naunheim KS, Coplin MA, et al. How tumor
stage affects surgeons surveillance strategies after lung can-
cer surgery. Chest 1997; 111:99102
76 Virgo KS, Naunheim KS, Coplin MA, et al. Lung cancer
patient follow-up. Chest 1998; 114:15191534
77 Shields TW. Postoperative lung cancer surveillance. Chest
1997; 111:1112
78 Lowe VJ, Naunheim KS. Positron emission tomography. Ann
Thorac Surg 1998; 65:18211829
79 Lowe VJ, Naunheim KS. Current role of positron emission
tomography in thoracic oncology. Thorax 1998; 53:703712
80 Kubota K, Yamada S, Ishiwata K, et al. Positron emission
tomography for treatment evaluation and recurrence detec-
tion compared with CT in long-term follow-up of cases of
lung cancer. Clin Nucl Med 1992; 17:877881
81 Patz EF, Lowe VJ, Hoffman JM, et al. Persistent or recurrent
bronchogenic carcinoma. Radiology 1994; 191:379382
82 Inoue T, Kim EE, Komaki R, et al. Detecting recurrent or
residual lung cancer with FDG-PET. J Nucl Med 1995;
36:788793
83 Duhaylongsod FG, Lowe VJ, Patz EF, et al. Detection of
primary and recurrent lung cancer by means of F-18 fluoro-
deoxyglucose positron emission tomography (FDG PET).
84 Bury T, Corhay JL, Duysinx B, et al. Value of FDG-PET in
detecting residual or recurrent nonsmall cell lung cancer. Eur
Respir J 1999; 14:13761380
85 Patz EF, Connolly J, Herndon J. Prognostic value of thoracic
FDG PET imaging after treatment for non-small cell lung
cancer. AJR Am J Roentgenol 2000; 174:769774
86 Stieber P, Zimmermann A, Reinmiedl J, et al. CYFRA 211 in
the early diagnosis of recurrent disease in nonsmall cell lung
carcinomas. Anticancer Res 1999; 19:26652668
87 Niho S, Nishiwaki Y, Goto K, et al. Significance of serum
pro-gastrin-releasing peptide as a predictor of relapse of small
88 Weigel TL, Kosco PJ, Dacic S, et al. Postoperative fluores-
cence bronchoscopic surveillance in non-small cell lung can-
cer patients. Ann Thorac Surg 2001; 71:967970
89 Gritz ER, Nisenbaum R, Elashoff RE, et al. Smoking behav-
ior following diagnosis in patients with stage I non-small cell
lung cancer. Cancer Causes Control 1991; 2:105112
90 Richardson GE, Tucker MA, Venzon DJ, et al. Smoking
cessation after successful treatment of small cell lung cancer
is associated with fewer smoking related second primary
cancers. Ann Intern Med 1993; 119:383390
91 Tucker MA, Murray N, Shaw EG, et al. Second primary
cancers related to smoking and treatment of small cell lung
2003;123;272-283 Chest
Gene L. Colice, Jeffrey Rubins and Michael Unger
Curative-Intent Therapy
Follow-up and Surveillance of the Lung Cancer Patient Following
& Services
Updated Information
S
References
S#BIBL
free at:
Citations
S#otherarticles
Reprints
2003;123;284-311 Chest
Paul A. Kvale, Michael Simoff and Udaya B. S. Prakash
Palliative Care
ISSN: 0012-3692.
Palliative Care*
Paul A. Kvale, MD, FCCP; Michael Simoff, MD, FCCP;
Udaya B. S. Prakash, MD, FCCP
The majority of patients who acquire lung cancer will have troublesome symptoms at some time
during the course of their disease. Some of the symptoms are common to many types of cancers,
while others are more often encountered with lung cancer than other primary sites. The most
common symptoms are pain, dyspnea, and cough. This document will address the management of
these symptoms, and it will also address the palliation of specific problems that are commonly
seen in lung cancer: metastases to the brain, spinal cord, and bones; hemoptysis; tracheoesoph-
ageal fistula; and obstruction of the superior vena cava. (CHEST 2003; 123:284S311S)
Key words: bone metastases; brain metastases; dyspnea; hemoptysis; interventional bronchoscopy; pain management;
pleural effusions; spinal cord metastases; superior vena cava syndrome; tracheoesophageal fistula
Abbreviations: AHCPR Agency for Health Care Policy and Research; APC argon plasma coagulation;
EBB endobronchial brachytherapy; NSAID nonsteroidal anti-inflammatory drug; NSCLC non-small cell lung
cancer; PDT photodynamic therapy; RCT randomized controlled trial; SCLC small cell lung cancer;
SVC superior vena cava; TEF tracheoesophageal fistula; WBRT whole-brain radiation therapy
S
ince the vast majority (86%) of patients with lung
cancer will die from their disease, it is intuitively
obvious that most such patients will have one or
more symptoms during the course of their disease.
These symptoms produce a clinically significant al-
teration in quality of life, andfor many of the
symptoms and the specific problems that the symp-
toms representa shortening of the quantity of life.
Symptoms that may require palliation include those
attributable to the primary lung cancer itself (dys-
pnea, hemoptysis), regional metastases within the
thorax (superior vena cava [SVC] syndrome, tracheo-
esophageal fistula [TEF], pleural effusions), or from
metastases to distant sites (brain, spinal cord, bone).
Pain is an ever-troublesome symptom for many
patients with lung cancer. Clinicians experienced in
managing patients with lung cancer must be conver-
sant with the many different ways to palliate the
symptoms that may occur with lung cancer.
This section of the evidence-based guidelines is
based on an extensive review of the medical litera-
ture. The Agency for Health Care Policy and Re-
search (AHCPR) guidelines for the management of
cancer pain was used in an abbreviated form for the
guidelines regarding management of pain in lung
cancer. Randomized controlled trials (RCTs) have
generally not been done for most aspects of palliative
care in lung cancer specifically, and meta-analyses
are not available. Three RCTs were identified that
studied surgical resection for brain metastases and
whole-brain radiation therapy (WBRT) for brain
metastases. One RCT was identified that studied the
effect of corticosteroids in bone metastases, spinal
cord compression, and brain metastases, respec-
tively. Most reports of the topics considered in this
section were case series.
Pain Control
A comprehensive document for the management
of cancer pain was developed and published in 1994
as part of a response to Public Law 101239 (the
Omnibus Reconciliation Act of 1989), under the
aegis of the AHCPR.
1
The comments in this section
are adapted from that resource, which was written by
a multidisciplinary panel of private-sector clinicians
and other experts convened by the AHCPR. Explicit,
science-based methods and expert clinical judgment
were used to develop specific statements. The scope
of that effort is beyond what can be discussed in
detail in this document, and the reader is referred to
that resource for additional information.
The causes of cancer pain include tumor progres-
sion and related pathology (eg, nerve damage), sur-
gery, and other procedures used for treatment and
diagnosis, toxic side effects of chemotherapy and
radiation, infection, and muscle aches when patients
*From the Division of Pulmonary, Critical Care, Allergy, Immu-
nology, and Sleep Disorders Medicine (Drs. Kvale and Simoff),
Henry Ford Health System, Detroit, MI; and Division of Tho-
racic Medicine (Dr. Prakash), Mayo Clinic, Rochester, MN.
Correspondence to: Paul A. Kvale, MD, FCCP, Division of
Pulmonary, Critical Care, Allergy, Immunology, and Sleep Dis-
orders Medicine, Henry Ford Health System, 2799 West Grand
Blvd, Detroit, MI 48202; e-mail: pkvale1@hfhs.org
limit their physical activity. Approximately 75% of
patients with advanced cancer have pain. Failure to
relieve pain leads to unnecessary suffering. De-
creased activity, anorexia, and sleep deprivation
caused by pain can further weaken already debili-
tated patients.
Effective management of pain from cancer can be
achieved in approximately 90% of patients. Proper
management of a patients pain involves more than
analgesia, and the program of pain control for any
one patient must be individualized. Approaches that
may augment analgesia include cognitive/behavioral
strategies, physical modalities, palliative radiation
and antineoplastic therapies, nerve blocks, and pal-
liative and ablative surgery.
Any analgesic medication program should be kept
as simple as possible, both with regard to the fre-
quency and route of administration. Oral medica-
tions are preferred, because of convenience and
cost-efficacy. If the patient cannot take medications
orally, rectal and transdermal routes should be con-
sidered because they are relatively noninvasive. IM
routes of administration should be avoided because
of the associated pain and inconvenience, and also
because of unreliable absorption.
A nonsteroidal anti-inflammatory drug (NSAID)
or acetaminophen should be used, unless there is a
contraindication to their use. If pain persists or
becomes worse, an opioid should be added and not
substituted. Using opioids and acetaminophen or
NSAIDs often provides more analgesia than can be
accomplished by either class of drug alone. Further,
the use of acetaminophen or NSAIDs may have a
dose-sparing effect for opioids, which can provide
the benefit of fewer side effects from the opioids.
When pain persists despite this approach, the dose of
opioids should be increased or a more potent agent
chosen. The World Health Organization ladder has
been shown to be an effective method to ensure the
rational titration of therapy for cancer pain (Fig 1).
2
Morphine is the most commonly used opioid for
moderate or severe pain. It is available in a wide
variety of dosage forms that include immediate and
controlled-release preparations. Morphine is rela-
tively inexpensive. Transdermal and rectal routes of
administration can be used for most patients who
cannot receive medications orally. Morphine, hydro-
codone, and oxymorphone suppositories are avail-
able. Fentanyl is the opioid most frequently used for
transdermal administration. Meperidine should not
be administered if it is anticipated that there will be
a continuous need for opioid medication. It has a
short duration of action, and its metabolite, norme-
peridine, is toxic and causes CNS stimulation with
dysphoria, agitation, and seizures.
Both the cancer patient and family members may
shun the use of opioids because of a fear of addic-
tion. Physicians must educate both the patient and
the family about pain and how it is to be managed as
part of the treatment plan. Effective pain control
begins by asking the patient about pain. An easily
administered pain rating scale should be used for
assessment of pain, both at the time of initial pre-
sentation and periodically at regular intervals during
the course of the disease. The most common pain
scales are numeric (0 to 10 pain intensity), simple
descriptive in nature (no pain, mild, moderate, se-
vere), and a visual analog scale.
Analgesic medications should be administered
around the clock with extra doses on an as-needed
basis, as this approach helps to prevent recurrence of
pain. A written pain management plan should be
given to the patient with cancer pain. Constipation is
a side effect of opioid medications. Constipation
should be anticipated, treated prophylactically, and
monitored constantly. Mild constipation can be man-
aged by an increase in fiber consumption and a mild
laxative such as milk of magnesia. Bulk-forming
laxatives such as fiber supplements should be
avoided. Unless there are contraindications, cathartic
agents should be administered on a regular schedule.
Adjuvant drugs may be used to enhance the
efficacy of opioids. Corticosteroids produce effects
that include mood elevation, relief of inflammation,
Figure 1. The World Health Organization three-step analgesic
ladder.
and reduction of cerebral or spinal cord edema when
there is intracranial metastasis or spinal cord com-
pression. Anticonvulsants such as phenytoin, car-
bamazepine, and clonazepam are used to manage
neuropathic pain. Tricyclic antidepressants are used
as an adjuvant to analgesics for the management of
neuropathic pain. They augment the effects of opi-
oids and have innate analgesic properties. Their
mood-elevating properties may be helpful as an
adjuvant to strict analgesics. Other adjunctive phar-
macologic approaches include neuroleptics such as
the major tranquilizers, hydroxyzine, bisphospho-
nates, and calcitonin for bone metastases.
There are many different nonpharmacologic
methods to manage pain, many of which are very
simple, effective, and inexpensive. A list of nonphar-
macologic methods to manage pain is outlined in
Table 1. For patients with intractable and persistent
pain despite use of all modalities that are known and
familiar to the practitioner, referral to a clinic that
specializes in the management of pain should be
considered. Pain-control specialists can help to select
additional methods that may improve the overall
palliation of pain.
Recommendations for Pain Control
1. All patients and their families must be reas-
sured that pain can be relieved safely and
effectively. Level of evidence: good; net ben-
efit: substantial; grade of recommendation: A
2. All patients should be questioned about their
pain, and the patients self-report of pain
should be the primary source of assessment.
Simple rating scales for pain should be used to
assess pain for all patients, and to document
the effectiveness of pain management at reg-
ular intervals during treatment. Level of evi-
dence: good; net benefit: moderate; grade of
recommendation: B
3. For all patients, medications that are used to
control pain should be individualized. Level of
evidence: fair; net benefit: substantial; grade
of recommendation: B
4. For all patients, medication administration
should be simple and noninvasive, whenever
possible. Level of evidence: fair; net benefit:
substantial; grade of recommendation: B
5. For all patients, mild-to-moderate pain should
be managed initially with acetaminophen or
an NSAID, assuming there are no contraindi-
cations to their use. Opioids should be admin-
istered when pain is more severe or when it
increases. Level of evidence: good; net bene-
fit: substantial; grade of recommendation: A
6. For patients with persisting pain, the dose of
opioid or its potency should be increased.
Level of evidence: fair; net benefit: substan-
tial; grade of recommendation: B
7. For patients with pain not controlled by pure
analgesic medications, adjunctive medications
such as tricyclic antidepressants, anticonvul-
sants, and neuroleptic agents will often aug-
ment the effects of pure analgesic medica-
tions. Level of evidence: fair; net benefit:
8. For patients who require medications to con-
trol cancer pain, the medications should be
administered around the clock with additional
as-needed doses. Level of evidence: fair; net
tion: B
9. For any patient, if it is anticipated that there
will be a continuous need for opioid medica-
tion, meperidine should not be used. It has a
short duration of action, and its metabolite,
normeperidine, is toxic and causes CNS stim-
ulation with dysphoria, agitation, and seizures.
Level of evidence: fair; net benefit: none;
grade of recommendation: D
10. For all patients, medications should be admin-
istered orally because of convenience and
cost-effectiveness. If medications cannot be
administered orally, rectal and transdermal
routes are preferred because they are rela-
tively noninvasive. Level of evidence: fair; net
benefit: small; grade of recommendation: C
Table 1Nonpharmacologic Methods to Control Pain
Noninvasive Methods Anesthetic Approaches Neurosurgical and Other Invasive Procedures
Cutaneous stimulation Nerve blocks Implantation of drug infusion systems
Maintenance of regular activity Peripheral neurectomy
Repositioning Dorsal rhizotomy
Avoidance of prolonged immobilization Anterolateral cordotomy
Psychosocial interventions Commissural myelotomy
Use of peer support groups Hypophysectomy
Education Neuroaugmentation (electrical stimulation of deep
brain structures)
Participation in care by pastoral members Debulking of tumor at a site of metastasis
Radiation Acupuncture
11. For all patients, medications should not be
administered IM because of pain and incon-
venience, and because IM medications are not
reliably absorbed. Level of evidence: fair; net
12. For all patients receiving opioids, constipation
is common and it should be anticipated,
treated prophylactically, and constantly mon-
itored. Level of evidence: fair; net benefit:
13. All patients should be given a written pain
management plan. Level of evidence: fair; net
tion: B
14. All patients should be encouraged to remain
active and to care for themselves whenever
possible. Prolonged immobilization should be
avoided whenever possible. Level of evidence:
fair; net benefit: moderate; grade of recom-
mendation: B
15. For patients whose pain is associated with
muscle tension and spasm, cutaneous stimu-
lation techniques such as heat and cold appli-
cations should be offered for pain relief. Level
of evidence: poor; net benefit: small; grade of
recommendation: C
16. For all patients, psychosocial methods of care
should be introduced early in the manage-
ment plan, but they should not be regarded as
a substitute for analgesia. Level of evidence:
fair; net benefit: substantial; grade of recom-
mendation: B
17. For interested patients and family, pastoral
care should be encouraged. Level of Evi-
dence: poor; net benefit: moderate; grade of
recommendation: C
18. When patients have metastases that have caused
pain, palliative radiation therapy should be of-
fered. Level of evidence: fair; net benefit: mod-
erate; grade of recommendation: B
19. For all patients with pain, referral to a special-
ized pain clinic should be considered. Level of
evidence: fair; net benefit: moderate; grade of
recommendation: B
Palliative Treatment of Bone Metastases
Metastatic lung cancer to bone is a manifestation
of stage IV disease, and it is an indication that care
for the patient will be palliative in nature. Elimina-
tion of pain is the primary goal of treatment options.
There are no randomized prospective studies that
directly compare radiation to pharmacotherapy for
the management of pain due to bony metastasis. If a
metastasis occurs in a weight-bearing bone, prophy-
lactic surgical stabilization should be considered
before a pathologic fracture occurs.
Pain caused by bone metastases has multiple
causes. Periosteal inflammation and elevation is the
most common mechanism behind the bone metas-
tasis. Lung cancer metastases to bone are predomi-
nately lytic. After controlling pain with pharmaco-
logic methods, treatment should be directed at
managing inflammation. External beam radiation
should therefore be considered as the initial non-
pharmacologic method. This technique uses energy
to diminish the local inflammatory response and
thereby eliminate the source of the pain. Other
nonpharmacologic methods to manage pain from
bone metastases include radioactive isotope infusion,
supportive measures for pain management, and di-
rect local management (surgery, nerve blocks, etc).
Ninety percent of patients with symptomatic bone
metastases obtain some pain relief with a low-dose,
brief course of palliative radiation therapy. One half
of the responding patients may have complete pain
relief.
3
For short-term improvement in bone pain,
8 Gy in a single fraction is as effective as higher
doses.
4,5
Although a single dose of radiation may be
effective, the duration of pain relief is less than with
higher fractionated doses of radiation therapy. Also,
if large fields are required, local inflammation and
edema may be a problem with a high single dose. A
high single dose is more appropriate for small ex-
tremity fields, provided internal organs are not in-
cluded, and for patients whose expected survival is
only a few months.
There is a suggestion that different lung cancer
cell types that metastasize to the bone may have
different responses to radiation therapy. For exam-
ple, 72% of metastatic adenocarcinoma has been
reported to respond as compared to 40% of patients
with metastatic squamous cell carcinoma.
6
Most
studies do not delineate between cell types but
rather separate only small cell lung cancer (SCLC)
from non-small cell lung cancer (NSCLC). The
reported response to radiation therapy for bony
metastases is 75 to 100% for complete to partial
relief of pain.
711
A prospective randomized study compared the
addition of methylprednisolone with external beam
radiation therapy to radiation therapy alone for bone
metastases. The group treated with the combination
of methylprednisolone and radiation had more rapid
and longer duration relief of pain. Stratification was
used based on the hydroxyproline/creatinine ratio of
3.6 mg/g). Patients with a higher ratio had more
improvement with the combination of external beam
radiation and methylprednisolone.
12
IV radioisotope infusion can also be used to
manage pain from bony metastases. Ethylenedia-
mine tetramethylene phosphonic acid is one such
agent. Like many of the studies of different methods
to treat bone metastases, only a small number of the
patients in the reviewed studies had pain due to
metastatic lung cancer; this constrains our interpre-
tation of the results. Pain relief was achieved in 83 to
93% of patients treated in this fashion.
13,14
Patients
with breast cancer responded better than patients
with lung cancer, but bone metastases from lung
cancer respond better than other primary sites.
15
Other radiopharmaceuticals that have been tried
for metastatic cancer to bones include strontium-89
chloride. Response rates vary considerably, with
some studies identifying little improvement
16
and
others demonstrating a 77% response rate.
17
Re-186
hydroxyethylidene diphosphonate has been studied
in patients with bony metastases, and was found not
to be effective in patients with lung cancer.
18
Signif-
icant data and support are limited for these tech-
niques except in case report format.
Adjunctive therapy with disodium pamidronate
has demonstrated good therapeutic response by it-
self, but more importantly when it is used in combi-
nation with radiotherapy for bony metastases. Re-
sponse rates of 92% were seen in a randomized study
with external beam radiation and pamidronate, vs
radiation alone (83%), pamidronate alone (85%), or
pamidronate in combination with chemotherapy
(87%).
19
An evidence-based review of the use of this
medication provided similar conclusions.
20
Calcitonin, both porcine and salmon, has been
studied in patients with bone pain. Porcine calcitonin
was used after failure of radiotherapy and analgesics,
with reduction in pain demonstrated in 63% of
patients.
21
Salmon calcitonin was studied in different
treatment schedules to identify proper dosing regi-
mens for the management of bony pain due to
metastatic NSCLC.
22
Other techniques of pain management that have
been tried all have limited evidence-based data and
are case series with mixed cancer populations. Per-
cutaneous ethanol injection into metastatic lesions
under CT guidance was associated with a reduction
in analgesic need in 74% of patients.
23
As cancers are
highly vascular, 88% of patients who were treated by
embolization of the bone tumor vasculature in a
small case series had reduced pain.
24
Pathologic fractures may occur when lung cancer
metastasizes to bones. Fracture of long bones signif-
icantly impairs functional status and quality of life.
The femur is at special risk because of its role in
weight bearing, and surgical intervention may be
needed. Other bones that may require palliative
surgical intervention include the tibia, hip (proximal
femur plus acetabulum), vertebrae, and the hu-
merus.
Prophylactic surgery is recommended for the fol-
lowing situations when long bones are involved:
persistent or increasing local pain despite the com-
pletion of radiation therapy; a solitary well-defined
lytic lesion circumferentially involving 50% of the
cortex; involvement of the proximal femur associated
with a fracture of the lesser trochanter; and diffuse
involvement of a long bone.
25
Contraindications to
surgical treatment of metastatic disease to long
bones include a survival expectancy 4 weeks, and
a poor general condition that is an obstacle to a safe
operation.
26
No randomized prospective controlled trials have
compared surgery alone, surgery plus radiation ther-
apy, or radiation therapy alone for metastatic long-
bone disease. All series that have analyzed operative
intervention have included metastatic bone disease
from multiple primary organ sites, with breast cancer
as the most common. Lung cancer usually is the
second most common primary site in reported series.
A retrospective study of 60 patients compared adju-
vant surgery plus radiation therapy (35 sites) to 29
sites that were treated with surgery alone. Univariate
analysis revealed that combined therapy (p 0.02)
and prefracture functional status (p 0.04) were the
only predictors of patients achieving a good func-
tional status after surgery. On multivariate analysis,
only postoperative radiation therapy was significantly
associated with attaining a good level of function
after surgery (p 0.02).
26
Intramedullary nailing is generally regarded as the
preferred operative approach to deal with metastatic
long bone disease. An ex vivo biomechanical analysis
of the forces required to fracture the humerus after
fixation of 50% hemicylindrical cortical central third
defects showed that intramedullary nailing was sig-
nificantly better than dynamic compression plating.
27
In general, endoprostheses and total arthroplasty are
required only for intracapsular or very proximal
lesions.
28
Operative intervention for metastatic frac-
tures of long bones provides a good functional result
in approximately 80 to 85% of patients; a good
analgesic effect is accomplished in nearly all patients.
Recommendations for Management of Bone
Metastases
20. For patients with bone metastases, external
radiation therapy is indicated to control local-
ized pain. Higher fractionated doses of exter-
nal radiation therapy provide the most pre-
dictable and longer lasting pain relief for bone
metastases. Level of evidence: fair; net bene-
fit: moderate; grade of recommendation: B
21. For most patients with pain from bone metas-
tases, a single large fraction of external radia-
tion will provide pain relief, but this technique
is best reserved for patients with survival
expectancy 3 months and for smaller ex-
tremity lesions. Level of evidence: fair; net
22. For patients with bone metastases, systemic
corticosteroids (prednisone, 20 to 40 mg/d),
when used together with external beam radi-
ation, may augment pain relief. Level of evi-
dence: fair; net benefit: small; grade of rec-
ommendation: C
23. In patients who do not respond to external
beam radiation for the relief of pain caused by
bony metastases, bisphosphonates can be ad-
ministered alone or as an adjunct to external
radiation therapy for bone metastases. Level
of evidence: fair; net benefit: moderate; grade
24. For patients with bone metastases for whom
external radiation is not effective, calcitonin
may provide pain relief. Level of evidence:
poor; net benefit: small; grade of recommen-
dation: C
25. In patients with bone metastases, a variety of
radiopharmaceuticals are available to treat
pain. They should be considered when anal-
gesics and external radiation therapy fail to
control pain. Level of evidence: poor; net
26. In patients with bone metastases, if survival is
expected for 4 weeks and general health
status is satisfactory, surgical fixation of a
symptomatic or an asymptomatic metastasis to
long and/or weight-bearing bones is indicated
to minimize the potential for a fracture. In-
tramedullary nailing is the preferred ap-
proach, especially for the femur or the hu-
merus. Level of evidence: fair; net benefit:
Palliative Treatment of Spinal Cord
Compression
Spinal cord compression by epidural tumor is an
important complication for many patients with lung
cancer, with an estimated frequency of 5% based on
autopsy data.
29
Spinal cord compression can be
classified anatomically as intramedullary, leptomen-
ingeal, and extradural. No studies focus on functional
results of treating intramedullary or leptomeningeal
compression of the spinal cord; paraplegia and death
occur rapidly in almost all such cases, and treatment
is merely supportive. Early detection of epidural
metastases with compression of the spinal cord and
prompt treatment appear to favorably affect out-
come. Epidural spinal cord compression is defined as
compression of the dural sac and its contents (spinal
cord and/or cauda equina) by an extradural tumor
mass. The minimum radiologic evidence for cord
compression is indentation of the theca at the level of
clinical features, which include any or all of the
following: pain (local or radicular), weakness, sensory
disturbance, and/or evidence of sphincter dysfunc-
tion.
There is good evidence to support the use of
high-dose dexamethasone (64 mg/d). One well-
designed RCT compared high-dose dexamethasone
to no dexamethasone in malignant spinal cord com-
pression treated with radiation therapy alone.
30
Eighty-one percent of patients in the high-dose
dexamethasone treatment arm who were ambulatory
before treatment remained ambulatory after treat-
ment, compared with 63% in the control arm. In
patients who are paretic or paraplegic before treat-
ment, there is a lesser likelihood that gait function
will be regained, but the addition of dexamethasone
appears to improve the probability of regaining the
ability to ambulate. Significant side effects occur in
11% of those who receive high-dose dexamethasone.
High-dose dexamethasone is therefore recom-
mended as an adjunct to radiation therapy in retain-
ing or restoring ambulation after treatment, but with
a relatively high incidence of serious side effects that
must be accepted. There is inconclusive evidence to
support the use of moderate-dose dexamethasone
(16 mg/d) plus radiation therapy for malignant epi-
dural spinal cord compression.
31
Methylpred-
nisolone has not been compared to dexamethasone
in head-to-head studies.
The evidence for radiation therapy with subclinical
epidural spinal cord compression is fair, and a com-
bination of high-dose steroids plus radiation should
be administered to patients who are not paretic and
ambulatory.
32
The dose prescription should be left to
the discretion of the prescribing radiation oncologist,
as no study has stratified the results by dosing
protocol. Most lesions can be managed with nonop-
erative aggressive treatment aimed at shrinking tu-
mor size and halting growth of the tumor. There is
reasonable evidence that there is a difference in
relapse rates between those who receive prophylac-
tic radiation and those who do not for patients
with asymptomatic epidural spinal cord compression,
but the optimal screening process has not been
elucidated.
31
Surgical intervention is limited to specific indica-
tions, including spinal instability, progressive neuro-
logic deterioration from bony collapse and compres-
sion, intractable pain, and failure of conservative
treatment.
32
Laminectomy alone was once the inter-
vention of choice, but it was associated with a high
rate of spinal instability and inferior ambulatory
outcomes compared with radiation therapy alone.
Vertebral body resection with stabilization has the
advantage of maintaining the structural integrity of
the spine and removing the bulk of bony disease.
There is, however, a higher complication rate and
perioperative mortality. No study has evaluated
these techniques against each other.
Radiation therapy alone should be the first line of
treatment for patients who are ambulatory. When
there is spinal instability, bony compression, or
paraplegia at the time of presentation, surgery
should be performed first. In a retrospective analysis
of 123 patients treated at a single institution between
1970 and 1996, the major wound complication rate
for patients who had radiation before surgical de-
compression and stabilization was 32%, as compared
with 12% for patients whose surgery was done first
(p 0.05). Patients treated initially with surgery had
better functional outcomes, with 75% of ambulatory
patients remaining ambulatory and continent 30 days
after treatment, compared to 50% for those whose
surgery followed radiation.
33
The type of surgical
decompression depends on the topography of the
metastasis: when the anterior or middle column are
involved, an anterior decompression should be per-
formed. When the posterior column is involved, a
posterior approach may be preferred. A combined
approach may be needed when circumferential in-
volvement is present. Reconstruction (cement or
prosthesis) is often needed.
34,35
Both surgical and
radiation therapy specialties recommend the routine
use of surgery for patients with bony compression
and for surgical salvage after progression on radia-
tion therapy.
27
Recommendations for the Palliation of Epidural
Spinal Cord Metastases
27. For patients with epidural spinal cord metas-
tases, prompt treatment favorably affects out-
come and should be given to all such patients.
28. For patients who are not paretic and ambula-
tory, a combination of high-dose steroids plus
radiation should be administered. High-dose
dexamethasone (64 mg/d) is recommended as
an adjunct to radiation therapy in retaining or
restoring ambulation after treatment, but with
a relatively high incidence of serious side
effects that must be accepted. Level of evi-
dence: fair; net benefit: substantial; grade of
recommendation: B
29. For patients with asymptomatic epidural spi-
nal cord compression, prophylactic radiation
should be prescribed. Level of evidence: fair;
net benefit: moderate; grade of recommenda-
tion: B
30. For patients with epidural spinal cord com-
pression and spinal instability, progressive
neurologic deterioration from bony collapse
and compression, intractable pain, and failure
of conservative treatment, surgical interven-
tion is indicated. Progression of neurologic
deficit while patients are receiving radiation is
also an indication for surgical stabilization.
Level of evidence: fair; net benefit: moderate;
31. When there is spinal instability, bony com-
pression, or paraplegia at the time of presen-
tation, surgery should be performed first and
should then be followed by radiation. Level
of evidence: poor; net benefit: moderate;
Palliative Treatment of Brain Metastases
Brain metastases are more common from lung
cancer than from any other primary site. Brain
metastases from NSCLC occur in approximately one
third of patients, and brain metastases occur in
40% of patients with SCLC.
36,37
If patients with
brain metastases are not treated, neurologic deteri-
oration occurs quickly.
38,39
There are four methods
currently available to treat patients with metastatic
lung cancer to the brain: (1) systemic corticosteroids,
used to ameliorate the brain edema that typically
accompanies intracranial metastases; (2) WBRT;
(3) surgical resection of the metastasis; and
(4) stereotactic radiosurgery.
Treatment with systemic glucocorticoids is known
to improve neurologic function only for a short time
(maximum 1 month).
40
Two thirds of patients will
have improvement in their neurologic signs and
symptoms with the use of steroids.
41
Dexamethasone
is the most commonly used glucocorticoid, because it
has minimal mineralocorticoid activity as compared
with other steroids. Conventional dosing with dexa-
methasone for brain tumor edema is 16 mg/d.
4244
When dexamethasone is administered in these doses
for 1 month, serious side effects are common.
45
Two consecutive, randomized, double-blind, pro-
spective controlled trials of patients with brain me-
tastases and Karnofsky scores 80 compared dexa-
methasone, 8 mg/d, or dexamethasone, 4 mg/d, to
dexamethasone, 16 mg/d.
46
Lower doses of dexa-
methasone were equally effective for improvement
in quality of life as compared with patients treated
with 16 mg/d, with significantly fewer toxic side
effects (cushingoid facies, peripheral edema,
steroid-induced myopathy) than in the group receiv-
ing 16 mg/d (p 0.03). One other study (nonran-
domized and nonblinded) also reported on the effect
of lower doses of dexamethasone; there were similar
benefits from lower doses and fewer toxic side
effects.
47
In a small pilot study, 12 patients with
intracranial metastases were initially administered 24
mg of dexamethasone IV q6h for 48 h, and then
randomized to receive either 4 mg of dexamethasone
po q6h for approximately 2 weeks during brain
irradiation or no further dexamethasone during the
radiotherapy.
48
Withholding steroids during the ra-
diotherapy did not result in pronounced deteriora-
tion of general performance status or neurologic
function at the conclusion of treatment or in reduc-
tion in overall survival. A multi-institutional prospec-
tive trial is needed to perform adequate statistical
evaluation of patients regarding the role of steroid
therapy in managing intracranial metastases. Until
such a study is done, the consensus of opinion
holds that dexamethasone, 16 mg/d, should be ad-
ministered for 4 weeks, during the time of WBRT,
and that it should then be rapidly tapered and
discontinued.
Because of the frequency with which brain metas-
tases occur in patients with SCLC, prophylactic
WBRT is routinely indicated in patients with limited
stage disease who achieve either a complete or a near
complete response in the thorax following combined
radiation and chemotherapy. When intracranial me-
tastases are known to be present with SCLC, WBRT
is again the primary method for palliating symptoms.
Patients with more than one intracranial metasta-
sis from NSCLC are generally treated with WBRT.
Median survival with this approach is 3 to 7 months,
depending on prognostic factors.
49
Currently, there are three treatment options avail-
able for patients with a known NSCLC and a solitary
intracranial metastasis: surgical resection, external
beam WBRT, and stereotactic radiosurgery
50
(see
also chapter on special treatment issues in this
guideline). Most often, some combination of these
methods of treatment is preferable. Almost all stud-
ies of patients with solitary intracranial metastases
that have compared two or more methods of treat-
ment have included patients with tumors from a
variety of primary sites, not solely lung cancer. Lung
cancer is almost always the most common primary
site in these studies; SCLC is usually an exclusion
criterion. Whereas data analyses are done on the
group as a whole, it is reasonable to apply the
conclusions to the subset of NSCLC patients with
solitary intracranial metastases.
Two randomized, prospective, controlled trials
have demonstrated a better outcome for a combina-
tion of WBRT plus surgical resection over WBRT
alone.
51,52
Surgery is appropriate for a solitary me-
tastasis in patients with good functional status and a
surgically accessible lesion. Median survival for the
patients treated with combination therapy was sig-
nificantly better in both studies as compared with
WBRT alone (a third randomized trial failed to show
a benefit from surgery, but more patients with active
systemic disease were included in this study
53
). In
one of the two studies that showed a significant
difference in median survival for the combined
approach, the differences were most pronounced for
patients with stable extracranial disease.
52
The rationale for adding WBRT to surgical resec-
tion in the setting of a solitary brain metastasis is
based on the notion that micrometastases cannot
reliably be detected with current technology. A
randomized, prospective, controlled trial that com-
pared postoperative WBRT plus surgical resection to
surgery alone demonstrated that recurrence of tu-
mor anywhere in the brain was less frequent in the
WBRT group than in the observation group (18% vs
70%, p 0.001).
54
The time to any brain recurrence
was also significantly longer in the WBRT group.
Overall survival was not different between the two
groups; thus, postoperative radiotherapy prevented
death due to neurologic causes but death due to
systemic cancer was more frequent.
There are no significant differences among various
conventional radiation therapy fractionation schemes
(20 Gy in 5 fractions, 30 Gy in 10 fractions, 40 Gy in
20 fractions). A common dose of radiation therapy
administered is 30 Gy given at 3 Gy per fraction in
10 fractions. A more protracted schedule is used for
patients who have limited or no evidence of systemic
disease or those who have undergone resection of a
single brain metastasis, since these patients have the
potential for long-term survival or even cure.
55,56
Side effects of WBRT may include measurable
deterioration of neuropsychological function.
Stereotactic radiosurgery utilizes a stereotactic
fixation system and noncoplanar convergent beams
that create a very sharp peripheral dose fall-off along
the edge of the target. Thus, the surrounding normal
tissues are spared while the radiation kills the tumor
cells; accordingly, a single large fraction of ionizing
radiation can be administered, making this method
of treatment an attractive alternative to treat lesions
whether surgically accessible or not. Stereotactic
radiosurgery is usually restricted to lesions 3 cm in
diameter.
No randomized prospective trials have compared
stereotactic radiosurgery to surgery. Many studies of
stereotactic radiosurgery for patients with intracra-
nial metastases have reported similar median survival
times to surgery as reported by others.
5761
A retro-
spective study has demonstrated equal local tumor
control rates and equal neurologic death rates be-
tween surgery and stereotactic radiosurgery.
62
A
prospective but nonrandomized study of patients
with lung cancer (both SCLC and NSCLC) demon-
strated significantly longer median survival for ste-
reotactic radiosurgery with or without WBRT over
WBRT alone (10.6 months and 9.3 months vs
5.7 months, p 0.0001).
63
A randomized study of
WBRT alone vs WBRT plus stereotactic radiosur-
gery in patients with two to four intracranial metas-
tases showed significantly improved local control
with a trend toward increased survival for WBRT
plus stereotactic radiosurgery.
64
Stereotactic radio-
surgery can be performed after brain recurrence in
patients who previously have had WBRT, surgical
excision of a metastasis, or both. Median survival in
a case series of patients with lung cancer whose brain
metastases were treated with stereotactic radiosur-
gery alone was 13.9 months, 14.5 months for stereo-
tactic radiosurgery plus WBRT, and 10 months for
patients treated with stereotactic radiosurgery for
recurrent brain metastases.
65
Recommendations for Palliative Treatment of Brain
Metastases From Lung Cancer
32. Patients with symptomatic brain metas-
tases should be treated with dexamethasone,
16 mg/d, for 4 weeks, during the course of
WBRT; dexamethasone should then be rap-
idly tapered and discontinued. Level of evi-
recommendation: B
33. Patients with multiple brain metastases from
lung cancer should be treated with WBRT.
34. For patients with intracranial metastases that
are not surgically accessible, or when two to
four intracranial metastases are present, or for
intracranial recurrence after surgery, stereo-
tactic radiosurgery, accompanied by WBRT,
can also be offered. Level of evidence: poor;
tion: C
Palliation of Dyspnea and Cough
Dyspnea is a subjective experience of difficult,
labored, and uncomfortable breathing. Dyspnea is a
common symptom of lung cancer, is most prevalent
in advanced disease, and may present as air hun-
ger. Dyspnea and cough are the most commonly
reported symptoms in lung cancer, with 15% of
patients having dyspnea at diagnosis and 65% at
some point during their illness.
66,67
One major can-
cer center in the United States reported that dyspnea
was the fourth most common symptom among pa-
tients with cancer who presented to the emergency
department.
68
A prospective cohort study of seri-
ously ill, hospitalized adults in five teaching hospitals
in the United States reported that among 939 pa-
tients with stage III or IV NSCLC, severe dyspnea
was recorded in 32%.
69
Patients with lung cancer
presenting to emergency departments with dyspnea
have a much shorter survival than patients with other
malignancies.
68
One study of 120 patients with stage
I through IV lung cancer used a questionnaire to
evaluate the importance of dyspnea and observed
that 87% had dyspnea, and patients with high dys-
pnea scores had lower quality of life.
70
The causes of dyspnea in patients with lung cancer
can be classified into five broad groups: (1) the result
of direct involvement of the respiratory system by
lung cancer, (2) the result of indirect respiratory
complications caused by lung cancer (postobstruc-
tive pneumonia, pleural effusion, etc), (3) the result
of specific therapies to treat lung cancer (radiation-
induced and chemotherapy-induced lung toxicity,
anemia, etc, (4) the result of respiratory complica-
tions that occur more frequently in these patients
(pulmonary embolism, lung infections, etc), and
(5) comorbid conditions (COPD, heart failure, prior
lung resection, malnutrition, etc).
Irrespective of the stage of lung cancer, dyspnea
usually impacts the patients physical, social, and
psychological well-being. Anxiety, fear of impending
death, and pain caused by lung cancer are among the
factors that contribute to the subjective symptoms of
dyspnea. A prospective study of 100 terminally ill
cancer patients (49 patients with lung cancer) ob-
served that dyspnea, measured on visual analog
scale, was significantly associated with anxiety
(p 0.001).
71
From the perspectives of the patient
and health-care providers, dyspnea can be perceived
as panic, chest congestion and tightness, and suffo-
cation. One study of 52 patients with lung cancer
noted that both physical and emotional sensations
were associated with descriptions of breathlessness,
such as the feeling of being unable to get enough
breath, or of panic or impending death.
70,72
In-
creased anxiety has been connected with worse
dyspnea in patients with obstructive lung disease,
chronic pulmonary disease, and/or cancer.
7375
One
study of 120 patients with stage I through IV lung
cancer observed no difference in dyspnea based on
cancer stage, cell type, or performance status. How-
ever, pain and anxiety scores were higher in patients
with high dyspnea scores.
70
Fatigue is a common symptom among patients
with lung cancer, particularly those with advanced
disease. One study of 227 cancer patients and 98
control subjects reported that the prevalence of
severe fatigue was 15% among patients with recently
diagnosed breast cancer, 16% among patients with
recently diagnosed prostate cancer, 50% among pa-
tients with inoperable NSCLC, and 78% among
patients receiving specialist inpatient palliative care.
Fatigue was significantly associated with the severity
of psychological symptoms (anxiety and depression)
and with the severity of pain and dyspnea.
76
Pharmacotherapy of Dyspnea
Pharmacologic treatments for dyspnea caused by
lung cancer have included oxygen, bronchodilators,
corticosteroids, antibiotics, and opioids. One retro-
spective study at a medical center specializing in
cancer assessed the resource utilization associated
with the management of dyspnea caused by lung
cancer in 45 patients. The most common therapies
administered in the emergency department were
oxygen (31%),
2
-agonists (14%), antibiotics (12%),
and opioids (11%).
68
Bronchodilators
Standard bronchodilators such as
2
-agonists, an-
ticholinergics, and aerosolized corticosteroids are
commonly prescribed to patients with lung cancer
who have underlying COPD or asthma. There is no
evidence that the presence of lung cancer induces
bronchospastic disease. However, the onset of lung
cancer in patients with underlying obstructive lung
diseases usually aggravates symptoms of preexisting
obstructive lung disease. There are not many studies
to prove a beneficial effect of bronchodilators in
patients with lung cancer. However, a prospective
study of 100 terminally ill cancer patients (49 pa-
tients with lung cancer) observed that the potentially
correctable causes of dyspnea included bronchos-
pasm (in 52%) and hypoxia (in 40%).
71
It is impor-
tant to ensure that bronchodilator therapy is opti-
mized if the patient has obstructive airways disease.
Corticosteroids
The role for systemic corticosteroids is limited for
relieving dyspnea from lung cancer. As is the case
with bronchodilator therapy, patients with obstruc-
tive airways disease may benefit from treatment with
systemic corticosteroids to decrease mucus produc-
tion and inflammatory changes in the airway mucosa.
It is also important to recognize that patients with
lung cancer who are actively receiving specific ther-
apy, such as radiotherapy and/or chemotherapy, may
have varying degrees of dyspnea.
77
This may reflect
pulmonary toxicity to such therapies. Pulmonary
parenchymal toxicity leading to dyspnea may require
discontinuation of tumor-specific therapies and ad-
ministration of systemic corticosteroids.
Oxygen
Supplemental oxygen is perhaps the most com-
monly prescribed therapy to relieve dyspnea in
68
Significant involvement
of the respiratory system by lung cancer or underly-
ing obstructive airways disease usually produces or
aggravates dyspnea and hypoxemia. A limited num-
ber of studies have shown the beneficial effects of
supplemental oxygen therapy. A prospective, dou-
ble-blind, crossover trial assessed the effects of
supplemental oxygen on the intensity of dyspnea in
14 patients with advanced cancer. Patients were
randomized to receive either oxygen or air delivered
at 5 L/min by mask. Dyspnea was evaluated with a
visual analog scale. The results showed that 12
patients consistently preferred oxygen to air, and
patients reported little or no benefit from air com-
pared with moderate to much benefit from oxygen.
78
Irrespective of the oxygenation status, supplemen-
tal oxygen therapy should be considered if patients
with lung cancer have dyspnea. Multiple blood gas
analyses should be avoided to justify oxygen therapy.
Percutaneous oximetry should suffice to assess ade-
quate oxygenation.
Analgesics
Dyspnea has been shown to be more severe in
patients with severe pain.
70,73
Dyspneic sensation
caused or aggravated by cancer-induced pain may
respond to nonnarcotic analgesic therapy. However,
dyspnea due to pain caused by bony metastases,
malignant pleural effusions, or fatigue is unlikely to
respond to conventional analgesic therapy. Such
circumstances require more aggressive pain control,
including palliative radiotherapy for skeletal metas-
tasis. In patients with dyspnea caused by milder pain
and discomfort, nonnarcotic analgesics should be
tried for a brief period.
Opioid Analgesics
Opioids are frequently used to alleviate dyspnea in
patients with advanced lung cancer, advanced ob-
structive airway disease, and cardiac failure.
79
A wide
variety of opioid analgesics have been used to control
both dyspnea and pain in patients with cancer of lung
and other organs. They include morphine, oxy-
codone, hydromorphone, and others. Opioids have
been used orally, parenterally, and by aerosol. It is
unclear if all opioids are equally efficacious in de-
creasing dyspnea perception in patients with lung
cancer. In a study of 104 patients with lung cancer,
opioids administered to treat pain did not decrease
dyspnea.
70
An open, uncontrolled study evaluated the role of
oral morphine to relieve dyspnea in 15 patients with
advanced malignancy receiving standard care and
noted that regular, titrated oral morphine may im-
prove dyspnea but can cause significant short-term
adverse effects.
79
The relief of dyspnea is usually
noted within 24 h and the relief stays at a plateau
with continued opioid therapy.
80
Continuous IV infusion of morphine has been
used in patients with terminal lung cancer with
severe dyspnea, unrelieved by oxygen, nonnarcotic
drugs, or intermittent bolus narcotics.
81
Even when
patients achieve good dyspnea relief, the major side
effect is sedation. Health-care providers, patients,
and family members should be cognizant of the
possibility of severe hypoventilation and hypercarbic
respiratory failure and death. This side effect also has
been described with inhaled morphine.
82
Other Methods
Nonpharmacologic, noninterventional methods
for the control of dyspnea include patient education
and intervention by allied health personnel. A mul-
ticenter RCT of 119 patients with SCLC or NSCLC
or with mesothelioma, who had completed first-line
treatment and reported dyspnea, used various strat-
egies. These included breathing control, activity
pacing, relaxation techniques, and psychosocial sup-
port, in addition to standard management and treat-
ment available for dyspnea. The group assigned to
intervention by nurses improved significantly at 8
weeks in breathlessness, performance status, and
physical and emotional status compared to the con-
trol group.
83,84
Pharmacotherapy of Cough
Cough is a frequent and distressing symptom in
patients with lung cancer. Cough can be dry or
associated with sputum production. Involvement of
any part of the respiratory system can lead to cough.
Among the initial symptoms of lung cancer, cough is
present in 65% and productive cough in 25% of
patients.
85
Cough can be the presenting or leading
symptom of lung cancer. It is more likely among
patients with lung cancer originating in the airways.
All pharmacologic therapies aimed at controlling
cough caused by lung cancer are symptomatic. Even
if complete cessation of cough is not possible, a
significant control of cough may help patients enjoy
cough-free periods. In late-stage cancer when no
specific therapy can address the cancer itself, control
of bothersome cough becomes a problem. The phar-
macologic agents available include the following.
Cough Suppressants
Nonopioid cough suppressants may work in a
small group of patients with advanced lung cancer.
Occasionally, even opioid-resistant cough may re-
spond to agents such as the peripherally acting
nonopioid drug benzonatate.
86
Bronchodilators
Bronchospasm can cause or contribute to cough. If
the patient with lung cancer also has underlying
bronchospastic obstructive airways disease, then
standard bronchodilator therapy may help alleviate
the cough.
One study tested the role of inhaled sodium
cromoglycate in 20 patients with NSCLC and cough
resistant to conventional treatment. The patients
were randomized to receive, in a double-blind trial,
either inhaled sodium cromoglycate or placebo. The
results showed that inhaled sodium cromoglycate
reduced cough in all patients with NSCLC.
87
Opioids
Opioids are the best cough suppressants in pa-
tients with lung cancer. Codeine is the most widely
used opioid. In advanced stages of lung cancer,
standard nonopioid cough suppressants may not
control the cough. Intractable or troublesome cough
should be treated with opioid agents. Caution should
be exercised in prescribing graduated doses of these
drugs because of the risk of respiratory suppression
and hypoventilation.
A double-blind, RCT regarding the treatment of
nonproductive cough was performed in 140 adults
with primary lung cancer or metastatic cancer of the
lungs. The therapeutic efficacy and the tolerability
of a 7-day treatment with levodropropizine drops
(75 mg tid) were evaluated in comparison with
dihydrocodeine drops (10 mg tid). Efficacy was
assessed on the basis of cough severity scores, the
number of night awakenings due to cough, and
overall estimate of antitussive efficacy. Tolerability
was evaluated by laboratory results, vital signs, and
any adverse event occurring during the clinical trial,
including the presence or absence of somnolence.
Subjective cough severity was significantly reduced
during treatment with levodropropizine and dihy-
drocodeine, the antitussive effect and its time profile
being similar for both drugs. Also, according to the
investigators evaluation, both levodropropizine and
dihydrocodeine treatment produced a significant
decrease in cough severity. Concurrently with the
relief of cough, the number of night awakenings was
decreased significantly by both drugs, with no differ-
ence between the two treatments. No change in
laboratory test values was considered clinically rele-
vant, and vital signs were not clinically affected. The
number of patients reporting adverse events was
similar in the levodropropizine group (n 6) and
dihydrocodeine group (n 4). However, the per-
centage of patients experiencing somnolence in the
group receiving levodropropizine (8%) was signifi-
cantly lower as compared with that of the dihydro-
codeine group (22%). These results confirm the
antitussive effectiveness of levodropropizine and
suggest a more favorable benefit/risk profile when
compared to dihydrocodeine.
88
However, levodro-
propizine is not available for use in the United
States.
Corticosteroids
There are no studies on steroids specifically for
cough in lung cancer. If cough is caused by radiation-
induced lung problems, then high-dose corticoste-
roid therapy may relieve a significant degree of
cough.
Lidocaine
There are no studies on the role of inhaled
lidocaine on cough in patients with lung cancer.
Palliation of Dyspnea Caused by Pleural
Effusions
Malignant pleural effusions occur in 7 to 15% of
patients with lung cancer,
8991
more than half of
whom acquire dyspnea.
92
The mechanism of dys-
pnea with pleural effusions is unclear. Mechanical
factors influencing the chest wall, mediastinum,
pleural space, and the lung itself all may contribute
to the sensation of dyspnea in the patient with a
pleural effusion. Several studies have evaluated var-
ious aspects of compliance, lung volume, muscle
strength, and gas exchange in patients before and
after therapeutic aspiration of pleural effusions.
9398
When a pleural effusion is identified and sus-
pected as the etiology of dyspnea in a patient with
lung cancer, the first requisite is to determine if the
effusion is malignant or due to some other cause.
The methods for this differentiation have been dis-
cussed in another section (see chapter on diagnosis
in this guideline).
The major indication for treating a malignant
pleural effusion is to relieve dyspnea. Often there are
multiple causes of dyspnea in patients with lung
cancer, so removal of the pleural fluid may or may
not provide adequate relief of dyspnea. Chest radio-
graphs, often including decubitus views, should be
assessed to determine if the pleural fluid is free
flowing or loculated. Contralateral shift of the medi-
astinum with large effusions suggests that evacuation
of the effusion should provide relief of dyspnea for
the patient. The next step is to perform a therapeutic
thoracentesis to assess the effects on breathlessness
after fluid removal, as well as the rate and degree of
reaccumulation of pleural fluid. If the lung is trapped
because of parenchymal or pleural disease, there will
be minimal relief of dyspnea and the lung will
not re-expand. The volume of fluid removed with
the initial thoracentesis should be no more than 1 to
1.5 L, stopping earlier should the patient have
dyspnea, chest pain, or cough. Removal of larger
amounts of pleural fluid may be associated with
re-expansion pulmonary edema, particularly if there
is coexisting endobronchial obstruction.
99
Although
complicated, this technique may minimize the risk of
re-expansion pulmonary edema and help assess for
the presence of a trapped lung at the time of the
diagnostic or therapeutic thoracentesis.
100,101
Pleural
pressure monitoring may be a more objective
assessment for trapped lung than chest radiograph
assessment.
If the initial thoracentesis provides relief of dys-
pnea and lung re-expansion is seen on postprocedure
chest radiography, reaccumulation of fluid can be
managed in two basic ways: intermittent therapeutic
thoracentesis, or insertion of a chest tube to com-
pletely evacuate the pleural fluid, followed by pleu-
rodesis.
99
Repeated therapeutic thoracentesis is a viable
option for patients with poor performance status or
with advanced disease. There are no studies that
compare repeated thoracentesis to other manage-
ment approaches. If the malignant pleural effusion
continues to accumulate, a more definitive proce-
dure can be considered. Chemical pleurodesis via
chest tube or medical thoracoscopy is the most
common and effective approach, but pleuroperito-
neal shunting, pleural drainage catheters, and sys-
temic therapy are other options.
The overall complete response rate to chemical
pleurodesis is 64%.
102
Further analysis according to
the type of agent used reveals that fibrosing agents as
a group are associated with a 75% complete response
rate and that talc, specifically, is associated with a
91% complete response rate. Antineoplastic agents
are less often successful, with a reported complete
response of 44%.
102
Reported complete response
rates to various sclerosing agents are listed in
Table 2.
Despite testing a variety of new chemical pleu-
rodesis and intrapleural therapy agents, none appear
to have the clinical success rate of talc, nor have most
of these agents been tested in a randomized fashion
against talc or one of the more commonly used
chemical pleurodesis agents. All chemical pleurode-
sis and intrapleural therapy agents that are available
have common as well as their own unique complica-
tions and constraints for use, which should be re-
viewed before a specific agent is chosen.
Pleuroperitoneal shunting is another technique to
manage malignant and other intractable pleural effu-
sions. All studies of pleuroperitoneal shunting are case
series. Many patients with malignant pleural effusions
lack the ability to actively utilize the pumping device,
which must be pushed at least 100 times or so daily to
overcome the positive peritoneal pressure.
126130
Another technique for managing malignant pleural
effusions is tunneled long-termcatheter drainage of the
pleural space. Case series suggest good results for the
relief of dyspnea over an extended time in patients with
malignant effusions. Although encouraging, many of
these studies are retrospective and there has been no
comparison to other treatment modalities.
131133
The treatment of choice for malignant effusions
due to SCLC is systemic chemotherapy. Many pa-
tients will respond with resolution of pleural effu-
sions and the associated dyspnea.
134
Bronchoscopic Methods to Palliate
Dyspnea and Cough
Introduction
Central airway obstruction refers to significant
obstruction of the trachea and main bronchi. Ob-
struction of lobar bronchi in a patient with limited
pulmonary reserve can also lead to considerable
respiratory distress. Central airway obstruction in
patients with lung cancer can present with life-
threatening dyspnea or hemoptysis. Most patients
with dyspnea caused by central airway obstruction
also complain of cough. The severity of dyspnea is
dependent on the extent of luminal involvement of
the airway, and the presence or absence of underly-
ing conditions such as COPD, cardiac failure, loss of
lung tissue from previous lung surgery, etc.
Extraluminal tumor compression of the major
airways, intraluminal tumor growth, or a combina-
tion of both can cause central airway obstruction.
Perhaps the most important aspect of managing
these patients is to first determine the anatomic type
of airway involvement. Clinical examination, imaging
techniques, and flow-volume curves may provide
diagnostic possibilities. However, the single most
important diagnostic procedure is bronchoscopy.
Bronchoscopy provides visual clues to the nature of
the obstructing lesion and the extent of luminal
narrowing, and it will determine if an intraluminal or
extraluminal process causes the obstruction. It will
also help in determining if bronchoscopic therapy is
feasible.
Almost all endobronchial therapies are palliative in
patients with lung cancer involving the major air-
ways. A small number of patients with in situ lung
cancer, who cannot undergo resection because of
comorbid conditions, may be cured with endobron-
chial therapies. The relief of cough is more trouble-
some because none of the therapies discussed below
will totally eradicate the tumor. Thus, even though
Table 2Comparison of Complete Response Rates Among Various Agents Used for Pleurodesis
Pleurodesis agent Complete Response Rate, % Source
Talc poudrage 91 113, 116, 118, 120122
Talc slurry 91 113, 116, 118, 119, 121, 122
Doxycycline 8085 110112
79 110
Bleomycin* 3679 (compared with
tetracycline)
109, 110, 113117
Batimastat (BB-94) 4450 123
Iodopovidone 64 124
Mitoxantrone 79 125
Mepacrine 86 125
Mitomycin C 41 126
OK-432 73 126
Intrapleural chemotherapy (cisplatin plus cytarabine) 49 127
Doxorubicin 40 128
Doxorubicin with LC9018 74 128
Doxorubicin microspheres 100 129
-interferon 22 130
Interleukin 2 45 131, 132
*Results displayed in studies where bleomycin was compared to tetracycline.
Tetracycline no longer available.
OK-432 associated with 80% complication rate; therefore, this agent is no longer used.
cough and hemoptysis may be controlled to some
extent by these methods, other types of therapy may
be required to control cough.
The degree of dyspnea and respiratory distress
should dictate the appropriate mode of endobron-
chial therapy. The currently available methods of
bronchoscopic therapy include debulking of intralu-
minal tumor growth (usually with rigid bronchosco-
py), balloon dilatation, laser therapy, electrocautery,
cryotherapy, argon plasma coagulation (APC), endo-
bronchial irradiation, or intraluminal stent place-
ment. All of the following therapeutic techniques
will provide significant relief of dyspnea and cough in
the majority of patients. It should be noted, however,
that not all techniques described here will accom-
plish dyspnea relief quickly (Table 3).
Types of Therapy
Endotracheal Intubation: This should be per-
formed only in a patient who faces impending death
because of tracheal obstruction and no broncho-
scopic therapy is available. If the obstruction is
located in a short segment of trachea and if bron-
choscopic examination reveals a lumen, an endotra-
cheal tube (largest diameter feasible) should be
inserted so the tip of the tube is beyond the distal
aspect of obstruction. Once the airway is secured and
oxygenation optimized, then consideration should be
given to more definitive therapy. This technique is
useful in both luminal and extraluminal obstructions.
The risk of bleeding during endotracheal intubation
and the difficulty of intubation should be recognized.
Rigid Bronchoscopy: Rigid bronchoscopy is the
quickest technique to relieve dyspnea caused by
either intraluminal or extraluminal obstruction of
major airways. The major advantages include the
availability of rigid bronchoscopes of various diame-
ters, ability to maintain an airway, the possibility of
delivering oxygen and anesthetic gases, and the
ability to employ other therapeutic techniques de-
scribed below. One retrospective study has evaluated
the role of emergency rigid bronchoscopic interven-
tion, including Nd-YAG laser resection or stenting,
in patients with acute respiratory failure from malig-
nant central airways obstruction. Airway strictures
were caused by primary bronchogenic lung cancer in
14 patients. Urgent bronchoscopic intervention per-
mitted immediate discontinuation of mechanical
ventilation in 52% of these patients (including 19
patients with benign lesions).
136
The rigid bronchoscope itself can be used as a
tumor-debulking instrument, much like coring an
apple. In many patients, this alone may suffice to
relieve dyspnea. In patients with extraluminal airway
obstruction, the rigid bronchoscope is essential to
insert silicone stents. A rigid bronchoscope is also
helpful in placement of self-expanding metal
stents.
137
Endobronchial Balloon Dilatation: This procedure
has a limited role in the treatment of major airway
obstruction by malignant tumors.
138
Balloon dilata-
tion, through either the flexible or rigid broncho-
scope, is best suited for stenoses that are short in
length.
139
Bronchoscopic balloon dilatation may help
prepare an obstructed airway for placement of
stents.
Laser: Endobronchial laser therapy is useful in
relieving obstruction caused by intraluminal lesions.
It has no role in treatment of obstruction caused by
extraluminal tumors. Either rigid or flexible bron-
choscopy can be used for application of laser energy,
even though the former accomplishes this more
quickly. Currently, various types of lasers are avail-
able for treatment of endobronchial tumors. These
include Nd-YAG, potassium triphosphate, and CO
2
Table 3Bronchoscopic Therapies*
Therapies Type of Lesion Type of Bronchoscope
Rapidity of
Positive Result
Repeatability of
Therapy
Mechanical debridement E/Sm R/F1
Stent E/W/Ex R/F2
Laser E R/F1
Electrocautery E R/F
APC E R/F
Cryotherapy E R/F
Balloon dilatation E/Sm/Ex R/F1
Brachytherapy E/Sm F
PDT E F
*From Ramser and Beamis.
135
R rigid bronchoscope; F flexible bronchoscope; E endoluminal lesion; Sm submucosal lesion;
Ex extraluminal compression; 1 rigid bronchoscope preferable; 2 Dumon stent requires rigid bronchoscope; Wallstents and Gianturco
stents require fluoroscopy; most rapid or repeatable.
laser units. Almost all bronchoscopic laser therapies
reported in the literature have employed Nd-YAG
laser for therapy of major airway lesions. Immediate
relief of airway occlusion and obstructive symptoms
can be expected in 90% of patients. Laser therapy
also helps in preparing the airway for insertion of
airway stents. Complications from laser therapy in-
clude severe hemorrhage, pneumothorax, and pneu-
momediastinum.
140,141
Disadvantages of laser appli-
cation are that it requires special training and
expensive equipment.
Electrocautery: Electrocautery application
through either a rigid or flexible bronchoscope em-
ploys alternating electrical current to produce coag-
ulation and vaporization of endobronchial lesions.
142
The rapidity with which the electrocautery tech-
nique opens obstructed airway is similar to that
achieved with laser therapy. Immediate relief of
dyspnea can be achieved with electrocautery in 55 to
75% of patients.
143147
A prospective study evaluated
the impact of bronchoscopic electrosurgery on the
need for bronchoscopic Nd-YAG laser in patients
with symptomatic airway lesions and observed that of
the 47 bronchoscopic electrosurgery procedures, 42
procedures (89%) were successful in alleviating the
obstruction, thus eliminating the need for laser. All
procedures were performed in the outpatient bron-
choscopy suite with the patient under conscious
sedation (morphine and midazolam) and topical
anesthesia with 2% lidocaine.
148
The advantages of
electrocautery include the less expensive equipment
(compared to laser) and the ease of use through
flexible or rigid bronchoscope. Complications in-
clude endobronchial fire, hemorrhage, and inadver-
tent electrical shock to the operator or patient.
Overall, electrocautery seems a good alternative to
the more expensive laser therapy.
APC: APC applies a technique to achieve noncon-
tact electrocoagulation of viable tissue. APC utilizes
electrically conductive argon plasma as a medium to
deliver high-frequency current via a flexible probe to
coagulate tissue. APC devitalizes tissue gradually by
producing temperatures that coagulate and desiccate
tissue. One retrospective study of 60 patients with
bronchogenic carcinoma,
43
metastatic tumors of air-
ways,
14
or benign bronchial disease
3
employed APC
therapy via flexible bronchoscopy to control hemop-
tysis, symptomatic airway obstruction, or both ob-
struction and hemoptysis. Patients with endoluminal
airway lesions had an overall decrease in mean
obstruction of 18 22%. All patients with obstruc-
tive lesions experienced symptom improvement, and
symptom control was maintained during a median
follow-up period of 53 days. There were no compli-
cations related to APC. The advantages of APC
include low cost, noncontact mode of therapy, easy
portability of equipment, and ease of use. The
noncontact feature of APC allows rapid coagulation
with minimal manipulation of and mechanical
trauma to the target tissue. The procedure can be
performed in an outpatient setting or at the bedside
in the ICU.
149
Cryotherapy: Cryotherapy employs cryoprobes,
through either a rigid or flexible bronchoscope, to apply
extremely cold temperatures to tumor tissue so that
malignant cells are devitalized and killed by repeated
cycles of cold application followed by thawing. Nitrous
oxide or liquid nitrogen is most commonly used to
produce temperatures of 80C.
150152
As is the case
with laser and electrocautery, cryotherapy can be used
to treat only intraluminal tumors. Subjective improve-
ments have been observed in 75% of patients with
malignant airway lesions.
153,154
The cryotherapy equip-
ment is less expensive and easier to use than laser
therapy. Complications are few and minor. Repeat
bronchoscopy is needed for continued therapy in many
patients. The major disadvantage of treating large
tumors in major airways is that cryotherapy requires far
more time to relieve obstruction. As a result, cryother-
apy is not an ideal therapeutic method to relieve
dyspnea caused by major airway lesions.
Stents: Airway prostheses or stents made of metal,
silicone, or other materials are available to relieve
airway obstruction caused by malignant tumor.
155,156
Stent therapy is applicable to both intraluminal and
extraluminal major airway obstructions. Stent ther-
apy is more effective in patients with tracheal or
main bronchial diseases than in those with airway
diseases that involve lobar and distal bronchi. Either
silicone or metallic stents can be used to treat
malignant airway lesions. Uncovered metallic stents
should not be inserted in patients with malignant
airway lesions because the growth of cancer through
the wire mesh negates the benefits of stent place-
ment.
137
Even after bronchoscopic debridement of
tumor and laser therapy, stent placement should be
considered to maintain airway patency. In one study
of 22 patients with severe malignant strictures, 34
endobronchial stents were implanted as a temporary
measure when patients received irradiation or che-
motherapy. Significant improvements of dyspnea
and partial oxygen pressure were observed, and in
50% of patients, the stents were removed after
successful tumor-specific therapy.
157
In another
study, among 34 patients with inoperable malignant
airway stenosis, covered metallic stents were im-
planted on emergent basis in 19 patients (56%)
because of life-threatening airway obstruction. Im-
mediate relief of dyspnea was achieved in 82% of the
patients, and significant improvements were ob-
served in airway diameter, vital capacity, and peak
expiratory flow.
136
All silicone stents require rigid bronchoscopy for
their insertion, manipulation, and removal,
158160
whereas metal stents can be inserted with the aid of
flexible bronchoscopy and/or fluoroscopic guidance.
Complications from silicone stents include migration
of stent and inspissation of thick mucus within the
stent lumen. Metallic stents are more likely to
promote growth of granulation tissue.
Brachytherapy: Brachytherapy is employed to de-
liver radiation therapy from within the airway lumen
to treat intraluminal malignant tumors.
161163
Brachytherapy is aimed at palliating malignant air-
way lesions in patients who have already received a
maximum dose of external beam radiation. Brachy-
therapy can be used as a stand-alone therapy or as
complimentary or combined therapy following exter-
nal beam radiation therapy, airway debulking (laser,
mechanical removal, etc), or after airway stent place-
ment. Even though earlier experience demonstrated
that endobronchial brachytherapy (EBB) alone re-
sulted in adequate symptomatic relief in a consider-
able number of patients,
164171
current evidence
indicates that brachytherapy as a complimentary
therapy provides better relief of dyspnea and other
symptoms than EBB alone.
168,172178
Relief from
dyspnea can be expected in 60% of patients and
can last for weeks to months. A phase II study
involving 30 patients with stage III NSCLC treated
with 60 Gy radiation therapy also used EBB and
reported palliation rates of 80% for dyspnea and 43%
for cough.
179
One prospective study of 342 patients
with endobronchial tumors treated by the combina-
tion of external beam radiation therapy (30 to 60 Gy)
and concomitant EBB during weeks 1, 3, and 5
observed a response rate of 85% for cough and 86%
for dyspnea.
178
Major complications of brachyther-
apy include fistula formation between the airways
and other thoracic structures in up to 8% of patients.
The risk of massive hemoptysis increases dramati-
cally when a fraction size of 15 Gy is used.
180
Photodynamic Therapy: Photodynamic therapy
(PDT) consists of using tumor-tagging compounds
such as hematoporphyrin derivative and Photofrin.
When tumor cells thus tagged are exposed to the
light of the proper wavelength, chemical reactions
cause death of malignant cells through the produc-
tion of toxic radicals. Patients with small ( 3 cm
2
)
epithelial cell malignancies are most likely to benefit
from this therapy. Complete response lasting for
12 months has been observed in 50% of pa-
tients.
181,182
Complications from PDT include pho-
totoxicity, hemoptysis, and obstruction of bronchi by
thick necrotic material.
The effectiveness of PDT for symptom palliation
and determined survival benefit has been evaluated
in patients with advanced inoperable bronchogenic
cancer and endobronchial luminal obstruction.
Among 100 such patients, 82% had received prior
chemotherapy and/or radiotherapy. On an average,
endoluminal obstruction diminished from 86 to 18%.
This study suggests that PDT is effective in palliation
of inoperable advanced lung cancer in a subset of
patients. However, the relief from obstruction is
slow,
183
and because of this slow response there is no
major role for PDT in the treatment of obstructing
lesions of major airways.
Surgery: Surgical resection of malignant tracheo-
bronchial tumors should be considered when un-
usual types of malignant tumors are encountered.
The types of tumors that are amenable to resection
and anastomosis include carcinoid, cylindroma, and
mucoepidermoid tumors. The length of involvement
of trachea or major bronchus should be short enough
for the surgeon to resect the tumor so that the
anastomotic site is free of malignant cells.
Recommendations for Palliation of Cough and
Dyspnea
35. In all patients with lung cancer, potentially
correctable causes of dyspnea, such as local-
ized obstruction of a major airway, a large
pleural effusion, or an exacerbation of coexist-
ing COPD, should be sought initially. Level of
evidence: poor; net benefit: moderate; grade
of recommendation: C
36. For all lung cancer patients with dyspnea,
pharmacologic approaches for the manage-
ment of dyspnea may include oxygen, bron-
chodilators, corticosteroids, antibiotics, and
opioids. Level of evidence: poor; net benefit:
moderate; grade of recommendation: C
nonpharmacologic, noninterventional treat-
ments including patient education and inter-
vention by allied health personnel should be
used to help control dyspnea, including
breathing control, activity pacing, relaxation
techniques, fans, and psychosocial support.
Level of evidence: poor; net benefit: moder-
ate; grade of recommendation: C
38. For all patients with lung cancer who continue
to have cough, opioids are the best cough
suppressants and should be used. Level of
recommendation: B
39. Patients with malignant pleural effusions that
cause dyspnea initially should be drained by
thoracentesis. Level of evidence: fair; net ben-
efit: substantial; grade of recommendation: C
40. Patients with lung cancer who have poor
performance status and limited life expect-
ancy, and with recurring malignant pleural
effusions, can be managed with repeated tho-
racenteses. Level of evidence: fair; net bene-
fit: small; grade of recommendation: C
41. Patients with NSCLC and better performance
status and recurrent malignant pleural effu-
sions, and whose lungs re-expand with initial
thoracentesis or thoracoscopy, should be fol-
lowed up by pleurodesis. Level of evidence:
good; net benefit: moderate; grade of recom-
mendation: B
42. In patients with SCLC, the treatment of choice
for malignant effusions is systemic chemother-
apy. Level of evidence: good; net benefit: mod-
43. For patients with central airway obstruction,
bronchoscopy should be done to determine
the type of airway obstruction (extraluminal
tumor compression of the major airways, in-
traluminal tumor growth, or both). Level of
44. In patients with central airway obstruction,
rapid relief of dyspnea can be accomplished
via bronchoscopy with removal of intraluminal
tumor (laser, electrocautery, APC) and/or by
inserting a stent. Other methods (cryotherapy,
brachytherapy, PDT) are effective but do not
relieve dyspnea as quickly. Level of evidence:
poor; net benefit: substantial; grade of recom-
mendation: C
Palliation of Hemoptysis
Hemoptysis is the presenting symptom in 7 to 10%
of patients with lung cancer. Approximately 20% will
have hemoptysis some time during their clinical
course, with 3% having terminal massive hemopty-
sis.
184187
For patients whose initial presenting symp-
tom is hemoptysis, surgical resection of the bleeding
lobe or the entire lung may be appropriate if the
cancer is confined to a hemithorax and amenable to
surgery with curative intent. Massive hemoptysis,
that which most commonly requires intervention,
has a broad definition as expectoration of at least
100 to 600 mL of blood in 24 h. Blood clot formation
obstructing the airways is suggested as the most
common cause of respiratory insufficiency from mas-
sive hemoptysis. Massive hemoptysis due to lung
cancer has a much poorer prognosis than hemoptysis
of other etiologies. The mortality of massive hemop-
tysis may be as high as 59 to 100% in patients with
bronchogenic carcinoma.
188
Surgery, a more defini-
tive therapeutic modality, is not on the algorithm for
intervention, as most lung cancer patients with mas-
sive hemoptysis have advanced disease and are al-
ready nonsurgical candidates.
Because the prognosis for survival is so grim
among lung cancer patients who acquire massive
hemoptysis, the patient may not want any kind of
intervention. If intervention is undertaken, the initial
priority in managing massive hemoptysis should be
to maintain an adequate airway.
189,190
This usually
requires endotracheal intubation, and a single-lumen
endotracheal tube is generally more beneficial than a
double-lumen endotracheal tube. Selective right or
left mainstem intubation can be performed to pro-
tect the nonbleeding lung. Double-lumen endotra-
cheal tubes are more difficult to place and position,
have smaller lumens, and do not allow a therapeutic
bronchoscope to be passed through each side of the
tube. This makes it difficult to further control and/or
suction the airways.
191
Bronchoscopy is typically needed to identify the
source of bleeding. Endoscopic management of he-
moptysis depends on whether the location of bleed-
ing is identified but a specific endoscopic cause for
the bleeding is not seen, as opposed to both site and
a specific cause of the hemoptysis (eg, bleeding from
an endobronchial tumor). When only the location of
bleeding is identified but no direct source is found,
endobronchial management options begin with tam-
ponade of the segment by continuous suctioning to
collapse the segment.
192
Vasoactive drugs such as
1:10,000 epinephrine solution can be instilled, al-
though this is most useful on visualized lesions.
193
Iced saline solution lavage has been proposed as a
temporizing technique to control hemoptysis,
194,195
and balloons may be used to tamponade the bleeding
site. It is suggested that balloons remain in place for
24 to 48 h to allow tamponade of hemoptysis.
196199
Use of Nd-YAG photocoagulation is an efficient tool
for the management of bleeding endobronchial le-
sions, with a reported response rate of 60%.
200,201
Electrocautery should produce similar results, but its
use to control hemoptysis has thus far been anec-
dotal. APC provided resolution of hemoptysis in
100% of patients with a 3-month follow-up.
149
If hemoptysis is determined to arise from an
endoscopically visualized lung cancer, bronchial ar-
tery embolization may temporize the bleeding. Most
reports of bronchial artery embolization are limited
by the few cases of lung cancer managed in almost all
studies.
202205
When bronchoscopy reveals that bleeding is com-
ing from an endoscopically visible, unresectable lung
cancer, external beam radiation should be the next
consideration.
206
This is true even when other tem-
porizing measures, such as those discussed earlier,
have been done. A high-dose initial fraction may
provide a more rapid response. Endobronchial radi-
ation has also been reported to be effective for the
control of bleeding from visible tumors.
Recommendations for the Palliation of Hemoptysis
45. In managing a patient with massive hemopty-
sis, the initial priority should be maintaining
adequate airway protection. If intubation is
required, a standard single-lumen endotra-
cheal tube should be used. Level of evidence:
poor; net benefit: moderate; grade of recom-
mendation: C
46. For patients with massive hemoptysis, bron-
choscopy is typically needed to identify the
source of bleeding. Early bronchoscopy to
assess the site of bleeding is recommended.
Level of evidence: poor; net benefit: substan-
tial; grade of recommendation: C
47. For patients with massive hemoptysis, endo-
bronchial management options begin with
tamponade. Effective adjunctive devices in-
clude APC, Nd-YAG laser, and electrocautery.
Level of evidence: fair; net benefit: small;
48. For patients with massive hemoptysis due to
lung cancer, bronchial artery embolization is a
temporizing treatment. Level of evidence:
dation: I
49. For patients with massive hemoptysis or per-
sistent large-volume hemoptysis that is deter-
mined to arise from an endoscopically visible,
unresectable lung cancer, external beam radi-
ation should be considered. Level of evidence:
mendation: B
Palliation of Malignant TEF
TEFs are serious complications of lung and esoph-
ageal cancer; TEFs are more common with esopha-
geal cancers than primary lung cancers. Life expect-
ancy with no therapy is estimated at 1 to 7 weeks.
Patients have repeated aspiration of food, gastric
contents, and saliva. This persistent aspiration leads
to patient distress due to coughing and shortness of
breath. Patients can acquire recurrent pneumonias
with persistent inflammation of the airways. Patients
frequently lose weight and become dehydrated be-
cause they cannot tolerate oral intake. Even with
abstinence from eating and drinking, most patients
have difficulty with controlling their own secretions.
Curative resection of the involved tracheal-
bronchial and/or esophageal segments in face of a
malignancy is inappropriate; most such patients are
near the end of their lives and palliation should be
the primary treatment objective. Esophageal bypass
procedures can be considered, but they have very
high morbidity and mortality and are inappropriate
as palliative tools in advanced lung cancer. The goals
of therapy are to restore patency of the trachea,
bronchi, and/or esophagus, to prevent spillage of
further material into the lung, and ensure that the
patient receives nutrition and fluid.
Double stenting of the tracheobronchial tree and
the esophagus appears to be the procedure that
yields the best overall results for symptomatic relief
for patients with this condition. Clinical series have
attempted either esophageal or tracheobronchial
stenting with mixed results. Most series with higher
success rates use a double-stenting technique.
207209
The addition of percutaneous enterogastric tube
placement can ensure proper nutrition and fluid
management for patients with this condition. Pa-
tients may be able to eat soft foods once double
stenting is performed, but maintaining adequacy of
fluid status and nutrition is often difficult.
Recommendations for the Palliative Treatment of
Malignant TEF
50. For patients with a malignant TEF or bron-
choesophageal fistula, stenting of both the
tracheobronchial tree and the esophagus is the
procedure that yields the best overall results
for symptomatic relief. Level of evidence:
mendation: C
51. For any patient with lung cancer and a TEF,
attempts at curative resection of the involved
trachea and/or bronchi and/or esophageal seg-
ments should not be done. Level of evidence:
fair; net benefit: none; grade of recommenda-
tion: D
52. For any patient with advanced lung cancer,
esophageal bypass procedures have very high
morbidity and mortality and should not be
done. Level of evidence: fair; net benefit:
Palliative Treatment of SVC Obstruction
The SVC may become obstructed by lung cancer,
either via direct extension or more commonly by
regional lymph node metastases. SVC syndrome
develops in 10% of patients with a right-sided ma-
lignant intrathoracic mass lesion.
210
SVC syndrome
includes symptoms that may be severe and debilitat-
ing, including congestion of the collateral veins of
the neck, anterior chest wall, face, eyelids and right
arm. Dyspnea, headache from cerebral venous hy-
pertension, and cyanosis occur less frequently. Typ-
ically, lifestyle is significantly impaired with SVC
syndrome. Consensus opinion once held that emer-
gency radiation therapy was always necessary to
alleviate the symptoms that are secondary to com-
pression of the SVC. However, the need for emer-
gency radiation therapy has been challenged, and
many symptomatic patients are better treated by
stenting.
211
Intravascular stenting with expandable
metallic stents has largely replaced surgical venous
bypass to obtain rapid relief of SVC syndrome from
malignant obstruction. Balloon angioplasty may be
needed initially to enlarge the vascular lumen and
allow proper stent placement. When thrombosis
occurs as a complication of SVC syndrome, local lytic
therapy may be of value to re-establish patency and
subsequently to allow insertion of a stent.
There is usually time to obtain a histologic diag-
nosis, which is important because patients with
SCLC are well managed by chemotherapy.
211
Whereas chemotherapy for SCLC may also improve
the symptoms associated with SVC syndrome, radi-
ation therapy is usually a part of treating SCLC with
SVC obstruction. Patients with SCLC and SVC
syndrome undergoing concurrent treatment with
chemotherapy and radiation therapy have improved
5-year survival (15 7%) [mean SD] compared
with SCLC patients without SVC syndrome
(9 2%, p 0.008).
212
Radiation therapy for SVC syndrome caused by
NSCLC is associated with a good symptomatic
response in 82% of patients within 1.7 0.9 weeks
(3 days to 4 weeks).
213
In a small series, 10 of 11
patients had complete relief of symptoms within
3 days after stent implantation; they remained symp-
tom free until death in follow-up (17 to 227 days).
214
In a nonrandomized study, expandable metallic
stents provided symptom relief (78%) equally often
as radiation therapy (80%); mean survival was iden-
tical.
215
Patients with intraluminal tumors had a
shorter mean survival (45 days) compared with those
who had extrinsic compression (199 days). Stent
placement was effective in relieving symptoms in
patients without a response to radiation therapy.
215
In another nonrandomized study comparing stent
insertion to radiation therapy, faster relief of symp-
toms and significantly greater improvement in SVC
obstruction score were seen with stent insertion
(p 0.0005 and 0.001, respectively).
216
Placement
of stents to obtain quick relief of SVC syndrome
symptoms does not preclude the use of radiation
therapy or chemotherapy. Complications attribut-
able to stent insertion (bleeding due to vascular
injury, thrombosis within the stent) may occur in a
minority of patients so treated.
Recommendations for Palliation of SVC
Obstruction
53. Lung cancer patients with symptomatic SVC
obstruction can be treated with radiation ther-
apy, insertion of a stent, or both. Level of
recommendation: B
The majority of patients who acquire lung cancer
will have troublesome symptoms during the course
of their disease. Pain, dyspnea, hemoptysis, and the
effects of regional or distant metastases to bones,
brain, or spinal cord are common. There are many
effective methods available to relieve these symp-
toms. Familiarity with the palliative approach to care
is crucial for a clinician to be competent in caring for
Recommendations for Pain Control
1. All patients and their families must be reas-
sured that pain can be relieved safely and
effectively. Level of evidence: good; net ben-
efit: substantial; grade of recommendation: A
2. All patients should be questioned about their
pain, and the patients self-report of pain
should be the primary source of assessment.
Simple rating scales for pain should be used to
assess pain for all patients, and to document
the effectiveness of pain management at reg-
ular intervals during treatment. Level of evi-
recommendation: B
3. For all patients, medications that are used to
control pain should be individualized. Level of
4. For all patients, medication administration
should be simple and noninvasive, whenever
possible. Level of evidence: fair; net benefit:
5. For all patients, mild-to-moderate pain should
be managed initially with acetaminophen or an
NSAID, assuming there are no contraindica-
tions to their use. Opioids should be adminis-
tered when pain is more severe or when it
increases. Level of evidence: good; net benefit:
substantial; grade of recommendation: A
6. For patients whose pain persists, the dose of
opioid or its potency should be increased.
7. For patients whose pain is not controlled by
pure analgesic medications, adjunctive medi-
cations such as tricyclic antidepressants, anti-
convulsants, and neuroleptic agents will often
augment the effects of pure analgesic medica-
tions. Level of evidence: fair; net benefit:
8. For all patients who require medications to
control cancer pain, the medications should
be administered around the clock with addi-
tional as-needed doses. Level of evidence:
mendation: B
9. For any patient, if it is anticipated that there
will be a continuous need for opioid medica-
tion, meperidine should not be administered.
It has a short duration of action, and its
metabolite, normeperidine, is toxic and causes
CNS stimulation with dysphoria, agitation,
and seizures. Level of evidence: fair; net
10. For all patients, medications should be admin-
istered orally because of convenience and
cost-effectiveness. If medications cannot be
taken orally, rectal and transdermal routes are
preferred because they are relatively noninva-
sive. Level of evidence: fair; net benefit: small;
11. For all patients, medications should not be
administered IM because of pain and incon-
venience, and because IM medications are not
reliably absorbed. Level of evidence: fair; net
12. For all patients receiving opioids, constipation
is common and it should be anticipated,
treated prophylactically, and constantly mon-
itored. Level of evidence: fair; net benefit:
13. All patients should be given a written pain
management plan. Level of evidence: fair; net
tion: B
14. All patients should be encouraged to remain
active and to care for themselves whenever
possible. Prolonged immobilization should be
avoided whenever possible. Level of evidence:
mendation: B
15. For patients whose pain is associated with
muscle tension and spasm, cutaneous stimu-
lation techniques, such as heat and cold appli-
cations, should be offered for pain relief.
Level of evidence: poor; net benefit: small;
16. For all patients, psychosocial methods of care
should be introduced early in the manage-
ment plan, but they should not be regarded as
a substitute for analgesia. Level of evidence:
mendation: B
17. For interested patients and family, pastoral
care should be encouraged. Level of evidence:
mendation: C
18. When patients have metastases that have caused
pain, palliative radiation therapy should be of-
fered. Level of evidence: fair; net benefit: mod-
19. For all patients with pain, referral to a special-
ized pain clinic should be considered. Level of
recommendation: B
Recommendations for Management of Bone
Metastases
20. For patients with bone metastases, external
radiation therapy is indicated to control local-
ized pain. Higher fractionated doses of exter-
nal radiation therapy provide the most pre-
dictable and longer-lasting pain relief for bone
metastases. Level of evidence: fair; net bene-
fit: moderate; grade of recommendation: B
21. For most patients with pain from bone metas-
tases, a single large fraction of external radia-
tion will provide pain relief, but this technique
is best reserved for patients with survival
expectancy 3 months and for smaller ex-
tremity lesions. Level of evidence: fair; net
22. For patients with bone metastases, systemic
corticosteroids (prednisone, 20 to 40 mg/d),
when used together with external beam radi-
ation, may augment pain relief. Level of evi-
dence: fair; net benefit: small; grade of rec-
ommendation: C
23. In patients who do not respond to external
beam radiation for the relief of pain caused by
bony metastases, bisphosphonates can be ad-
ministered alone or as an adjunct to external
radiation therapy for bone metastases. Level
of evidence: fair; net benefit: moderate; grade
24. For patients with bone metastases for whom
external radiation is not effective, calcitonin
may provide pain relief. Level of evidence:
dation: C
25. In patients with bone metastases, a variety of
radiopharmaceuticals are available to treat
pain. They should be considered when anal-
gesics and external radiation therapy fail to
control pain. Level of evidence: poor; net
26. In patients with bone metastases, if survival is
expected for 4 weeks and general health
status is satisfactory, surgical fixation of a
symptomatic or an asymptomatic metastasis to
long and/or weight-bearing bones is indicated
to minimize the potential for a fracture. In-
tramedullary nailing is the preferred ap-
proach, especially for the femur or the hu-
merus. Level of evidence: fair; net benefit:
Recommendations for the Palliation of Epidural
Spinal Cord Metastases
27. For patients with epidural spinal cord metas-
tases, prompt treatment favorably affects out-
come and should be administered to all such
patients. Level of evidence: fair; net benefit:
28. For patients who are not paretic and ambula-
tory, a combination of high-dose steroids plus
radiation should be administered. High-dose
dexamethasone, 64 mg/d, is recommended as
an adjunct to radiation therapy in retaining or
restoring ambulation after treatment, but with
a relatively high incidence of serious side
effects that must be accepted. Level of evi-
dence: fair; net benefit: substantial; grade of
recommendation: B
29. For patients with asymptomatic epidural spi-
nal cord compression, prophylactic radiation
should be prescribed. Level of evidence: fair;
tion: B
30. For patients with epidural spinal cord com-
pression and spinal instability, progressive
neurologic deterioration from bony collapse
and compression, intractable pain, and failure
of conservative treatment, surgical interven-
tion is indicated. Progression of neurologic
deficit while patients are receiving radiation is
also an indication for surgical stabilization.
31. When there is spinal instability, bony com-
pression, or paraplegia at the time of presen-
tation, surgery should be performed first and
should then be followed by radiation. Level of
evidence: poor; net benefit: moderate; grade
Recommendations for Palliative Treatment of Brain
Metastases From Lung Cancer
32. Patients with symptomatic brain metas-
tases should be treated with dexamethasone,
16 mg/d, for 4 weeks during the course of
WBRT; dexamethasone should then be rap-
idly tapered and discontinued. Level of evi-
recommendation: B
33. Patients with multiple brain metastases from
lung cancer should be treated with WBRT.
34. For patients with intracranial metastases that
are not surgically accessible, or when two to
four intracranial metastases are present, or for
intracranial recurrence after surgery, stereo-
tactic radiosurgery, accompanied by WBRT,
can also be offered. Level of evidence: poor;
tion: C
Recommendations for Palliation of Cough and
Dyspnea
35. In all patients with lung cancer, potentially
correctable causes of dyspnea, such as local-
ized obstruction of a major airway, a large
pleural effusion, or an exacerbation of coexist-
ing COPD, should be sought initially. Level
of evidence: poor; net benefit: moderate;
pharmacologic approaches for the manage-
ment of dyspnea may include oxygen, bron-
chodilators, corticosteroids, antibiotics, and
opioids. Level of evidence: poor; net benefit:
moderate; grade of recommendation: C
nonpharmacologic, noninterventional treat-
ments, including patient education and inter-
vention by allied health personnel, should
be used to help control dyspnea, including
breathing control, activity pacing, relaxation
techniques, fans, and psychosocial support.
Level of evidence: poor; net benefit: moder-
ate; grade of recommendation: C
38. For all patients with lung cancer who continue
to have cough, opioids are the best cough
suppressants and should be used. Level of
recommendation: B
39. Patients with malignant pleural effusions that
cause dyspnea initially should be drained by
thoracentesis. Level of evidence: fair; net ben-
efit: substantial; grade of recommendation: C
40. Patients with lung cancer who have poor
performance status and limited life expect-
ancy, and with recurring malignant pleural
effusions, can be managed with repeated tho-
racenteses. Level of evidence: fair; net bene-
fit: small; grade of recommendation: C
41. NSCLC patients with better performance sta-
tus and recurrent malignant pleural effusions,
and whose lungs re-expand with initial thora-
centesis or thoracoscopy, should be followed
up with pleurodesis. Level of evidence: good;
tion: B
42. In patients with SCLC, the treatment of choice
for malignant effusions is systemic chemother-
apy. Level of evidence: good; net benefit: mod-
43. For patients with central airway obstruction,
bronchoscopy should be done to determine
the type of airway obstruction (extraluminal
tumor compression of the major airways, in-
traluminal tumor growth, or both). Level of
44. In patients with central airway obstruction,
rapid relief of dyspnea can be accomplished
via bronchoscopy with removal of intraluminal
tumor (laser, electrocautery, APC) and/or by
inserting a stent. Other methods (cryotherapy,
brachytherapy, PDT) are effective but do not
relieve dyspnea as quickly. Level of evidence:
poor; net benefit: substantial; grade of recom-
mendation: C
Recommendations for the Palliation of Hemoptysis
45. In managing a patient with massive hemopty-
sis, the initial priority should be maintaining
adequate airway protection. If intubation is
required, a standard single-lumen endotra-
cheal tube should be used. Level of evidence:
mendation: C
46. For patients with massive hemoptysis, bron-
choscopy is typically needed to identify the
source of bleeding. Early bronchoscopy to
assess the site of bleeding is recommended.
Level of evidence: poor; net benefit: substan-
47. For patients with massive hemoptysis, endo-
bronchial management options begin with
tamponade. Effective adjunctive devices in-
clude APC, Nd-YAG laser, and electrocautery.
Level of evidence: fair; net benefit: small;
48. For patients with massive hemoptysis due to
lung cancer, bronchial artery embolization is a
temporizing treatment. Level of evidence:
dation: I
49. For patients with massive hemoptysis or per-
sistent large-volume hemoptysis that is deter-
mined to arise from an endoscopically visible,
unresectable lung cancer, external beam radi-
ation should be considered. Level of evidence:
mendation: B
Recommendations for the Palliative Treatment of
Malignant TEF
50. For patients with a malignant TEF or bron-
choesophageal fistula, stenting of both the
tracheobronchial tree and the esophagus is the
procedure that yields the best overall results
for symptomatic relief. Level of evidence:
mendation: C
51. For any patient with lung cancer and a TEF,
attempts at curative resection of the involved
trachea and/or bronchi and/or esophageal seg-
ments should not be done. Level of evidence:
fair; net benefit: none; grade of recommenda-
tion: D
52. For any patient with advanced lung cancer,
esophageal bypass procedures have very high
morbidity and mortality and should not be
done. Level of evidence: fair; net benefit:
Recommendations for Palliation of SVC
Obstruction
53. Lung cancer patients with symptomatic SVC
obstruction can be treated with radiation ther-
apy, insertion of a stent, or both. Level of
recommendation: B
References
1 Jacox A, Carr DB, Payne R, et al. Management of cancer
pain: clinical practice guideline No. 9. Rockville, MD:
Agency for Health Care Policy and Research, U.S. Depart-
ment of Health and Human Services, Public Health Service,
March 1994; AHCPR Publication No. 940592
2 World Health Organization. Cancer pain relief and palliative
care: Report of a WHO expert committee. World Health
Organization Technical Report Series, 804. Geneva, Swit-
zerland: World Health Organization, 1990; 175
3 Hoegler D. Radiotherapy for palliation of symptoms in
incurable cancer. Curr Probl Cancer 1997; 21:129183
4 Medical Research Council Lung Cancer Working Party.
Inoperable non-small-cell lung cancer (NSCLC): a Medical
Research Council randomized trial of palliative radiotherapy
with two fractions or ten fractions. Br J Cancer 1991;
63:265270
5 Medical Research Council Lung Cancer Working Party. A
Medical Research Council (MRC) randomised trial of pal-
liative radiotherapy with two fractions or a single fraction in
patients with inoperable non-small-cell lung cancer
(NSCLC) and poor performance status. Br J Cancer 1992;
65:934941
6 Murai N, Koga K, Nagamachi S, et al. Radiotherapy in bone
metastases, with special reference to its effect on relieving
pain. Gan No Rinsho 1989; 35:11491152
7 Hirano T, Baba H, Iwasaki K, et al. Curative local control of
malignant tumors in the acetabulum by intraoperative radio-
therapy. Arch Orthop Trauma Surg 1994; 113:215217
8 Salazar OM, Sandhu T, da Motta NW, et al. Fractionated
half-body irradiation (HBI) for the rapid palliation of wide-
spread, symptomatic, metastatic bone disease: a randomized
Phase III trial of the International Atomic Energy Agency
(IAEA). Int J Radiat Oncol Biol Phys 2001; 50:765775
9 Koswig S, Buchali A, Bohmer D, et al. Palliative radiother-
apy of bone metastases: a retrospective analysis of 176
patients. Strahlenther Onkol 1999; 175:509514
10 Koswig S, Budach V. Remineralization and pain relief in
bone metastases after different radiotherapy fractions (10
times 3 Gy vs 1 time 8 Gy): a prospective study. Strahlenther
Onkol 1999; 175:500508
11 Hayashi S, Hoshi H, Iida T, et al. Multi-fractionated wide-
field radiation therapy for palliation of multiple symptomatic
bone metastases from solid tumors. Radiat Med 1999;
17:411416
12 Teshima T, Inouye T, Ikeda H, et al. Symptomatic relief for
patients with osseous metastasis treated with radiation and
methylprednisolone: a prospective randomized study. Radiat
Med 1996; 14:185188
13 Deng H, Luo S, Tan T, et al. 153Sm-EDTMP for moderate
and severe bone cancer pain. Hua Xi Yi Ke Da Xue Xue Bao
Med Sci 1995; 26:391394
14 Jiang CY, Shu BL, Zhang YJ. The value of Sm-153-EDTMP
for treatment of metastatic bone pain and improving quality
of life. Zhonghua Zhong Liu Za Zhi 1994; 16:118121
15 Tian JH, Zhang JM, Hou QT, et al. Multicentre trial on the
efficacy and toxicity of single-dose samarium-153-ethylene
diamine tetramethylene phosphonate as a palliative treat-
ment for painful skeletal metastases in China. Eur J Nucl
Med 1999; 26:27
16 Lee CK, Aeppli DM, Unger J, et al. Strontium-89 chloride
(Metastron) for palliative treatment of bony metastases: the
University of Minnesota experience. Am J Clin Oncol 1996;
19:102107
17 Firusian N. Endo-osseous isotope therapy of malignant
skeletal disease. Cancer Res Clin Oncol 1978; 91:143156
18 Kucuk NO, Ibis E, Aras G, et al. Palliative analgesic effect of
Re-186 HEDP in various cancer patients with bone metas-
tases. Ann Nucl Med 2000; 14:239245
19 Shucai Z, Guimei L, Fanbin H. A clinical trial of Bonin in
bone metastases of lung cancer. Chin J Clin Oncol 1999;
26:445447
20 Bloomfield DJ. Should bisphosphonates be part of the
standard therapy of patients with multiple myeloma or bone
metastases from other cancers? An evidence-based review.
J Clin Oncol 1998; 16:12181225
21 Egawa J, Kawada Y, Abe M, et al. Effect of porcine
calcitonin on pain caused by cancer-induced bone destruc-
tion. Gan No Rinsho 1984; 30:251258
22 Schiraldi GF, Soresi E, Locicero S, et al. Salmon calcitonin
in cancer pain: comparison between two different treatment
schedules. Int J Clin Pharmacol Ther Toxicol 1987; 25:229
232
23 Gangi A, Kastler B, Klinkert A, et al. Injection of alcohol into
bone metastases under CT guidance. J Comput Assist
Tomogr 1994; 18:932935
24 Gorich J, Solymosi L, Hasan I, et al. Embolization of bone
metastases. Radiologe 1995; 35:5559
25 Haentjens P, Casteleyn PP, Opdecam P. Evaluation of
impending fractures and indications for prophylactic fixation
of metastases in long bones: review of the literature. Acta
Orthop Belg 1993; 59(Suppl 1):611
26 Broos P, Reynders P, van den Bogert W, et al. Surgical
treatment of metastatic fracture of the femur improvement
of quality of life. Acta Orthop Belg 1993; 59(Suppl 1):5256
27 Damron TA, Rock MG, Choudhury SN, et al. Biomechani-
cal analysis of prophylactic fixation for middle third humeral
impending pathologic fractures. Clin Orthop 1999; 363:
240248
28 Fourneau I, Broos P. Pathologic fractures due to metastatic
disease: a retrospective study of 160 surgically treated
fractures. Acta Chir Belg 1998; 98:255260
29 Turner S, Marosszeky B, Timms I, et al. Malignant spinal
cord compression: prospective evaluation. Int J Radiat Oncol
Biol Phys 1993: 26:141146
30 Sorenson S, Helweg-Larsen S, et al. Effect of high-dose
dexamethasone in carcinomatous metastatic spinal cord
compression treated with radiotherapy: a randomised trial.
Eur J Cancer 1994; 30A:2227
31 Loblaw DA, Laperriere NJ. Emergency treatment of malig-
nant extradural spinal cord compression: an evidence-based
guideline. J Clin Oncol 1998; 16:16131624
32 Jenis LG, Dunn EJ, An HS. Metastatic disease of the
cervical spine: a review. Clin Orthop 1999; 359:89103
33 Ghogawala Z, Mansfield FL, Borges LF. Spinal radiation
before surgical compression adversely affects outcomes of
surgery for symptomatic metastatic spinal cord compression.
Spine 2001; 26:818824
34 Chataigner H, Onimus M. Surgery in spinal metastasis
without spinal cord compression: indications and strategy
related to the risk of recurrence. Eur Spine J 2000; 9:523
527
35 Sundaresan N, Galicich JH, Bains MS, et al. Vertebral body
resection in the treatment of cancer involving the spine.
Cancer 1984; 53:13931396
36 Newman SJ, Hansen HH. Proceedings: frequency, diagnosis
and treatment of brain metastases in 247 consecutive pa-
tients with bronchogenic carcinoma. Cancer 1974; 33:492
496
37 Hirsch FR, Paulson OB, Hansen HH, et al. Intracranial
metastases in small cell carcinoma of the lung: prognostic
aspects. Cancer 1983; 51:529533
38 Andrews RJ, Gluck DS, Konchingeri RH. Surgical resection
of brain metastases from lung cancer. Acta Neurochir
(Wien) 1996; 138:382389
39 Kelly K, Bunn PA Jr. Is it time to reevaluate our approach to
the treatment of brain metastases in patients with non-small
cell lung cancer? Lung Cancer 1998; 20:8591
40 Cairncross JG, Kim JH, Posner J. Radiation therapy for
brain metastases. Ann Neurol 1980; 7:529541
41 Coia LR, Aaronson N, Linggood R, et al. A report of the
consensus workshop panel on the treatment of brain metas-
tases. Int J Radiat Oncol Biol Phys 1992; 23:223227
42 Shapiro WR. Intracranial neoplasms. In: Rosenberg RN, ed.
Comprehensive neurology. New York, NY: Raven Press,
1991; 157200
43 Weiss HD. Neoplasms. In: Samuel MA, ed. Manual of
neurology. 4th ed. Boston, MA: Little, Brown, 1991; 213
239
44 French LA, Galicich JH. The use of steroids for control of
cerebral edema. Clin Neurosurg 1964; 10:212223
45 Weissman DE, Dufer D, Vogel V, et al. Corticosteroid
toxicity in neuro-oncology patients. J Neurooncol 1987;
5:125128
46 Vecht CJ, Hovestadt A, Verbiest HBC, et al. Dose-effect
relationship of dexamethasone on Karnofsky performance in
metastatic brain tumors: a randomized study of doses of 4, 8,
and 16 mg per day. Neurology 1994; 44:675680
47 Weissman DE, Janjan NA, Erickson B, et al. Twice-daily
tapering dexamethasone treatment during cranial radiation
for newly diagnosed brain metastases. J Neurooncol 1991;
11:235239
48 Wolfson AH, Snodgrass SM, Schwade JG, et al. The role of
steroids in the management of metastatic carcinoma to the
brain: a pilot prospective trial. Am J Clin Oncol 1994;
17:234238
49 Diener-West M, Dobbins TW, et al. Identification of an
optimal subgroup for treatment evaluation of patients with
brain metastases using RTOG study 7916. Int J Radiat Oncol
Biol Phys 1989; 16:669673
50 Rock JP, Haines S, Recht L, et al. Practice parameters for
the management of single brain metastasis. Neurosurg
Focus 2000; 9:19
51 Patchell RA, Tibbs PA, Walsh JW, et al. A randomized trial
of surgery in the treatment of single metastases to the brain.
N Engl J Med 1990; 322:494500
52 Vecht CJ, Haaxma-Reiche H, Noordijk EM, et al. Treat-
ment of single brain metastasis: radiotherapy alone or
combined with neurosurgery? Ann Neurol 1993; 33:583590
53 Mintz AH, Kestle J, Rathbone MP, et al. A randomized trial
to assess the efficacy of surgery in addition to radiotherapy in
patients with single brain metastasis. Cancer 1996; 78:1470
1476
54 Patchell RA, Tibbs PA, Regine WF, et al. Postoperative
radiotherapy in the treatment of single metastases to the
brain: randomized trial. JAMA 1998; 280:14851489
55 Borgelt B, Gelber RD, Kramer S, et al. The palliation of
brain metastases: final results of the first two studies by the
Radiation Therapy Oncology Group. Int J Radiat Oncol Biol
Phys 1980; 6:19
56 Kurtz J, Gelber R, Brady LW, et al. The palliation of brain
metastases in a favorable patient population: a randomized
clinical trial by the Radiation Therapy Oncology Group. Int
57 Zouhair A, Tercier PA, Fankhauser H, et al. Stereotactic
radiotherapy of brain metastases: experience in Lausanne.
Schweiz Med Wochenschr 1997; 40:16521656
58 Shirato H, Takmura A, Tomita M, et al. Stereotactic irradi-
ation without whole-brain irradiation for single brain metas-
tasis. Int J Radiat Oncol Biol Phys 1997; 37:385391
59 Flickinger JC, Kondziolka D, Lunsford LD, et al. A multi-
institutional experience with stereotactic radiosurgery for
solitary brain metastasis. Int J Radiat Oncol Biol Phys 1994;
28:797802
60 Adler Jr, Cox RS, Kaplan I, et al. Stereotactic radiosurgical
treatment of brain metastases. J Neurosurg 1992; 76:444
449
61 Alexander E 3rd, Moriarty TM, Davis RB, et al. Stereotactic
radiosurgery for the definitive, noninvasive treatment of
brain metastases. J Natl Cancer Inst 1995; 87348740
62 Muacevic A, Kreth FW, Horstmann GA, et al. Surgery and
radiotherapy compared with gamma knife radiosurgery in
the treatment of solitary cerebral metastases of small diam-
eter. J Neurosurg 1999; 91:3543
63 Li B, Yu J, Suntharalingam M, et al. Comparison of three
treatment options for single brain metastasis from lung
cancer. Int J Cancer 2000; 90:3745
64 Kondziolka D, Patel A, Lunsford LD, et al. Stereotactic
radiosurgery plus whole brain radiotherapy vs radiotherapy
alone for patients with multiple brain metastases. Int J
65 Hoffman R, Sneed PK, McDermott MW, et al. Radiosur-
gery for brain metastases from primary lung carcinoma.
Cancer J 2001; 7:121131
66 Muers MF, Round CE. Palliation of symptoms in non-small
cell lung cancer: a study by the Yorkshire Regional Cancer
Organisation Thoracic Group. Thorax 1993; 48:339343
67 Reuben DB, Mor V. Dyspnea in terminally ill cancer
patients. Chest 1986; 89:234236
68 Escalante CP, Martin CG, Elting LS, et al. Dyspnea in
cancer patients. Etiology, resource utilization, and survival-
implications in a managed care world. Cancer 1996; 78:
13141319
69 Claessens MT, Lynn J, Zhong Z, et al. Dying with lung
cancer or chronic obstructive pulmonary disease: insights
from SUPPORT. Study to Understand Prognoses and Pref-
erences for Outcomes and Risks of Treatments. J Am
Geriatr Soc 2000; 48(5Suppl):S146S153
70 Smith EL, Hann DM, Ahles TA, et al. Dyspnea, anxiety,
body consciousness, and quality of life in patients with lung
cancer. J Pain Symptom Manage 2001; 21:323329
71 Dudgeon DJ, Lertzman M. Dyspnea in the advanced cancer
patient. J Pain Symptom Manage 1998; 16:212219
72 ODriscoll M, Corner J, Bailey C. The experience of breath-
lessness in lung cancer. Eur J Cancer Care (Engl) 1999;
8:3743
73 Bernhard J, Ganz PA. Psychosocial issues in lung cancer
patients: part 1. Chest 1991; 99:216223
74 Kellner R, Samet J, Pathak D. Dyspnea, anxiety, and
depression in chronic respiratory impairment. Gen Hosp
Psychiatry 1992; 14:2028
75 Gift AG, Cahill CA. Psychophysiologic aspects of dyspnea in
chronic obstructive pulmonary disease: a pilot study. Heart
Lung 1990; 19:252257
76 Stone P, Richards M, AHern R, et al. A study to investigate
the prevalence, severity and correlates of fatigue among
patients with cancer in comparison with a control group of
volunteers without cancer. Ann Oncol 2000; 11:561567
77 Abratt RP, Willcox PA. The effect of irradiation on lung
function and perfusion in patients with lung cancer. Int J
78 Bruera E, de Stoutz N, Velasco-Leiva A, et al. Effects of
oxygen on dyspnoea in hypoxaemic terminal-cancer patients.
Lancet 1993; 342:1314
79 Boyd KJ, Kelly M. Oral morphine as symptomatic treatment
of dyspnoea in patients with advanced cancer. Palliat Med
1997; 11:277281
80 Zeppetella G. Nebulized morphine in the palliation of
dyspnoea. Palliat Med 1997; 11:267275
81 Cohen MH, Anderson AJ, Krasnow SH, et al. Continuous
intravenous infusion of morphine for severe dyspnea. South
Med J 1991; 84:229234
82 Lang E, Jedeikin R. Acute respiratory depression as a
complication of nebulized morphine. Can J Anaesth 1998;
45:6062
83 Bredin M, Corner J, Krishnasamy M, et al. Multicentre
randomised controlled trial of nursing intervention for
breathlessness in patients with lung cancer. BMJ 1999;
318:901904
84 Corner J, Plant H, AHern R, et al. Non-pharmacological
intervention for breathlessness in lung cancer. Palliat Med
1996; 10:299305
85 Vaaler AK, Forrester JM, Lesar M, et al. Obstructive
atelectasis in patients with small cell lung cancer: incidence
and response to treatment. Chest 1997; 111:115120
86 Doona M, Walsh D. Benzonatate for opioid-resistant cough
in advanced cancer. Palliat Med 1998; 12:5558
87 Moroni M, Porta C, Gualtieri G, et al. Inhaled sodium
cromoglycate to treat cough in advanced lung cancer pa-
tients. Br J Cancer 1996; 74:309311
88 Luporini G, Barni S, Marchi E, et al. Efficacy and safety of
levodropropizine and dihydrocodeine on nonproductive
cough in primary and metastatic lung cancer. Eur Respir J
1998; 12:97101
89 Cohen S, Hossain SA. Primary carcinoma of the lung: a
review of 417 histologically proved cases. Dis Chest 1966;
49:6774
90 Emerson GL, Emerson MS, Sherwood CE. The natural
history of carcinoma of the lung. J Thorac Surg 1959;
37:291304
91 Johnston WW. The malignant pleural effusion: a review of
cytopathologic diagnoses of 584 specimens from 472 con-
secutive patients. Cancer 1985; 56:905909
92 Chernow B, Sahn SA. Carcinomatous involvement of the
pleura: an analysis of 96 patients. Am J Med 1977; 63:695
702
93 Estenne M, Yernault JC, De Troyer A. Mechanism of relief
of dyspnea after thoracocentesis in patients with large
pleural effusions. Am J Med 1983; 74:813819
94 Light RW, Stansbury DW, Brown SE. The relationship
between pleural pressures and changes in pulmonary func-
tion after therapeutic thoracentesis. Am Rev Respir Dis
1986; 133:658661
95 Agusti AG, Cardus J, Roca J, et al. Ventilation-perfusion
mismatch in patients with pleural effusion: effects of thora-
centesis. Am J Respir Crit Care Med 1997; 156:12051209
96 Karetzky MS, Kothari GA, Fourre JA, et al. Effect of
thoracentesis on arterial oxygen tension. Respiration 1978;
36:96103
97 Brown NE, Zamel N, Aberman A. Changes in pulmonary
mechanics and gas exchange following thoracocentesis.
Chest 1978; 74:540542
98 Krell WS, Rodarte JR. Effects of acute pleural effusion on
respiratory system mechanics in dogs. J Appl Physiol 1985;
59:14581463
99 Management of malignant pleural effusions. Am J Respir
Crit Care Med 2000; 162:19872001
100 Light RW, Jenkinson SG, Minh VD, et al. Observations on
pleural fluid pressures as fluid is withdrawn during thora-
centesis. Am Rev Respir Dis 1980; 121:799804
101 Lan RS, Lo SK, Chuang ML, et al. Elastance of the pleural
space: a predictor for the outcome of pleurodesis in patients
with malignant pleural effusion. Ann Intern Med 1997;
126:768774
102 Walker-Renard PB, Vaughan LM, Sahn SA. Chemical pleu-
rodesis for malignant pleural effusions. Ann Intern Med
1994; 120:5664
103 Hartman DL, Gaither JM, Kesler KA, et al. Comparison of
insufflated talc under thoracoscopic guidance with standard
tetracycline and bleomycin pleurodesis for control of malig-
nant pleural effusions. Thorac Cardiovasc Surg 1993; 105:
743747
104 Hamed H, Fentiman IS, Chaudary MA, et al. Comparison of
intracavitary bleomycin and talc for control of pleural effu-
sions secondary to carcinoma of the breast. Br J Surg 1989;
76:12661267
105 Fentiman IS, Rubens RD, Hayward JL. A comparison of
intracavitary talc and tetracycline for the control of pleural
effusions secondary to breast cancer. Eur J Cancer Clin
Oncol 1986; 22:10791081
106 Todd TR, Delarue NC, Ilves R, et al. Talc poudrage for
malignant pleural effusion. Chest 1980; 78:542
107 Fentiman IS, Rubens RD, Hayward JL. Control of pleural
effusions in patients with breast cancer: a randomized trial.
Cancer 1983; 52:737739
108 Yim AP, Chan AT, Lee TW, et al. Thoracoscopic talc
insufflation vs talc slurry for symptomatic malignant pleural
effusion. Ann Thorac Surg 1996; 62:1655
109 Kennedy L, Rusch VW, Strange C, et al. Pleurodesis using
talc slurry. Chest 1994; 106:342346
110 Patz EF Jr, McAdams HP, Erasmus JJ, et al. Sclerotherapy
for malignant pleural effusions: a prospective randomized
trial of bleomycin vs doxycycline with small-bore catheter
drainage. Chest 1998; 113:13051311
111 Heffner JE, Standerfer RJ, Torstveit J, et al. Clinical efficacy
of doxycycline for pleurodesis. Chest 1994; 105:17431747
112 Pulsiripunya C, Youngchaiyud P, Pushpakom R, et al. The
efficacy of doxycycline as a pleural sclerosing agent in
malignant pleural effusion: a prospective study. Respirology
1996; 1:6972
113 Moffett MJ, Ruckdeschel JC. Bleomycin and tetracycline in
malignant pleural effusions: a review. Semin Oncol 1992;
19(2Suppl5):5962
114 Martinez-Moragon E, Aparicio J, Rogado MC, et al. Pleu-
rodesis in malignant pleural effusions: a randomized study of
tetracycline vs bleomycin. Eur Respir J 1997; 10:23802383
115 Zimmer PW, Hill M, Casey K, et al. Prospective randomized
trial of talc slurry vs bleomycin in pleurodesis for symptom-
atic malignant pleural effusions. Chest 1997; 112:430434
116 Macaulay VM, OByrne KJ, Saunders MP, et al. Phase I
study of intrapleural batimastat (BB-94), a matrix metallo-
proteinase inhibitor, in the treatment of malignant pleural
effusions. Clin Cancer Res 1999; 5:513520
117 Kelly-Garcia J, Roman-Berumen JF, Ibarra-Perez C. Io-
dopovidone and bleomycin pleurodesis for effusions due to
malignant epithelial neoplasms. Arch Med Res 1997; 28:
583585
118 Bjermer L, Gruber A, Sue-Chu M, et al. Effects of in-
trapleural mitoxantrone and mepacrine on malignant pleural
effusion: a randomized study. Eur J Cancer 1995; 31A:2203
2208
119 Luh KT, Yang PC, Kuo SH, et al. Comparison of OK-432
and mitomycin C pleurodesis for malignant pleural effusion
caused by lung cancer: a randomized trial. Cancer 1992;
69:674679
120 Figlin R, Mendoza E, Piantadosi S, et al. Intrapleural
chemotherapy without pleurodesis for malignant pleural
effusions: LCSG Trial 861. Chest 1994; 106(6Suppl):363S
366S
121 Masuno T, Kishimoto S, Ogura T, et al. A comparative trial
of LC9018 plus doxorubicin and doxorubicin alone for the
treatment of malignant pleural effusion secondary to lung
cancer. Cancer 1991; 68:14951500
122 Ike O, Shimizu V, Hitomi S, et al. Treatment of malignant
pleural effusions with doxorubicin hydrochloride-containing
poly(L-lactic acid) microspheres. Chest 1991; 99:911915
123 Rosso R, Rimoldi R, Salvati F, et al. Intrapleural natural beta
interferon in the treatment of malignant pleural effusions.
Oncology 1988; 45:253256
124 Astoul P, Viallat JR, Laurent JC, et al. Intrapleural recom-
binant IL-2 in passive immunotherapy for malignant pleural
effusion. Chest 1993; 103:209213
125 Viallat JR, Boutin C, Rey F, et al. Intrapleural immunother-
apy with escalating doses of interleukin-2 in metastatic
pleural effusions. Cancer 1993; 71:40674076
126 Ponn RB, Blancaflor J, DAgostino RS, et al. Pleuroperito-
neal shunting for intractable pleural effusions. Ann Thorac
Surg 1991; 51:605609
127 Pope AR, Joseph JH. Pleuroperitoneal shunt for pneumo-
nectomy cavity malignant effusion. Chest 1989; 96:686688
128 Schulze M, Boehle AS, Kurdow R, et al. Effective treatment
of malignant pleural effusion by minimal invasive thoracic
surgery: thoracoscopic talc pleurodesis and pleuroperitoneal
shunts in 101 patients. Ann Thorac Surg 2001; 71:1809
1812
129 Reich H, Beattie EJ, Harvey JC. Pleuroperitoneal shunt for
malignant pleural effusions: a one-year experience. Semin
Surg Oncol 1993; 9:160162
130 Petrou M, Kaplan D, Goldstraw P. Management of recur-
rent malignant pleural effusions: the complementary role of
talc pleurodesis and pleuroperitoneal shunting. Cancer
1995; 75:801805
131 Chen YM, Shih JF, Yang KY, et al. Usefulness of pig-tail
catheter for palliative drainage of malignant pleural effu-
sions in cancer patients. Support Care Cancer 2000; 8:423
426
132 Pien GW, Gant MJ, Washam CL, et al. Use of an implant-
able pleural catheter for trapped lung syndrome in patients
with malignant pleural effusion. Chest 2001; 119:16411646
133 Pollak JS, Burdge CM, Rosenblatt M, et al. Treatment of
malignant pleural effusion with tunneled long-term drainage
catheters. J Vasc Interv Radiol 2001; 12:201208
134 Livingston RB, McCracken JD, Trauth CJ, et al. Isolated
pleural effusion in small cell lung carcinoma: favorable
prognosis. Chest 1982; 81:208211
135 Ramser ER, Beamis JF Jr. Laser bronchoscopy, Clin Chest
Med 1995; 16:415426
136 Colt HG, Harrell JH. Therapeutic rigid bronchoscopy allows
level of care changes in patients with acute respiratory
failure from central airways obstruction. Chest 1997; 112:
202206
137 Miyazawa T, Yamakido M, Ikeda S, et al. Implantation of
ultraflex nitinol stents in malignant tracheobronchial steno-
ses. Chest 2000; 118:959965
138 Hautmann H, Gamarra F, Pfeifer KJ, et al. Fiberoptic
bronchoscopic balloon dilatation in malignant tracheo-
bronchial disease: indications and results. Chest 2001;
120:4349
139 Ball JB, Delaney JC, Evans CC, et al. Endoscopic bougie
and balloon dilatation of multiple bronchial stenoses: 10 year
follow up. Thorax 1991; 46:933935
140 Cavaliere S, Foccoli P, Toninelli C, et al. Nd-YAG laser
therapy in lung cancer: an 11-year experience with 2,253
applications in 1,585 patients. J Bronchol 1994; 1:105111
141 Cavaliere S, Venuta F, Foccoli P, et al. Endoscopic treat-
ment of malignant airway obstructions in 2,008 patients.
Chest 1996; 110:15361542
142 Sheski FD, Mathur PN. Cryotherapy, electrocautery, and
brachytherapy. Clin Chest Med 1999; 20:123138
143 Homasson JP, Bell NJ. Endobronchial electrocautery. Se-
min Respir Crit Care Med 1997; 18:535543
144 Petrou M, Kaplan D, Goldstraw P. Bronchoscopic dia-
thermy resection and stent insertion: a cost effective treat-
ment for tracheobronchial obstruction. Thorax 1993; 48:
11561159
145 Sutedja G, van Kralingen K, Schramel FM, et al. Fibreoptic
bronchoscopic electrosurgery under local anaesthesia for
rapid palliation in patients with central airway malignancies.:
a preliminary report. Thorax 1994; 49:12431246
146 van Boxem TJ, Venmans BJ, van Mourik JC, et al. Broncho-
scopic treatment of intraluminal typical carcinoid: a pilot
study. Thorac Cardiovasc Surg 1998; 116:402406
147 van Boxem TJ, Venmans BJ, Schramel FM, et al. Radio-
graphically occult lung cancer treated with fibreoptic bron-
choscopic electrocautery: a pilot study of a simple and
inexpensive technique. Eur Respir J 1998; 11:169172
148 Coulter TD, Mehta AC. The heat is on: impact of endo-
bronchial electrosurgery on the need for Nd:YAG laser
photoresection. Chest 2000; 118:516521
149 Morice RC, Ece T, Ece F, et al. Endobronchial argon
plasma coagulation for treatment of hemoptysis and neoplas-
tic airway obstruction. Chest 2001; 119:781787
150 Mathur PN, Wolf KM, Busk MF, et al. Fiberoptic broncho-
scopic cryotherapy in the management of tracheobronchial
obstruction. Chest 1996; 110:718723
151 Marasso A, Gallo E, Massaglia GM, et al. Cryosurgery in
bronchoscopic treatment of tracheobronchial stenosis: indi-
cations, limits, personal experience. Chest 1993; 103:472
474
152 Rodgers BM, Moazam F, Talbert JL. Endotracheal cryo-
therapy in the treatment of refractory airway strictures. Ann
Thorac Surg 1983; 35:5257
153 Maiwand MO, Homasson JP. Cryotherapy for tracheobron-
chial disorders. Clin Chest Med 1995; 16:427443
154 Walsh DA, Maiwand MO, Nath AR, et al. Bronchoscopic
cryotherapy for advanced bronchial carcinoma. Thorax 1990;
45:509513
155 Bolliger CT, Heitz M, Hauser R, et al. An airway Wallstent
for the treatment of tracheobronchial malignancies. Thorax
1996; 51:11271129
156 Monnier P, Mudry A, Stanzel F, et al. The use of the
covered Wallstent for the palliative treatment of inoperable
tracheobronchial cancers: a prospective, multicenter study.
Chest 1996; 110:11611168
157 Witt C, Dinges S, Schmidt B, et al. Temporary tracheobron-
chial stenting in malignant stenoses. Eur J Cancer 1997;
33:204208
158 Dumon JF. A dedicated tracheobronchial stent. Chest 1990;
97:328332
159 Homasson JP, Renault P, Angebault M, et al. Bronchoscopic
cryotherapy for airway strictures caused by tumors. Chest
1986; 90:159164
160 Dumon JF, Cavaliere S, Diaz-Jimenez P, et al. Seven-year
experience with the Dumon prosthesis. J Bronchol 1996;
3:610
161 Schray MF, McDougall JC, Martinez A, et al. Management
of malignant airway compromise with laser and low dose rate
brachytherapy: The Mayo Clinic experience. Chest 1988;
93:264269
162 Villanueva AG, Lo TC, Beamis JF Jr. Endobronchial brachy-
therapy. Clin Chest Med 1995; 16:445454
163 Ofiara L, Roman T, Schwartzman K, et al. Local determi-
nants of response to endobronchial high-dose rate brachy-
therapy in bronchogenic carcinoma. Chest 1997; 112:946
953
164 Bedwinek J, Petty A, Bruton C, et al. The use of high dose
rate endobronchial brachytherapy to palliate symptomatic
endobronchial recurrence of previously irradiated broncho-
genic carcinoma. Int J Radiat Oncol Biol Phys 1992; 22:
2330
165 Gauwitz M, Ellerbroek N, Komaki R, et al. High dose
endobronchial irradiation in recurrent bronchogenic carci-
noma. Int J Radiat Oncol Biol Phys 1992; 23:397400
166 Hernandez P, Gursahaney A Roman T, et al. High dose rate
brachytherapy for the local control of endobronchial carci-
noma following external irradiation. Thorax 1996; 51:354
358
167 Micke O, Prott FJ, Schafer U, et al. Endoluminal HDR-
brachytherapy as a palliative treatment of patients with
the recurrence of a non-small cell bronchial carcinoma
after prior radiotherapy. Strahlenther Onkol 1995; 171:
554559
168 Nori D, Allison R, Kaplan B, et al. High dose-rate intralu-
minal irradiation in bronchogenic carcinoma: technique and
results. Chest 1993; 104:10061011
169 Pisch J, Villamena PC, Harvey JC, et al. High dose-rate
endobronchial irradiation in malignant airway obstruction.
Chest 1993; 104:721725
170 Seagren SL, Harrell JH, Horn RA. High dose rate intralu-
minal irradiation in recurrent endobronchial carcinoma.
Chest 1985; 88:810814
171 Sutedja G, Baris G, Schaake-Koning C, et al. High dose rate
brachytherapy in patients with local recurrences after radio-
therapy of non-small cell lung cancer. Int J Radiat Oncol
Biol Phys 1992; 24:551553
172 Langendijk H, de Jong J, Tjwa M, et al. External irradiation
vs external irradiation plus endobronchial brachytherapy in
inoperable non-small cell lung cancer: a prospective ran-
domized study. Radiother Oncol 2001; 58:257268
173 Burt PA, ODriscoll BR, Notley HM, et al. Intraluminal
irradiation for the palliation of lung cancer with high dose
rate micro-Selectron. Thorax 1990; 45:765768
174 Chang LF, Horvath J, Peyton W, et al. High dose rate
afterloading intraluminal brachytherapy in malignant airway
obstruction of lung cancer. Int J Radiat Oncol Biol Phys
1994; 28:589596
175 Gollins SW, Burt PA, Barber PV, et al. High dose rate
intraluminal radiotherapy for carcinoma of the bronchus:
outcome of treatment of 406 patients. Radiother Oncol
1994; 33:3140
176 Huber RM, Fischer R, Hautmann H. Palliative endobron-
chial brachytherapy for central lung tumors: a prospective,
randomized comparison of two fractionation schedules.
Chest 1995; 107:463470
177 Mehta MP, Shahabi S, Jarjour NN, et al. Endobronchial
irradiation for malignant airway obstruction. Int J Radiat
Oncol Biol Phys 1989; 17:847851
178 Speiser BL, Spratling L. Remote afterloading brachytherapy
for the local control of endobronchial carcinoma. Int J
179 Anacak Y, Mogulkoc N, Ozkok S, et al. High dose rate
endobronchial brachytherapy in combination with external
beam radiotherapy for stage III non-small cell lung cancer.
Lung Cancer 2001; 34:253259
180 Langendijk JA, Tjwa MK, de Jong JM, et al. Massive
haemoptysis after radiotherapy in inoperable non-small cell
lung carcinoma: is endobronchial brachytherapy really a risk
factor? Radiother Oncol 1998; 49:175183
181 Edell ES, Cortese DA. Bronchoscopic phototherapy with
hematoporphyrin derivative for treatment of localized bron-
chogenic carcinoma: a 5-year experience. Mayo Clin Proc
1987; 62:814
182 Cortese DA, Edell ES, Kinsey JH. Photodynamic therapy
for early stage squamous cell carcinoma of the lung. Mayo
Clin Proc 1997; 72:595602
183 Moghissi K, Dixon K, Stringer M, et al. The place of
bronchoscopic photodynamic therapy in advanced unresect-
able lung cancer: experience of 100 cases. Eur J Cardiotho-
rac Surg 1999; 15:16
184 Miller RR, McGregor DH. Hemorrhage from carcinoma of
the lung. Cancer 1980; 46:200205
185 Chute CG, Greenberg ER, Baron J, et al. Presenting
conditions of 1539 population-based lung cancer patients by
cell type and stage in New Hampshire and Vermont. Cancer
1985; 56:21072111
186 Hyde L, Hyde CI. Clinical manifestations of lung cancer.
Chest 1974; 65:299306
187 Grippi MA. Clinical aspects of lung cancer. Semin Roent-
genol 1990; 25:1224
188 Corey R, Hla KM. Major and massive hemoptysis: reassess-
ment of conservative management. Am J Med Sci 1987;
294:301309
189 Cahill BC, Ingbar DH. Massive hemoptysis: assessment and
management. Clin Chest Med 1994; 15:147167
190 Jean-Baptiste E. Clinical assessment and management of
massive hemoptysis. Crit Care Med 2000; 28:16421647
191 Strange C. Double-lumen endotracheal tubes. Clin Chest
Med 1991; 12:497506
192 Zavala DC. Pulmonary hemorrhage in fiberoptic transbron-
chial biopsy. Chest 1976; 70:584588
193 Worth H, Breuer HWM, Charchut S, et al. Endobronchial
vs intravenous application of glypressin for the therapy and
prevention of lung bleeding during bronchoscopy. Am Rev
Respir Dis 1987; 135(4 part 2):A108
194 Sahebjami H. Iced saline lavage during bronchoscopy. Chest
1976; 69:13113
195 Conlan AA, Hurwitz SS. Management of massive haemop-
tysis with the rigid bronchoscope and cold saline lavage.
Thorax 1980; 35:90190
196 Hiebert CA. Balloon catheter control of life-threatening
hemoptysis. Chest 1974; 66:30830
197 Gottlieb LS, Hillberg R. Endobronchial tamponade therapy
for intractable hemoptysis. Chest 1975; 67:482483
198 Saw EC, Gottlieb LS, Yokahama T, et al. Flexible fiberoptic
bronchoscopy and endobronchial tamponade in the manage-
ment of hemoptysis. Chest 1976; 70:58951
199 Swersky RB, Chang JB, Wisoff BG, et al. Endobronchial
balloon tamponade of hemoptysis in patients with cystic
fibrosis. Ann Thorac Surg 1979; 27:262264
200 Hetzel MR, Smith SG. Endoscopic palliation of tracheo-
bronchial malignancies. Thorax 1991; 46:325333
201 Jain PR, Dedhia HV, Lapp NL, et al. Nd-YAG laser followed
by radiation for treatment of malignant airway lesions.
Lasers Surg Med 1985; 5:4753
202 Mal H, Rullon I, Mellot F, et al. Immediate and long-term
results of bronchial artery embolization for life-threatening
hemoptysis. Chest 1999; 115:9961001
203 White RI Jr. Bronchial artery embolotherapy for control of
acute hemoptysis: analysis of outcome. Chest 1999; 115:
912915
204 Osaki S, Nakanishi Y, Wataya H, et al. Prognosis of bronchial
artery embolization in the management of hemoptysis.
Respiration 2000; 67:412416
205 Knott-Craig CJ, Oostuizen JG, Rossouw G, et al. Manage-
ment and prognosis of massive hemoptysis: recent experi-
ence with 120 patients. Thorac Cardiovasc Surg 1993;
105:394397
206 Hoegler D. Radiotherapy for palliation of symptoms in
incurable cancer. Curr Probl Cancer 1997; 21:129183
207 Freitag L, Tekolf E, Steveling H, et al. Management of
malignant esophago-tracheal fistulas with airway stenting
and double stenting. Chest 1996; 110:11551160
208 Colt HG, Meric B, Dumon JF. Double stents for carcinoma
of the esophagus invading the tracheo-bronchial tree. Gas-
trointest Endosc 1992; 38:485489
209 Alexiou C, Neuhaus H, Kau RJ, et al. Treatment of esoph-
agorespiratory fistula by insertion of an esophageal Mont-
gomery and tracheal dynamic stent after failure of conven-
tional endoprosthesis. ORL J Otorhinolaryngol Relat Spec
1998; 60:5154
210 Baker GL, Barnes HJ. Superior vena cava syndrome:
etiology, diagnosis, and treatment. Am J Crit Care 1992;
1:5464
211 Muers MF. Lung cancer: quality of life and symptom
control. Eur Respir J 2001; 6:305329
212 Wurschmidt F, Bunemann H, Heilmann HP. Small cell lung
cancer with and without superior vena cava syndrome: a
multivariate analysis of prognostic factors in 408 cases. Int J
213 Kawanami S, Imada H, Terashima H, et al. Radiotherapy of
superior vena cava syndrome. Jpn J Lung Cancer 1996;
36:745752
214 Sasano S, Onuki T, Mae M, et al. Wallstent endovascular
prosthesis for the treatment of superior vena cava syndrome.
Jpn Thorac Cardiovasc Surg 2001; 49:165170
215 Tanigawa N, Sawada S, Mishima K, et al. Clinical outcome
of stenting in superior vena cava syndrome associated with
malignant tumors: comparison with conventional treatment.
Acta Radiol 1998; 39:669674
216 Nicholson AA, Ettles DF, Arnold A, et al. Treatment of
malignant superior vena cava obstruction: metal stents or
radiation therapy. J Vasc Interv Radiol 1997; 8:781788
2003;123;284-311 Chest
Paul A. Kvale, Michael Simoff and Udaya B. S. Prakash
Palliative Care
& Services
Updated Information
S
References
S#BIBL
free at:
Citations
S#otherarticles
Reprints
2003;123;312-331 Chest
Ackerman, Melinda Thompson and F. Hammond Cole, Jr
John P. Griffin, Judith E. Nelson, Kathryn A. Koch, Harvey B. Niell, Terrence F.
End-of-Life Care in Patients With Lung Cancer
ISSN: 0012-3692.
End-of-Life Care in Patients With Lung
Cancer*
John P. Griffin, MD, FCCP; Judith E. Nelson, MD, JD, FCCP;
Kathryn A. Koch, MD, FCCP; Harvey B. Niell, MD;
Terrence F. Ackerman, PhD; Melinda Thompson, MD, JD; and
F. Hammond Cole, Jr., MD, FCCP
Evidence-based practice guidelines for end-of-life care for patients with lung cancer have been
previously available only from the British health-care system. Currently in this setting, there has
been increasing concern in attaining control of the physical, psychological, social, and spiritual
distress of the patient and family. This American College of Chest Physicians-sponsored
multidisciplinary panel has generated recommendations for improving quality of life after
examining the English-language literature for answers to some of the most important questions
in end-of-life care. Communication between the doctor, patient, and family is central to the active
total care of patients with disease that is not responsive to curative treatment. The advance care
directive, which has been slowly evolving and is presently limited in application and often
circumstantially ineffective, better protects patient autonomy. The problem-solving capability of
the hospital ethics committee has been poorly utilized, often due to a lack of understanding of its
composition and function. Cost considerations and a sense of futility have confused caregivers as
to the potentially important role of the critical care specialist in this scenario. Symptomatic and
supportive care provided in a timely and consistent fashion in the hospice environment, which
treats the patient and family at home, has been increasingly used, and at this time is the best
model for end-of-life care in the United States. (CHEST 2003; 123:312S331S)
Key words: advance directive; communication; critical care; end-of-life care; hospice; hospital ethics committee; lung
cancer; practice guidelines
Abbreviations: CPR cardiopulmonary resuscitation; DNR do not resuscitate; HEC hospital ethics committee;
SUPPORT Study to Understand Prognoses and Preferences for Risks of Treatment; WHO World Health
Organization
A
fter years of neglect, care at the end of life is
receiving increasing attention and concern.
When end of life is near, the patient is suffering the
effects of a progressive and mortal illness, and is
coping not only with bodily symptoms, but also with
the existential crisis of approaching death.
1
The
purpose of this communication is to offer guidelines
in this important area, specific to lung cancer, the
most frequent cancer killer of men and women.
Although the imperative of care is providing optimal
symptom relief and alleviation of suffering, there is
clear evidence in the current medical literature that
we are failing to do this.
2,3
Despite wide dissemina-
tion of pain management guidelines,
4
many patients
with lung cancer continue to suffer not only from
pain, but also from other troubling symptoms and
interpersonal scenarios in their final days. The most
effective approach to providing better care in pa-
tients with other diseases has been the use of clinical
practice guidelines based on the delivery of evi-
dence-based medicine.
5
The ethical and professional
challenge to do so is as important as the obligation to
cure.
6
End-of-life care is defined as the active, total care
of patients whose disease is not responsive to cura-
tive treatment.
7
The philosophy of this care is to
attain maximal quality of life through control of the
myriad physical, psychological, social, and spiritual
distress of the patient and family.
*From the From the Division of Pulmonary and Critical Care
Medicine (Drs. Griffin and Thompson), Division of Hematology/
Oncology (Dr. Niell), Department of Medicine, the Department
of Human Values and Ethics (Dr. Ackerman), the Section of
Thoracic and Cardiovascular Surgery (Dr. Cole), Department of
Surgery, College of Medicine, The University of Tennessee
Health Science Center, Memphis, TN; the Division of Pulmo-
nary and Critical Care Medicine (Dr. Nelson), Department of
Medicine, Mount Sinai Medical Center, New York, NY; and the
Division of Pulmonary and Critical Care Medicine (Dr. Koch),
Department of Internal Medicine, University of Florida Health
Science Center, Jacksonville, FL.
Correspondence to: John P. Griffin, MD, FCCP, Division of
Pulmonary and Critical Care Medicine, University of Tennessee,
956 Court Ave, Room H 314, Memphis, TN 38163; e-mail:
jpgriffin@utmem.edu
Central to ensuring quality of all care at the end of
life is communication between the doctor, patient,
and family.
8
Teaching how to break bad news has
been the subject of 166 articles from 1975 to 1999,
the majority published in the past 5 years, but
15% were based on controlled trials.
9
In addition
to the many important tenets relative to the patient,
the family and its ethnic, cultural, and religious roots
must be taken into account. Although many profes-
sionals feel awkward in talking about the end of life,
family members face similar challenges in expressing
their feelings and asking questions about prognosis.
In a study from eight cancer centers, doctors consid-
ered that they had more trouble communicating with
families than with the patients themselves.
10
Rela-
tives often felt left out or in the way, which is
particularly disturbing since impending death has
such a profound impact on the family, with members
often recalling in exquisite detail the lack of sensi-
tivity of the doctor and staff. Such memories affect
the grieving process, especially how attentive the
doctor and staff were in controlling the patients
distress and physical symptoms.
11,12
In a study of the
implications for relatives of 200 consecutive, nonsur-
gically treated patients with lung cancer, monthly
case histories, questioning of nursing staff and house
physicians, was supplemented by home visits of
nearest relatives to discuss illness and death.
13
Ninety percent of patients died within 1 year, with
approximately 40% having no pain, and approxi-
mately 25% neither needed nor received any treat-
ment. Approximately 80% of nearest relatives said
the illness was not as bad as anticipated, and approx-
imately 20% accepted the clinical course as about
what they had expected. Spending time with the
relatives who are in the front line and need all the
encouragement they can get enhances mutual con-
fidence between the patient and his medical team.
Approximately two thirds of close relatives found
that a simple sketch of the likely general deteriora-
tion of the patient with loss of weight and strength
and a fortnight in bed at the end made the illness
sufficiently tangible that they could deal with it, but
in approximately 25% the death was still unexpected,
such as with the 10 patients who had sudden massive
hemoptysis. Of the approximate 25% who were
dissatisfied with management of the illness, pain
relief, delayed diagnosis, and nursing care setting
were the main problems. Steps advised to improve
these deficiencies were as follows: (1) training of
doctors in communication skills as critical to ensure
quality end-of-life care, (2) clinical research to de-
termine the best teaching methods, and (3) educa-
tion of family members in the end-of-life physical
caregiver role and its psychological impact.
1
Achieving effective pain management has been a
priority over the past decade. The American Pain
Society, Agency for Health Care Research and Pol-
icy, World Health Organization (WHO), and Na-
tional Comprehensive Cancer Network guidelines
provide algorithms for decision making in pain man-
agement.
7,1416
Pain is one of the most prevalent
symptoms across terminal illnesses, affecting more
than one third of patients, and also a source of great
fear in anticipation of final days of agony. Consider-
ing all types of malignancy, 70 to 90% of patients
have pain, and 50% die without adequate pain
relief.
17
Pain also impairs psychosocial functioning,
causes anxiety and depression, and limits capacity for
enjoyment at the end of life. The American Pain
Society, National Comprehensive Cancer Network,
and Agency for Health Care Research and Quality
guidelines provide algorithms for management of
nociceptive and neuropathic pain of varying severity
and chronicity.
18
Identification of type of pain, use of
tools to assess pain severity and response to treat-
ment, evaluation of effect of interventions on mental
alertness, and flexibility of treatment regimens are
mainstays of pain management. Clinician education
of proper dosing and medication combinations facil-
itates better care, and use of adjunctive psychotropic
drugs and behavioral interventions are effective.
Implementation of guidelines has been impaired by
misconceptions about dependence and addiction,
risks of over sedation, and regulatory problems of
opiates. (Additional information on management of
pain in specific clinical scenarios due to metastatic
lung cancer is contained in the chapter on palliative
care in these guidelines.)
Review of the Literature
A comprehensive search covering the past five
years of English-language medical literature for
practice guidelines on end-of-life care for patients
with lung cancer, has revealed only specific contri-
butions from the United Kingdom. These guidelines,
while comprehensive for their National Health Ser-
vice, are difficult to apply in medical practice in the
United States but are listed for our consideration.
British Thoracic Society
The following recommendations
19
are based on
evidence from expert committee reports or the
clinical experience of respected authorities: (1) ob-
taining agreement about the initial organization of
the palliative phase with the patient, his or her
family, and the primary care physician should be the
responsibility of the relevant specialist; (2) all pa-
tients should receive regular follow-up by a member
of the managing team; frequency depends on cir-
cumstances, and arrangements should be according
to the needs and wishes of patients and their care-
givers; (3) detailed coordination and liaison between
the patient, his or her primary care physician, and
specialists are usually best done by a specialist nurse;
(4) all patients should be made aware of the doctor(s)
supervising their care, and who is the assigned
specialist nurse; (5) patients should be aware of who
and how to call for urgent problems; (6) patients
wishes should be explicitly sought when there are
major decisions to be made about changes in the
palliative care pattern; (7) all cancer specialty units
should have the facility to admit patients directly
from primary care for symptom control; and (8) all
correspondence detailing care plans or reviews
should be copied to primary care physicians and to
specialist nurses.
Scottish Intercollegiate Guidelines Network
20
are based on
evidence from expert committee reports or the
clinical experience of respected authorities: (1)
palliation of symptoms, physical and nonphysical,
should be an integral part of the care of all patients
with cancer; (2) referral to a specialist palliative
care unit should be considered to augment support
for the patient and caregivers; (3) locally agreed
standards of care incorporating current research
and best practice should be implemented for each
modality of cancer nursing; (4) collaborative care
should ensure best practice and care of patients in
both primary and secondary settings; (5) ideally,
patients with lung cancer and their relatives
should have easy access to an appropriately trained
specialist nurse throughout their illness; and
(6) nursing care of patients with lung cancer
should be viewed as collaborative within the wider
care team and should focus on family centered
care.
National Health Service Improving Outcomes in
Lung Cancer
21
are based on
evidence from nonrandomized trials or observational
studies: (1) at every stage, patients and their relatives
should be offered clear, full and prompt information
in both verbal and written form; (2) all health
professionals involved in each patients care should
know what the patient has been told; (3) short
courses of palliative radiotherapy should be available
without delay for patients with chest pain due to
cancer; (4) specialist palliative care should be avail-
able for all patients, both within the hospital and in
the community.
Based on evidence from randomized controlled
trials or systematic reviews of such trials, an addi-
tional recommendation would be that effective pain
relief should be available promptly, using WHO or
other guidelines with stepwise analgesia, and ade-
quacy of control assessed.
How Important Is Communication With
Patient and Family in End-of-Life Care in
Patients With Lung Cancer?
There is an increasing body of literature that
reveals that patients want more information from
their clinician than they receive. They want an
opportunity to discuss their preferences and their
goals of treatment within the context of the medical
facts of their illness. They want to discuss plans for
the end of life, but they may need the clinician to
initiate the discussion. Death has the power to
surprise even the prepared, but many of our patients
are not prepared.
We do not yet know how improving communica-
tion between patient and physician will change the
outcome of an illness.
22
We do know that we are not
doing a very good job of communicating with our
patients about how they want to face whatever it is
that is about to happen.
2327
Objective studies of
interactions between physicians and cancer patients
reveal that clinicians spend little time probing the
psychosocial aspects of the patients illness.
28,29
In
particular, we are failing to make their actual treat-
ment match their preferences for treatment. Better
communication at the patient-surrogate-physician
level is needed to improve congruence between
patient wishes and the life-sustaining care they actu-
ally receive.
Nowhere has this issue been explored better than
in the Study to Understand Prognoses and Prefer-
ences for Risks of Treatment (SUPPORT).
30,31
In a
huge multicenter study that spanned 5 years, it was
demonstrated after a 2-year baseline observation
period, during which time 4,301 patients were en-
rolled, there was a mismatch between patient treat-
ment preferences and treatment actually received
and that there were clear barriers to communica-
tion.
17
These were seriously ill patients with a pre-
dicted 50% 6-month mortality. After a year of ana-
lyzing results and planning an intervention that, it
was hoped, would improve that match, there was a
2-year intervention period that expanded the total
patient database to 9,105 seriously ill patients.
Trained nurses talked to patients and families to
obtain additional information about patient prefer-
ences. They provided this information, as well as
objective prediction of patient outcome, to the clini-
cians. The clinicians discussed this information with
patients and families in only 15% of cases.
17
Even if patients have written advance directives,
their wishes may not be honored. Only 618 of 3,058
patients (20.2%) in the baseline SUPPORT group
were found to have advance directives, and in only
70 of 618 patients (11.3%) were these mentioned in
the medical record.
32
Only slightly more than half of
patients who wished to forego cardiopulmonary
resuscitation (CPR) in this group had a do-not-
resuscitate (DNR) order, and half of the written
DNR orders were placed in the medical record
within 48 h of death.
17,32
Their physicians accurately identified patient prefer-
ences for only 861 of 1,513 patients (56.9%) preferring
CPR, and for only 380 of 808 patients (47%) preferring
DNR in the baseline study group.
33
The intervention
produced literally no change in DNR orders or in days
spent in pain or other undesirable states.
17
It slightly
improved concurrence between patients and physicians
on the CPR/DNR question. There is no substitute
for the physician in exploring patient prognosis and
preferences.
34
If matters are left until the patient is too ill to
participate in decision making, and in the absence of
adequate preparation through a preliminary discus-
sion with the patient, there is no guarantee that the
family or surrogate will adequately represent what
the patient would have wanted.
3539
Patient self-
expressed preferences are relatively stable over
time
40
; however, in SUPPORT, 17% of baseline
patients and 20% of intervention patients did reverse
their preferences for DNR.
17
Talking About Life and Death
It is remarkably difficult to talk about death and its
meaning.
29,41
Clinicians may have their own personal
fears and a death anxiety.
29
They may lack training,
knowledge, and experience in giving bad news.
Formal training communication skills and increasing
availability of a wide variety of resources on commu-
nication can enhance the clinicians ability to relay
difficult information.
29,4245
This can be done with-
out increasing the patients emotional distress and
without lengthening the patient visit, as shown in a
prospective randomized trial of 69 physicians with an
8-h videotaped educational intervention on commu-
nication.
46
Too often we leave discussion about difficult mat-
ters until the ultimate point in the ICU or when
death is imminent and in a setting of crisis.
47
Virtu-
ally all patients with lung cancer are at significant risk
of death. We do not convey this knowledge to them
effectively.
23,24
In a prospective study of 326 patients
with cancer at five Chicago hospices, physicians
formulated prognoses in 300 of 311 evaluable pa-
tients (96.5%).
23
Physicians reported that they would
not communicate any survival estimate 22.7% of the
time, and would communicate survival estimates
different from the ones they had actually formulated
40.3% of the time. Of these discrepant communica-
tions, 70.2% were more optimistic than the actual
prediction.
Certainly the patient might be cured by interven-
tion, or might die from an unrelated event before the
disease has a chance to kill, but the seeds that the
patient may not be cured should be sown at the very
beginning of the management process.
48
A patient-
centered model of care from the very beginning
enhances and facilitates later transition to a palliative
approach to care.
49
At the same time that the diagnosis is made and
decisions about options in treatment are being ex-
plored, the physician should initiate discussion about
the patients concerns, preferences, and goals in life
as well as goals in treatment. Given the medical facts
of the patients situation (the particular circum-
stances of the case), given who the patient is as an
individual (what is important to him), how do the
different options in care match up with optimal
quality of life for however much time remains? From
these elements, the clinician and patient can weave a
plan of care, acknowledging that the plan is subject
to change based on the patients changing circum-
stances (failure to respond to a chosen option in
treatment, development of new and disturbing
symptoms). The elements necessary to formulate
this care plan include the following
50
: (1) a compre-
hensive assessment of the medical facts of the indi-
vidual patients situation, to include a realistic and as
accurate as possible appraisal (not too rosy, not too
grim) of what is likely to happen; (2) an understand-
ing of who that person is and what he or she values
in life (as well as in death); (3) together, these form
the assessment of what kind of quality of life the
patient could expect from different therapeutic
choices; and finally (4) extrinsic factors must be
included (resources, family situation, etc).
The patient should be encouraged to make appro-
priate arrangements for personal matters. These
include updating his or her will, thinking about
end-of-life care and advance directives, and discuss-
ing these with an appropriate surrogate. All of these
discussions should be both compassionate and cul-
turally sensitive. Even something as simple as an
extra 40 s of compassion on an educational videotape
about breast cancer has been shown to reduce
anxiety in 123 breast cancer survivors and 87 women
without breast cancer, although it did not improve
recall of the educational information.
51
Above all, the
patient should be allowed to express their reaction to
the situation.
52
If there will be a day-to-day contact person who
will coordinate care for the patient with all of the
different consultants and resources, that person
should be present at these discussions to facilitate
congruence between patient preferences and care
delivered. These conversations should be open
minded and nonjudgmental; we must make sure that
our own prejudices and beliefs do not overly color
this joint discussion-making process.
53
Consider a
written treatment plan to enhance communication
and understanding of the plan among all parties.
24
Predictions: Truth and Uncertainty
No time frame should be placed on what the
individual patient is likely to experience with differ-
ent choices in medical treatment. Although they
need to know what kind of result most patients like
themselves experience (eg, 20% 5-year survival), it
must be very clear that we cannot predict with
certainty what that individuals outcome will be for
different choices in treatment. Most people want to
know the probability of being pain free and able to
care for themselves in the immediate future. They
want to know what to expect 2 months from now and
6 months from now.
31
The statement that every patient is unique
should not be used to avoid this careful discussion of
prognosis. There is a strange collusion that may
occur between patient and doctor. In a study of 35
patients with small cell lung cancer in the Nether-
lands, false optimism was observed routinely during
early phases of treatment.
54
The doctor does and
does not want to tell the patient of a dismal progno-
sis. The patient does and does not want to know.
Both may become focused on the short-term goals of
treatment to the neglect of the long-term goals as a
result of this dynamic. Care must be given to ensure
that prognostic information is not distorted, as sur-
vival may be systematically overestimated by clini-
cians.
23
In SUPPORT, 4,028 physicians subjective predic-
tions of patient outcome during the intervention
were obtained, paired with objective prediction and
actual mortality. Physicians were slightly more pes-
simistic, but also more accurate than the objective
model in the patients with higher probability of
mortality, but slightly more optimistic and less accu-
rate for the patients with lower probability of mor-
tality.
55
Other studies have suggested that physician
predictions may be too optimistic.
56
There can be multiple layers of uncertainty in this
situation; since we physicians are not good at esti-
mating when an individual is going to die, we may
not communicate our estimate accurately to the
patient, and the patient may misinterpret or reject
our estimate.
54,57
Prognostic information should be
vague enough to include the truth (the median time
frame until death) but specific enough to help
people make appropriate plans.
There is no substitute to talking to the patient, but
the family, particularly the patients surrogate,
should also be included to facilitate both congruence
between the patients preferences and actual treat-
ment and also family knowledge about the patients
illness and expected outcome. Quill and Brody
58
propose an enhanced autonomy model of decision
making in which active dialogue between patient and
physician enables the patient to participate in deci-
sions as fully informed of medical realities as possi-
ble. Consider using the SPIKES six-step proto-
col
59
: S get the Setting right; P understand the
patients Perception of the illness; I obtain the
Invitation to impart information; K provide
Knowledge and education; E respond to the pa-
tients Emotion with empathy; and S provide
Summary strategy. Communication skills of junior
clinicians may be improved by including them in
these conversations to provide real-time role model-
ing for their own professional growth in communi-
cations skills.
Transitions to Palliative and Comfort Measures
Patients not only want to be free from pain and
suffering as they die, they also wish to have the
opportunity to make peace with God, to resolve
personal conflicts, and to make financial plans before
death.
60
A little compassion goes a long way toward
facilitating patient comfort in the face of difficult
decisions.
51
The first hurdle is to acknowledge that the patient
is likely to dieto use the D word. A euphemism
simply will not do; it is too subject to misinterpreta-
tion. If the patient (or family) is first asked if they
want complete information, there is no reason to
withhold the information that the patient is likely to
die, and every reason to share this knowledge. This
will enable a transition in focus of care, and enhance
planning and preparation for death.
52,6163
When approaching the patient for whom primary
treatment for lung cancer has failed, or for whom the
treatment has become worse than the disease, a
comprehensive re-evaluation of goals of treatment
must be sought. No longer can the patient realisti-
cally include cure as a treatment goal, even if that
was a possibility to begin with. The choices in
therapeutic options are increasingly restrained as the
potential benefits of treatment diminish and as the
real burdens of treatment escalate. The level of
treatment must be adapted to the individual patients
life goals in the face of death. Potentially useful
questions to facilitate this transition were reviewed
by Lo et al
52
for the American College of Physicians/
American Society of Internal Medicine end-of-life
care consensus panel.
All too frequently, the caregivers are not suffi-
ciently in tune with the patients experience of illness
and suffering and fail to address these aspects of the
patients illness adequately.
49
For other cancers,
inadequate attention to this has been associated with
an increase in patient utilization of homeopathic or
alternative medicine. This is presumably out of
frustration at the inability of the primary team to deal
effectively with such aspects of care.
6466
Including a
palliative care or hospice team with the cancer team
has mutual advantages to the clinicians and provides
the patient joint access not only to individuals skilled
in trying to cure the primary disease and in managing
life, but also to individuals skilled at assessing pain
and emotional and spiritual distress, and managing
death.
67,68
During whatever period of time the patient will
have, the focus of treatment will be on enabling
the dying person to live until death occurs. Goals
of treatment would be to optimize physical per-
formance and minimize suffering so that the pa-
tient can perform at his or her own maximum
potential. Goals of treatment should also include
assistance with the personal tools to make amends
with others as needed, to say goodbye, and to
pursue matters with the deepest meaning to the
patient so that he or she can face the end of life
with a sense of completion.
69
Treatment should be
directed toward extending life rather than pro-
longing death, toward reducing suffering both
physical and spiritual, toward achieving accep-
tance rather than denial or delusion.
Giving bad news is a difficult task and requires
physician competence and facility with communi-
cation under stressful circumstances. Physicians
should avail themselves of the increasing body of
educational resources to improve communication
at the end of life. Strategies for facilitating a
transition of focus of care to palliative and comfort
measures include the following
29,59,70
: (1) develop-
ing rapport; (2) finding out what the patient and
family already know; (3) identifying preferences
for receipt of information (amount and complexi-
ty); (4) giving the information in a sensitive but
straight-forward manner; (5) responding to emo-
tions; (6) establishing the overall goals of treat-
ment given the patients personal goals, the med-
ical facts, and the available technology; (7) and
finalizing the care plan, selecting elements based
on goals of treatment.
Giving bad news, establishing the care plan, and
counseling in advance care planning can all be coded
in the United States under current rules as counsel-
ing time. If the patient is already critical and the
discussion produces a change in care plan, it can be
coded as critical care time. In the critical patient, it
must be clear that any change in treatment plan does
not represent a change in caring.
71
Involving the patient and family in decision
making enables them to have as much control as
possible over the dying process. Death is the
ultimate outcome for every life. In general, people
meet their deaths the same way they approach life;
if they have time before their death, then they
need to have the tools to help them use this time
wisely. We need to do our best to give them both
the time and the tools.
Recommendations
1. For patients with lung cancer at the end of life,
it is recommended that clinicians increase their
focus on the patients experience of illness to
improve congruence of treatment with patient
goals and preferences: (a) be realistic, practical,
sensitive and compassionate; (b) listen; (c) al-
low/invite the patient to express his or her
reaction to the situation; (d) provide a contact
person; and (e) continually reassess the pa-
tients goals of therapy as part of treatment
planning. Evidence: poor; benefit: substantial;
2. For all patients with lung cancer, end-of-life
planning should be integrated as a component
of assessment of goals of treatment, and treat-
ment planning. Evidence: poor; benefit: sub-
stantial; grade of recommendation: C
3. For patients with lung cancer, an experienced
clinician should inform the patient of the diagno-
sis and its meaning. The day-to-day contact per-
son should also be present at this meeting and
should coordinate care. Evidence: poor; benefit:
substantial; grade of recommendation: C
4. Clinicians treating patients with lung cancer
should avail themselves of the increasing body
of educational resources to improve communi-
cation at the end of life. Evidence: fair; benefit:
5. With patients with lung cancer, hospice and/or
the palliative care service should be involved
early in the patients treatment, as part of the
team. Evidence: poor; benefit: substantial;
How Effective Are Advance Directives in
Fulfilling the Level of Care Requested
by End-of-Life Lung Cancer Patients?
After Cruzan vs Director, MDH
72
and the Patient
Self Determination Act,
73
most patients are aware
that they have the right to an advance directive.
There is widespread support of these documents
both within the medical community and among lay
individuals
74
; however, the medical and legal
sources, as given below, reveal that both physicians
and patients frequently find them less than ideal, and
therefore they are often rendered ineffective.
Advance directives are legal documents: living
wills and durable powers of attorney for health
care.
75
They acknowledge patient autonomy
74,76
and
attempt to document the patients wishes regarding
medical care when he or she is no longer competent.
Despite all good intentions, neither of these docu-
ments has always achieved its desired goals. With
regards to the durable power of attorney for health
care, the proxies are unable to choose consistently
[sic] treatment options conforming to patient wishes,
. . . [and] they fall short of providing a meaningful
extension of patient autonomy.
35,74,77,78
With regard
to living wills, the words frequently mean different
things to the physician, the patient, and the fami-
ly.
74,79
They only take effect when a patient is
terminalbut when is a patient terminal?
80
The
documents do not itemize all the possible treatment
options and, understandably, from the medical per-
spective, they cannot. Physicians with the best of
intentions occasionally feel the need to have the
assistance of other individuals when interpreting
them.
74
There are studies that show they do not
always affect whether or not a patient has an at-
tempted resuscitation, that the patients do not con-
tinue to receive the expected level of care, and that
they have little impact on the pattern of care when
compared to those patients without an advance
directive.
32,81,82
Even if the medical community
could put these findings aside, there is an overabid-
ing concern and considerable evidence within the
legal community that there are inherent problems
with the statutes themselves, as well as the Patient
Self Determination Act, which can cause them to be
ineffective.
8385
One study within the medical community was
SUPPORT published by Teno and colleagues
32
in
1994. This was a very large cohort study of 3,000
hospitalized patients who were seriously ill. Some of
the patients had executed living wills, and others had
not. The study found that there was no difference
between the groups as to whether or not a DNR
order was placed; each group was as likely to have
an attempted resuscitation.
32
These patients are
described only as seriously ill. Were all terminal? If
not, the conclusion of the study that living wills made
no difference as to the care rendered is inappropri-
ate. We as physicians must understand, and patients
must be educated, to the fact that a living will does
not automatically imply, or give the physician per-
mission to write a DNR order. These are two
separate issues, and they must not be considered
synonymous. First, a living will addresses only the
health-care preferences in the event of terminal
illness and patient incompetence. A seriously ill
incompetent patient may present to you for care, and
have executed a living will that is placed on the chart;
but if the patient is not terminal, it is not appropriate
to state the patient is DNR based on the living will
alone. And second, a DNR order can be written on
behalf of any patient regardless of the state of health;
patients have the right to refuse care.
85
This issue
can only be dealt with by further physician and
patient education. Physician education should be a
required topic in continuing medical education
(needed for licensure in many, if not all states), and
patient education should come from a joint effort
between physicians and attorneys to publish in the
lay press.
A second study was that of Danis et al
81
published
in 1991. The preferences of 175 nursing home
patients regarding their future care was recorded.
The results showed that approximately three fourths
of the patients had care consistent with their wishes,
but even with these patients, there was inconsistency
as to the level of aggressiveness of their care.
Tonelli
74
stated [that] such a finding hints that the
emphasis on advance directives as a way to limit
inappropriately aggressive care may be misguided,
with the real threat to patient autonomy coming
from the side of undertreatment. In keeping with
this finding, it is advisable that physicians be edu-
cated, collectively and repeatedly, by the legal pro-
fession on the interpretation of living wills, not only
on the statutory form in their respective state, but
also the current legal interpretation of various phras-
es; our legal system is not static.
It was the conclusion of Johnson et al
82
from a
1991 and 1992 study, that the presence or absence of
the advance directive (living will and/or durable
power of attorney for health care) had little impact
on the pattern of care. This study looked at 304
patients from a community teaching hospital. A
prospective nonrandomized cohort data collection
and analysis study was undertaken. Review of the
data showed that the patients ages and duration of
stay in ICU were quite similar within each group.
Data also showed that the percentage of deaths was
essentially the same. The authors admit the limita-
tions of the study, but nonetheless believed that
ultimate patient demise was not affected by the
presence or absence of the advance directive. The
study attributes this finding to the fact that in many
of the patients admitted without advance directives,
contact with family or legal representative allowed a
DNR order to be placed on the chart.
82
In addition to the inadequacies mentioned in the
medical literature, the legal sources reveal the fol-
lowing: (1) advance directives are written in such a
manner as to record a patients wishes at the time the
instrument is executed; they presume that a patients
values will not change as the patient grows older or
different circumstances arise
83,84
; (2) the Patient Self
Determination Act was enacted for the purpose of
protecting patient autonomy and self-determination
by informing patients that they have the right to
execute a living will; in the form in which it presently
exists, simply following the letter of the law (fol-
lowing the law precisely, as most hospitals have
done) has reduced this act to a process likened to
reading a person his Miranda Rights; when patients
are informed, very little is done to inform these
patients of how to meaningfully and effectively ex-
ercise this right
84
; (3) in the majority of states that
allow the execution of a living will, the enforcement
of living wills is in serious jeopardy; Specifically, if
the physician refuses to honor the living will, in most
of these states there is no legal recourse which allows
the individual to enforce this right.
85
In light of all of the above, it is imperative that we
look at the original intent of Kutner,
8688
and then
Bok,
89
when the idea of living wills was put forth in
the 1960s and 1970s. When [they] were initially
conceived it was recognized that they can only be
vague guides that must be interpreted by surrogates
at some later time.
90
But who are these surrogates
to be? There are advantages and disadvantages to
naming any particular category of persons, ie, the
physician, the next of kin, or a designee of the court,
the surrogate.
Difficult as it may be, it is a necessity that the
approach of our society to advance directives evolve.
Otherwise, the difficulties that are encountered and
troublesome for all parties will continue. The ad-
vance health-care directive is an idea that has been
proposed and codified in 13 states.
91
These statutes
combine the elements of a living will and durable
power of attorney for health care into one docu-
ment,
91
thus acknowledging patient autonomy, pa-
tient preferences, and the fact that a patients desires
change over time as comorbid conditions arise and
life circumstances change.
Recommendations
6. Each patient with lung cancer should be asked
if he or she has an advance directive, and the
clinician should assume responsibility for plac-
ing it in the chart. If the patient does not have
an advance directive, the clinician should sug-
gest preparing one. Evidence: poor; benefit:
7. With patients for whom there are questions
about the validity or interpretation of an ad-
vance directive, seek guidance from the hospi-
tal legal counsel or ethics committee. Evidence:
poor; benefit: substantial; grade of recommen-
dation: C
What Is the Role of the Hospital Ethics
Committee in Resolving End-of-Life
Problems in Patients With Lung Cancer?
The proliferation of hospital ethics committees
(HECs) was precipitated by public controversy that
burgeoned in the 1970s. Beginning with the case of
Karen Quinlan in 1975, a series of highly publicized
legal cases disclosed numerous unresolved issues
regarding end-of-life care.
92
These cases reflected
the substantial uncertainty of patients, physicians,
families and health-care institutions regarding the
appropriate limits of life-prolonging treatment and
the locus of decision-making responsibilities. At the
same time, the cumbersomeness of legal review, the
adversarial nature of the process, and the inconsis-
tency of the results suggested that the courts were
not an adaptable mechanism for the routine review
of cases regarding end-of-life care. In 1983, the
Presidents Commission for the Study of Ethical
Problems in Medicine and Biomedical and Behav-
ioral Research tentatively endorsed the use of HECs
as a review mechanism that might provide a via
media between purely private decision making and
recourse to the courts when difficult end-of-life
issues emerge in patient care.
93
The development of
HECs was given further impetus in 1992 when the
Joint Commission on Accreditation of Health Care
Organizations required that health-care institutions
establish formal mechanisms for addressing ethical
issues relating to patient care.
94
Although the Joint Commission on Accredita-
tion of Health Care Organizations did not require
the establishment of HECs as the sole means for
providing these mechanisms, HECs were widely
accepted as the most appropriate means for satis-
fying the mandate. At the time of its 1983 report,
a national survey undertaken at the request of the
Presidents Commission for the Study of Ethical
Problems in Medicine and Biomedical and Behav-
ioral Research revealed that only 4.3% of Ameri-
can hospitals with 200 beds had ethics commit-
tees, and 64.7% of facilities with ethics
committees were teaching hospitals. Overall, only
1% of US hospitals had ethics committees, and
none were found in hospitals with 200 beds.
95
A
separate national survey conducted in 2001 illus-
trates the sweeping transformation engendered by
the recommendations of the Presidents Commis-
sion for the Study of Ethical Problems in Medicine
and Biomedical and Behavioral Research and the
standards of the Joint Commission on Accredita-
tion of Health Care Organizations. Of 346 ran-
domly sampled hospitals, 93% now have function-
ing HECs, with a median period of operation of 7
years.
96
These committees engage in several crit-
ical functions within health-care facilities. All
HECs responding to the survey indicated that they
are involved in the development of policy, partic-
ularly as relates to the withholding of life-prolong-
ing treatment, accounting for 30% of committee
time. Education of committee members accounted
for an additional 30% of committee time. Finally,
86% of HECs surveyed reported that they engage
in clinical case consultation, accounting for 20% of
committee time.
The Context of Ethics Consultation
A prominent feature of modern society is the
pervasive pluralism of norms and values cherished by
different individuals. This pluralism is highlighted in
the health-care setting, where individuals interact
who are drawn from widely diverse professional,
cultural, and religious communities. As a result,
uncertainties or conflicts may arise about the inter-
pretation of moral norms or values in clinical care
decisions at the end of life.
Case consultation provided by an HEC may in-
volve individual members, teams, or the committee
as a whole. It is a service that is intended to assist
patients, families, health-care professionals, and
other involved parties in addressing uncertainties or
conflicts regarding moral norms or values affecting
patient care. Numerous issues engendering uncer-
tainty or conflict arise in the context of end-of-life
care. These issues may involve matters such as
informed consent, refusal of treatment, patient de-
cision-making competence, medical futility, inter-
pretation of advance directives, patient confidential-
ity, surrogate preferences, decision making for
patients without family, the distinction between pal-
liation and euthanasia, and appropriate allocation of
resources. The clinical situations engendering these
ethical issues also often involve complex legal, affec-
tive, interpersonal, organizational, and economic
factors.
The Goals of Ethics Consultation
The fundamental goal of ethics consultation is to
improve the quality of patient care by facilitating the
identification, analysis, and resolution of ethical is-
sues that arise in particular clinical cases.
97,98
There
are three ways in which ethics consultants can
facilitate the resolution of moral problems that arise
in providing care to patients with lung cancer.
99
One
is by providing basic information to the parties
consulted regarding relevant statutes, court cases,
institutional policies, and professional guidelines.
For example, state statutes may specify the priority
status of surrogate decision makers when the patient
is no longer competent, or outline procedures for
implementing advance directives. Similarly, institu-
tional policies may formulate specific procedures to
be followed when surrogate decision makers are not
available for incompetent patients. Ethics consult-
ants can apprise the relevant parties about these
substantive and procedural rules that may constrain
the range of choices within particular situations.
The second way in which ethics consultants can
assist in resolving ethical issues involves identifying
and analyzing the uncertainty or conflict of values
and norms that prompted the consultation. One part
of this process involves identifying the relevant moral
principles that apply in the particular case and
clarifying the content of their requirements for
patient care. Another component is assisting the
relevant parties in examining alternative courses of
action and the manner in which these alternatives
may respect or fail to respect the moral principles
identified. Finally, the ethics consultant may suggest
one or more possible solutions to the issue raised
that may provide the fullest consideration for the
moral concerns identified.
The third manner in which ethics consultants can
assist in resolving ethical issues involves facilitating
communication and negotiation among the physi-
cian, patient, and family.
98
Part of this function
involves helping the different parties involved in
clarifying their own values and concerns about the
situation. Another component involves assisting the
parties in communicating their concerns to one
another and in explaining the rationale for the
outcomes they prefer. Finally, the ethics consultant
can assist the parties involved in negotiating and
building a shared commitment with respect to how
the ethical issue raised by the case should be re-
solved.
Inappropriate Roles for Ethics Consultants
The assistance of an ethics consultant is properly
sought when a clinical situation causes conflicts or
uncertainty about the interpretation of moral norms
or values in the care of the patient.
99
The ethics
consultant facilitates the interpersonal process of
resolving these moral conflicts or uncertainties. This
role must be distinguished from several other activ-
ities. One is the function of addressing instances of
immoral or unprofessional conduct on the part of
health professionals. This kind of problem should be
referred to appropriate professional or peer review
groups. Another inappropriate function involves
serving as an advocate for a patient whose interests
or rights have been violated. While part of the
function of the ethics consultant is to ensure that the
patients perspective is properly considered in resolv-
ing end-of-life issues, this does not involve rectifying
prior encroachments on the interests or rights of
patients. The ethics consultant should also not be
regarded as a psychological counselor who assists
persons in understanding and resolving their own
emotional difficulties related to the clinical situation.
While ethical issues in patient care may be stressful
for patients, families, and health professionals, the
ethics consultants role is to assist the involved
parties in clarifying the relevant norms and values
that should guide their decisions. In addition, the
ethics consultant should not be regarded as someone
whose function is to provide the correct answer to
the ethical issues that have been raised. Rather, the
solution to an ethically troubling situation should be
viewed as a course of action to be constructed based
on a shared understanding of the relevant parties
regarding the meaning and interpretation of the
relevant moral norms and values. Finally, the ethics
consultant or HEC should not be assigned decision-
making authority regarding the resolution of a moral
issue in end-of-life care. This function would im-
properly preempt the decision-making authority as-
signed to patients, families, and physicians.
Utilization of HECs for Ethics Consultation
In the most recent national survey of HECs, the
275 committees performing case consultations
reported an average of 8.1 formal consultations and
4.3 informal consultations per year.
96
Patient auton-
omy and decision-making competence were ad-
dressed in 38% of the cases, improving communica-
tions in 35%, end-of-life issues in 7%, use of new
technologies or research in 7%, and cost-contain-
ment issues in 3%. In another study of a busy
consultation service in a teaching hospital, 51 con-
sultation requests over 12 months involved issues
about withholding life-sustaining treatment (49%),
DNR orders (37%), and clarification of legal require-
ments (31%).
100
A third study reviewing 46 case
consultations indicated that major issues raised for
review included withdrawing or withholding treat-
ment (38 cases), appropriateness of current treat-
ment (28 cases), resuscitation (22 cases), competency
or refusal of treatment (12 cases), and family de-
mands for inappropriate treatment (8 cases).
101
In
another study reviewing 104 case consultations in a
community hospital, consultation requests con-
cerned decisions to forego life-sustaining treatment
(74%), resolution of disagreements (46%), and de-
termination of patient competence (30%).
102
Thus,
available studies clearly suggest that the consultation
services of HECs are predominantly utilized in
addressing issues related to patient autonomy and
competence, withholding or withdrawal of life-
prolonging treatment, communication issues, and
surrogate preferences. Because these categories re-
flect the most prominent ethical issues posed in
end-of-life care for patients with lung cancer, ethics
consultation services may have particular relevance
in these settings.
The Effectiveness of Ethics Consultation
Assessment of the outcomes of ethics consultation
in end-of-life decision making is limited. Four po-
tential benchmarks for evaluation have been identi-
fied: (1) case management that conforms to ethical
norms and standards, (2) satisfaction of involved
parties regarding the results of consultation,
(3) resolution of initial conflict between the parties
involved, and (4) the educational impact of the ethics
consultation.
103
Evaluation of whether ethics consultation results
in management decisions that conform to ethical
standards requires that there is widely shared agree-
ments about the content of such standards. In some
aspects of end-of-life care, consensus standards have
emerged. For example, the major components of an
adequately informed consent have been well formu-
lated.
104
The right of competent adult patients to
refuse treatment, including life-prolonging proce-
dures, is well established.
92
The role of advance
directives in determining care for incompetent pa-
tients has been widely recognized in state law.
105
Criteria for decision making by surrogates have been
formulated.
106
However, despite the consensus in
many areas related to end-of-life care, there have
been no studies to date that assess whether ethics
consultation improves the extent to which end-of-life
care decisions satisfy ethical norms of professional
practice.
Several studies have examined the level of satis-
faction among parties who have requested or re-
ceived the assistance of ethics consultants. A review
of 46 case consultations found that 90% of attend-
ing physicians found consultation to be helpful in
clarifying ethical issues, educating the health-care
team, increasing the confidence about clinical deci-
sions, and improving patient management.
101
In
another evaluation study of 51 ethics consultations,
physicians considered the consultation very impor-
tant in 71% of the cases in clarifying the ethical
issues, learning about medical ethics, or in determin-
ing the management of the patient.
100
Finally, in a
2-year study of 104 ethics consultations, the request-
ing physician found the consultation helpful or
very helpful in 86% of the cases.
102
Thus, ethics
consultation appears to result in a high level of
satisfaction among parties either requesting or par-
ticipating in the process.
No studies have examined in detail the impact of
ethics consultations in resolving conflicts among the
partiesphysician, patient, and familyinvolved in
clinical care decisions. However, in one review of
16 case consultations, only 2 of 40 staff members and
2 of 6 family members indicated that they disagreed
with the consultants recommendations.
107
In an-
other study, physicians rated the consult as helpful in
75% of cases in which communication with the
family or other staff was a problem.
108
The educational impact of ethics consultation has
not been directly examined by evaluating the knowl-
edge of relevant factors before and after consulta-
tion. However, in three studies, the participants
reported that the consultation was helpful in identi-
fying ethical issues, clarifying relevant concepts and
norms, and in providing practical knowledge useful
in clinical case management.
100102
The Process of Ethics Consultation
Given the relative newness of the ethics consulta-
tion as a function of HECs, there are many variations
in the process utilized.
109
One area of variable
practice relates to access to ethics consultation.
Some protocols permit only physicians to request
ethics consultations, while others allow requests
from other professional staff members, as well as
from patients and families. Another variable relates
to process of case review. In some settings, the case
presentation and deliberative process include only
the attending physician and the consultants, while in
other settings professional staff, patient, and family
may be included in the process of presenting and
deliberating about the resolution of the case. A third
variable relates to the results that the consultation is
intended to achieve. These might involve only a full
discussion of the clinical care options, identification
of a range of acceptable alternatives, or the recom-
mendation of a preferred option. Another variable
relates to the form in which the results of consulta-
tion should be disclosed. Disclosure may be in the
form of a verbal or written report, including a
consultation summary inserted in the medical
record. Finally, there are alternatives regarding the
parties to whom the consultation results should be
disclosed, including the attending physician, relevant
staff members, patients and families. In light of
multiple choices available for fashioning the process
of ethics consultation, it is important that the re-
questing party and consultants clarify beforehand
who will be involved, what the objective of the
consultative process should be, how the results will
be formulated, and to whom the results will be
conveyed.
Recommendations
8. In making end-of-life decisions for patients
with lung cancer, ethics consultation by HECs
should be requested when assistance is needed
in clarifying applicable law and policy related to
patient autonomy and competence, informed
consent, withholding life-prolonging treat-
ments, surrogate preferences, decision making
for patients without family, and resource allo-
cation, as well as determining how ethical
norms should be interpreted, or negotiating
interpersonal conflicts among patients, fami-
lies, and physicians. Evidence: poor; benefit:
9. In end-of-life care for patients with lung can-
cer, given the potential variations in ethics
consultations, the requesting party and the
consultant should clarify beforehand the spe-
cific objectives of the consultation, the selec-
tion of the participants, the process to be used
in deliberation or negotiation, and the manner
in which results will be disclosed and recorded.
Evidence: poor; benefit: substantial; grade of
recommendation: C
What Is the Role of the Critical Care
Specialist in the End-of-Life Care of Lung
Cancer Patients?
Spiraling costs and reform of health care have
intensified pressure to reduce use of expensive re-
sources, such as critical care services, for patients
with poor prognoses, including those with advance
lung cancer. This is understandable, considering that
outcomes of ICU treatment for cancer patients are
often unfavorable and that intensive care consumes a
very large and disproportionate share of health-
related expenditures. Potential distress for patients
and families from invasive interventions in the criti-
cal care setting is another important cause for con-
cern. It is evident from existing literature and clinical
experience, however, that thousands of patients with
lung cancer and other malignancies are still admitted
to ICUs each year, with reported benefit for some
patients. In addition, despite an improved ability to
predict ICU outcome using cancer-specific models,
certainty remains elusive and, for many patients, only
a therapeutic trial can determine the effectiveness of
intensive care with an acceptable degree of confi-
dence. It is therefore reasonable to expect that
patients with lung cancer will continue to receive
care in ICUs, including end-of-life care, and it is
appropriate to review the role of the critical care
specialist in such care.
Lung Cancer in Critical Care
More than 5 million patients are admitted to ICUs
in the United States each year,
110
including a sub-
group with lung cancer that is difficult to quantify
with precision from existing data. A large-scale sur-
vey in 1993,
111
revealed that cancer-related organ
failure was the admitting diagnosis for approximately
1% of admissions to the nations ICUs. Among
cancer patients included in recent studies of ICU
outcomes, lung cancer was the most common solid
tumor and among the most common malignancies
overall, accounting for between 20% and 30% of
solid tumors in critically ill patients,
25,112
and 16% of
all malignancies in cancer patients admitted to a
medical ICUs.
113
Indications for ICU admission of
patients with lung cancer are diverse, ranging from
perioperative care of patients undergoing resectional
surgery with curative intent to respiratory failure
from exacerbation of underlying chronic lung dis-
ease, intercurrent compromise by infection, hemor-
rhage, or pleural effusion, or advancing malignancy
with bronchial obstruction. Appropriately, most cli-
nicians, including intensivists, wish to maximize
treatment of reversible, acute illness even in the
context of ongoing, underlying disease, while limiting
ICU care for patients with irreversible complica-
tions. The distinction may be difficult to draw,
however, at the time point of presentation to an ICU.
Outcomes of ICU Treatment for Cancer Patients
Decision making about ICU treatment for the
patient with lung cancer should be informed by
available knowledge about prognosis. Although no
mortality prediction model has been developed pro-
spectively and specifically for lung cancer patients
with critical illness, a large, multicenter, prospective
study by Groeger et al
114
derived and validated a
multivariable logistic regression model to estimate
the probability of hospital mortality among adult
cancer patients admitted to the ICU. This cancer-
specific model incorporates a manageable number of
readily available clinical variables and is better cali-
brated than general ICU prognostic scoring systems,
which underestimate the risk of mortality for cancer
patients.
111,115
A separate analysis of prospectively
collected data was subsequently performed,
112
to
identify predictors of hospital mortality for cancer
patients requiring mechanical ventilation, which re-
quires prospective and independent validation in the
future. Overall, the observed hospital mortality for
cancer patients receiving ICU treatment was 42%,
but 76% for patients requiring mechanical ventila-
tion. Among the latter group, progression or recur-
rence of cancer, cardiac arrhythmias, need for vaso-
pressor therapy, and presence of disseminated
intravascular coagulation, were among seven vari-
ables associated with an increased risk of death,
whereas prior surgery with curative intent was pro-
tective.
112
These models, like other scoring systems,
should not be used in isolation to predict outcome or
foreclose ICU treatment for individual patients, but
are useful to reduce uncertainty and improve repro-
ducibility in clinical decision making.
116
A retrospective study
117
documented hospital
mortality of 75% among 57 patients with primary
lung cancer who were admitted to a medical ICUs,
identifying acute pulmonary disease (such as infec-
tion or ARDS) and Karnofsky performance status
prior to hospital admission as factors predictive of
ICU and hospital death; among hospital survivors
(n 14), median postdischarge survival was
32 weeks for patients with stage I or II disease, and
16 weeks for stage III or IV. Among 44 lung cancer
patients included in a retrospective study of critically
ill cancer patients,
113
approximately one half of
whom received mechanical ventilation, hospital mor-
tality was 48%. An earlier retrospective study,
118
involving lung cancer patients without prior surgical
resection who required mechanical ventilation for
respiratory failure, found that 39 of 46 patients died
receiving mechanical ventilation and 10% sur-
vived the hospitalization; no patient was liberated
from mechanical ventilation after 6 days, a finding
consistent with other data associating prolonged
mechanical ventilation with dismal outcomes for
cancer patients.
119
At the present time, no data exist
with respect to functional status or quality of life of
patients with lung cancer surviving ICU treatment.
Burdens of ICU Treatment
Potential benefits of critical care for the patient
with lung cancer must be weighed against burdens
that may be associated with such treatment. Of
particular concern is emerging evidence of physical
and psychological suffering among ICU patients,
including patients with lung cancer and other malig-
nancies. A prospective study of symptom experience
self-reported in real-time by cancer patients treated
in a medical ICUs,
120
including patients and nonsur-
vivors receiving mechanical ventilation, revealed that
distressing symptoms such as pain, discomfort, dys-
pnea, depression, and anxiety were prevalent, often
at high levels of severity. In the large-scale
SUPPORT
121
for seriously ill patients, including a
subgroup in ICUs, 35% and 46% of lung cancer
patients dying within 2 months reported severe pain
and dyspnea, respectively, for at least half of the time
during the observed hospitalization. Symptom bur-
den is relevant in ICU decision making, although
expert palliative care can be expected to improve
patient and family comfort and should be integrated
into treatment plans for all critically ill patients
especially those, such as patients with lung cancer,
who remain at high risk for hospital death. Financial
and other burdens for families of such patients may
be significant.
For the health-care system as a whole, the costs of
treating cancer patients in ICUs are considerable.
ICU patients account for almost 20% of the average
hospitals operating budget, but only 5 to 6% of total
patient days and, among patients consuming the
most expensive ICU resources, mortality rates are
particularly high.
119,122
A 1993 analysis,
119
based on
hospital charges for patients treated in the ICUs of a
university-affiliated, tertiary care cancer center esti-
mated as $150,000 the cost per year of life gained
at home for those with solid tumors surviving 3
months, and as $1.1 million for hospital nonsurvi-
vors, who represented 41% of the solid tumor
group. Data such as these may influence societal
determination of the net value of some or all ICU
interventions in the treatment of patients with lung
cancer and other conditions associated with high risk
of hospital morbidity and mortality. Clinical decision
making, however, is appropriately based on a balanc-
ing of potential medical benefits and burdens for the
individual patient.
During critical illness, most patients lack decision-
making capacity,
123
yet their preferences for life-
supporting treatment and other components of crit-
ical care are poorly predicted by surrogates,
including close relatives and long-standing primary
doctors.
77,124
Unfortunately, communication dur-
ing
125
critical illness is often inadequate and the
preferences of most patients are not known, al-
though data indicate that patients are generally
willing to discuss preferences, prognosis, and care
goals, if the opportunity is provided.
126
A large-scale
observational study,
127
conducted in 15 ICUs in four
countries recently demonstrated that explicit ad-
vance directives continue to be uncommon during
the first 24 h in the ICU. Among patients with
explicit directives, the directive was DNR for 50%,
whereas the default directive implied for all pa-
tients who had not explicitly stated their preference
was to perform CPR. Even explicit directives are
often contingent on a prognosis that may still be
uncertain at the onset of critical illness, and other
limitations of such directives have also been identi-
fied.
81,128
To the extent that critical illness can be
anticipated, early discussion with the patients with
lung cancer about likely outcomes and burdens of
ICU treatment will help to promote concordance of
care goals and plan with the patients own prefer-
ences.
Provision of Palliative Care by the Critical Care
Specialist
For patients with lung cancer treated in ICUs, as
for other patients at substantial risk of hospital death,
it is not appropriate to postpone palliative care until
death is imminent. Death may come suddenly and
unexpectedly, as it did for many patients with cancer
and other serious illnesses in SUPPORT, whose
median predicted 2-month survival was 20% on the
day before they died.
129
In addition, accumulating
evidence from critical care and other settings sug-
gests that patient suffering is associated with unfa-
vorable outcomes including higher mortality,
whereas relief of distressing symptoms and improved
communication about treatment goals may promote
favorable clinical and utilization outcomes. Consen-
sus is therefore growing in support of an integrated
approach, combining palliative care with critical care
for all ICU patients, including those still pursuing
life-prolonging therapies as well as those more obvi-
ously at the end of life.
130
In some institutions,
clinicians with expertise in palliative medicine, spe-
cific areas of physical or psychological symptom
management, and/or ethics may be available to assist
those with primary responsibility for critically ill
patients, but the critical care specialist should have
fundamental knowledge and skills in management of
comfort and communication needs of ICU patients
and their families. Randomized controlled trials of
palliative interventions for critically ill patients are a
research priority, but these will require further
development of appropriate instruments and out-
come measures,
131
and may always be limited by
practical and ethical constraints. At the present time,
practice recommendations are guided by evidence
from observational studies, nonrandomized or un-
controlled interventional trials, qualitative research,
and expert opinion.
Recommendations
10. For the patient with lung cancer, decision
making about ICU treatment should incorpo-
rate available knowledge about prognosis, in-
cluding the use of a cancer-specific outcome
prediction model to complement clinical judg-
ment, and weigh reasonably expected benefits
of critical care against potential burdens, in-
cluding distressing physical and psychological
symptoms. Evidence: poor; benefit: substan-
11. In the inoperable or unresectable patient with
lung cancer, prolonged mechanical ventilation
is discouraged in view of dismal reported
outcomes. Evidence: fair; benefit: small; grade
of recommendation: D
12. In the critically ill patient with lung cancer,
palliative care, including expert management
of symptoms and effective communication
about appropriate goals of treatment, should
not be postponed until death is imminent, but
should be an integral component of the diag-
nostic and treatment plan for all patients,
including those still pursuing life-prolonging
therapies as well as those more obviously at
the end of life. Evidence: poor; benefit: sub-
What Is the Frequency of Use and
Satisfaction With Hospice Environment in
the End-of-Life Care of Patients With
Lung Cancer?
Lung cancer is the most common cause of cancer
deaths in both men and women, with only 15% of all
patients living for 5 years.
132
Since the majority of
patients with lung cancer will die of their disease, it
is imperative that end-of-life care be incorporated
into their routine management. The goal of this
palliative end-of-life care should be to achieve the
best possible quality of life for patients and their
families. In 1998, the American Society of Clinical
Oncology Task Force on Cancer Care at the End of
Life developed a statement strongly supporting the
need to provide excellent end-of-life care for patients
with cancer.
133
The task force emphasized a humane
system of cancer care and identified several princi-
ples for dealing with end-of-life care. They recom-
mended that cancer care should be centered on a
long-standing and continuous relationship with the
primary care provider and the patient. Cancer care
should be responsive to the patients wishes, and
should be based on truthful, sensitive, empathic
communication with the patient while optimizing
quality of life throughout the course of an illness
through meticulous attention to the myriad physical,
spiritual, and psychosocial needs of the patient and
family.
Patients often fear a lonely, painful death with
technologies that are out of their control and only
delay the natural process of dying. Cancer care
should be a longitudinal involvement from diagnosis,
treatment of the cancer and its symptoms, managing
recurrences, and end-of-life supportive care. The
physician must recognize a turning point in the
patients condition when anticancer treatments fail
to work and physical and emotional support become
the primary mode of treatment.
The physician should be aware that symptoms of
weakness, pain, fatigue, and nausea and vomiting are
continuing clinical problems throughout the course
of a terminal illness.
134,135
In a study evaluating
end-of-life care in 939 patients in five teaching
hospitals between 1989 and 1994, severe dyspnea
occurred in 32% of patients and severe pain in 28%
of patients.
121,134
These and other symptoms are the
focus of end-of-life care in the patient with lung
cancer.
Dyspnea can become an extremely important
challenge at the end of life in patients with lung
cancer. The uncomfortable sensation of labored
breath is a terrifying experience and often increases
near the end of life. Careful evaluation of the cause
of dyspnea can often lead to temporary control of
this symptom. Treatment of pulmonary infection,
chronic lung disease, cardiac failure, and arrhythmias
may improve the patients comfort. Obstructive pul-
monary lesions with atelectasis may be treated with
short courses of radiation therapy.
136
Malignant ef-
fusions can often be treated with thoracentesis.
Parenteral opioids can be used for the treatment of
dyspnea without causing significant deterioration of
pulmonary function.
57,137
Pain is often a major symptom of patients with
lung cancer, and the treatment of pain by physicians
is often inadequate. Pain increases many physical
symptoms such as fatigue, sleeplessness, constipa-
tion, and nausea.
138
Pain can lead to a sense of loss of
control and diminished usefulness. It also can cause
distress in the family causing physical and financial
burdens. Also the patient sees pain as a sign that the
disease is worsening. Several excellent reviews have
been written on the management of pain in cancer
patients.
69,139,140
Several important considerations
about pain control should be emphasized when
managing terminal patients with lung cancer. With
careful evaluation and management, pain should be
controlled in 85 to 95% of cases.
141
The WHO has
developed a useful treatment ladder for starting
treatment with escalation of drug doses as needed.
142
Oral morphine should be the standard opioid ther-
apy, using around the clock analgesia with break-
through doses when needed using whatever dose
controls the pain.
A very common complication of lung cancer is
nausea and vomiting. Prevalence of this symptom is
40 to 46% in the last 6 weeks of life
143
Causes for
nausea and vomiting should be worked up and
treated aggressively. Fear, anxiety, and pain can
cause nausea and vomiting, and should be treated.
Gastric irritation should be treated with acid-
blocking agents. Overfeeding should be avoided
because it can cause nausea and vomiting, and
hepatic metastases can be treated with nonsteroidal
anti-inflammatory drugs and corticosteroids with cy-
toprotectants. Other symptoms that may occur in
lung cancer such as weight loss, insomnia, and
anorexia may become irrelevant at the end of life,
but troublesome oral and respiratory secretions,
myoclonus, and drowsiness may be important end-
of-life symptoms that require careful management.
Hospice is the best model for end-of-life care in
the United States. This approach to end-of-life care
attempts to treat the patient and family at home.
Under the Tax Equity and Fiscal Responsibility Act
of August 1982,
144
terminally ill patients 65 years
old became eligible for hospice services under the
Hospice Benefit of the Medicare program. In order
to qualify, one had to be terminally ill with 6
months to live. Medicaid covers hospice care in 43
states and the District of Columbia. Unfortunately,
this form of care is often underutilized or utilized too
late to be helpful to the patient. A recent study
145
evaluating 6,457 patients enrolled in hospice pro-
grams and followed up for a minimum of 27 months
found that the most common diagnosis was lung
cancer, making up 21% of all patients. Patients with
lung cancer had a median survival of 36 days with
15% dying within 7 days. These data suggest that
patients with lung cancer may be admitted to hospice
programs too late in their disease to fully benefit
from hospice care. It has been shown that the
advantage of hospice care is that it can be delivered
in the patients home and the patient can die at
home,
146
it can optimize pain relief,
147
it increases
patient satisfaction,
148
and some studies suggest it
may be more cost-effective than hospital care.
149
Hospice provides nursing care, physician services,
medical appliances, drugs, short-term hospitaliza-
tions, services of homemakers, home health aids,
physical occupational and speech therapy, psycho-
logical counseling, and social services. Eighty per-
cent of hospice services are provided in the home.
There are a number of clinical barriers that exist
that interfere with this optimal level of end-of-life
care. Physicians are reluctant to talk about death at
the end of life, and they may consider progression of
disease to be a therapeutic failure. Fear of opioid
addiction by the physician and patient may lead to
underreporting of pain and undertreatment with
narcotics. Further barriers to optimal care include
the lack of available multimodality palliative care
teams, fragmentation of care by multiple physicians,
and no designated team leader directing the overall
care. It is important for the primary care provider to
be well versed in palliative care programs incorpo-
rating pharmacists, psychologists, nurses, social
workers, pastoral care providers, pain specialists, and
ethicists.
There are multitudes of educational deficiencies
that limit the optimal delivery of palliative care in the
patient with terminal cancer. The American Medical
Association Graduate Medical Education Report for
1993 noted that only 26% of US residency programs
offered a formal course in end-of life care.
150
Most
physicians do not receive adequate education in
end-of-life care in medical schools.
151
Physicians are
left learning on the job and often lack role models
with expertise in caring for the terminally ill. Courses
during clinical clerkship, residency, and fellowship
should provide didactic and practical experience in
palliative care with a team approach.
Comprehensive research into the multiple aspects
of palliative care has received little funding and is
completely lacking in many areas. Pain management
has been the subject of extensive research, but
cachexia, asthenia, and chronic nausea have received
little attention. Other areas receiving limited inves-
tigation have been anxiety, depression, and suicide.
Studies have shown that spirituality has a positive
impact on the quality of life of patients with terminal
cancer.
152
Recommendations
13. For patients with lung cancer at the end of
life, the goal of palliative care should be to
achieve the best quality of life for the patients
and their families. Evidence: poor; benefit:
14. In patients with lung cancer receiving hospice
care, end-of-life management needs to be
considered part of the longitudinal care of
these patients. Evidence: fair; benefit: Sub-
stantial; grade of recommendation: B
15. At the end of life in patients with lung cancer,
multimodality palliative care teams should be
developed and encouraged to participate in
their management. Evidence: fair; benefit:
Conclusion
No American clinical practice guideline is cur-
rently available that is specifically for end-of-life care
for patients with bronchogenic carcinoma. Since the
majority of those patients with an established diag-
nosis of non-small-cell lung cancer will die within 1
year, it is imperative that, with the possible exception
of approximately 20% who might benefit from a
complete surgical excision, extensive communication
about the entire expected course of their disease
occur between patients, their physicians, and their
families at the earliest opportunity. It is unfortunate
that in its present form and application, the advance
care directive has been difficult to utilize effectively
in the terminal care of patients with lung cancer. The
HEC, with a multidisciplinary team quite capable of
assisting in the resolution of many problems con-
cerning these patients and their families, is rarely
consulted in clinical practice. Although use of a
critical care specialist is usually considered too ex-
pensive and likely futile, there are nevertheless many
circumstances in which a concurrent and possibly
reversible condition might be successfully treated in
this setting. The growing popularity of the hospice
movement is largely due to its proven effectiveness
in the more satisfactory control of troublesome
symptoms in the patient with far-advanced lung
cancer, as well as its ability to provide a return to the
home environment with participation of family and
friends in end-of-life care.
life, it is recommended that clinicians increase
their focus on the patients experience of
illness to improve congruence of treatment
with patient goals and preferences: (a) be
realistic, practical, sensitive, and compassion-
ate; (b) listen; (c) allow/invite the patient to
express his or her reaction to the situation;
(d) provide a contact person; (e) and contin-
ually reassess the patients goals of therapy as
part of treatment planning. Evidence: poor;
tion: C
2. For all patients with lung cancer, end-of-life
planning should be integrated as a component
of assessment of goals of treatment and treat-
ment planning. Evidence: poor; benefit: sub-
3. For patients with lung cancer, an experienced
clinician should inform the patient of the
diagnosis and its meaning. The day-to-day
contact person should also be present at this
meeting and should coordinate care. Evi-
dence: poor; benefit: substantial; grade of
recommendation: C
4. Clinicians treating patients with lung cancer
should avail themselves of the increasing body
of educational resources to improve commu-
nication at the end of life. Evidence: fair;
tion: B
5. With patients with lung cancer, hospice and/or
the palliative care service should be involved
early in the patients treatment, as part of the
team. Evidence: poor; benefit: substantial;
6. Each patient with lung cancer should be asked
if he or she has an advance directive, and the
clinician should assume responsibility for plac-
ing it in the chart. Evidence: poor; benefit:
7. With patients for whom there are questions
about the validity or interpretation of an ad-
vance directive, seek guidance from the hos-
pital legal counsel or ethics committee. Evi-
dence: poor; benefit: substantial; grade of
recommendation: C
8. In making end-of-life decisions for patients with
lung cancer, ethics consultation by HECs should
be requested when assistance is needed in clar-
ifying applicable law and policy related to pa-
tient autonomy and competence, informed con-
sent, withholding life-prolonging treatments,
surrogate preferences, decision making for pa-
tients without family, and resource allocation, as
well as determining how ethical norms should
be interpreted, or negotiating interpersonal con-
flicts among patients, families, and physicians.
Evidence: poor; benefit: substantial; grade of
recommendation: C
9. In end-of-life care for patients with lung
cancer, given the potential variations in ethics
consultations, the requesting party and the
consultant should clarify beforehand the spe-
cific objectives of the consultation, the selec-
tion of the participants, the process to be used
in deliberation or negotiation, and the manner
in which results will be disclosed and re-
corded. Evidence: poor; benefit: substantial;
10. For the patient with lung cancer, decision
making about ICU treatment should incorpo-
rate available knowledge about prognosis, in-
cluding the use of a cancer-specific outcome
prediction model to complement clinical judg-
ment, and weigh reasonably expected benefits
of critical care against potential burdens, in-
cluding distressing physical and psychological
symptoms. Evidence: poor; benefit: substan-
11. In the inoperable or unresectable patient with
lung cancer, prolonged mechanical ventilation
is discouraged in view of dismal reported
outcomes. Evidence: fair; benefit: small; grade
12. In the critically ill patient with lung cancer,
palliative care, including expert management
of symptoms and effective communication
about appropriate goals of treatment, should
not be postponed until death is imminent, but
should be an integral component of the diag-
nostic and treatment plan for all patients,
including those still pursuing life-prolonging
therapies as well as those more obviously at
the end of life. Evidence: poor; benefit: sub-
life, the goal of palliative care should be to
achieve the best quality of life for the patients
and their families. Evidence: poor; benefit:
Substantial; grade of recommendation: C
14. In patients with lung cancer receiving hospice
care, end-of-life management needs to be
considered part of the longitudinal care of
these patients. Evidence: fair; benefit: sub-
stantial; grade of recommendation: B
15. At the end of life in patients with lung cancer,
multimodality palliative care teams should be
developed and encouraged to participate in
patient management. Evidence: fair; benefit:
References
1 Holland JC, Chertkov L. Clinical practice guidelines for
the management of psychosocial and physical symptoms of
cancer: improving palliative care for cancer; summary and
recommendations: National Academy Press, 2001; 199232
2 Carver AC, Foley KM. Palliative care. In: Holland JC, Frei
III E, Bast Jr. RC, et al, eds. Cancer medicine. Hamilton,
Ontario, Canada: BC Decker, 2000; 9921000
3 American Society Clinical Oncology. Outcomes of cancer
treatment for technology assessment and cancer treatment
guidelines. J Clin Oncol 1996; 14:671679
4 Jacox A, Carr DB, Payne R, et al. Management of cancer
pain: clinical practice guideline. Rockville, MD: US Depart-
ment of Health and Human Services, 1994; AHCPR Publi-
cation No. 940592
5 Hewitt M, Simone JV, eds. Ensuring quality cancer care.
6 Pellagrino ED. Ethical issues in palliative care. In: Chochi-
nov HM, Breithart W, eds. Handbook of psychiatry in
palliative medicine. New York, NY: Oxford University Press,
2000; 337356
7 World Health Organization. Symptom relief in terminal
illness. Geneva, Switzerland: World Health Organization,
1998
8 Girgis A, Sanson-Fisher RW. Breaking bad news: consensus
guidelines for medical practitioners. J Clin Oncol 1995;
13:24492456
9 Holland JC, Almanza J. Giving bad news: is there a kinder,
gentler way? Cancer 1999; 86:738740
10 Speice J, Harkness J, Laneri R, et al. Involving family
members in cancer care: focus group considerations of
patients and oncological providers. Psychooncology 2000;
9:101112
11 Chochinov HM, Holland JC, Katz IY. Bereavement: a
special issue in oncology. Psychooncology 1998; 7:1016
1032
12 Zisook S. Understanding and managing bereavement in
palliative care. In: Chochinov HM, Breithart W, eds. Hand-
book of psychiatry in palliative medicine. New York, NY:
Oxford University Press, 2000; 331334
13 Jones JS. Answering the relatives of lung cancer patients.
Br J Dis Chest 1984; 78:388391
14 American Pain Society Quality Care Committee. Quality
improvement guidelines for the treatment of acute pain and
cancer pain. JAMA 1995; 274:18741880
15 Carr JA, Payne R. Management of cancer pain: adults quick
reference Guide No. 9. Rockville, MD: Agency for Health
Care Policy, US Department of Health and Human Ser-
vices, 1994
16 McGivney WT. The National Comprehensive Cancer Net-
work: present and future directions. Cancer 1998; 82:(10
Suppl):20572060
17 SUPPORT principal investigators. A controlled trial to
improve care for seriously ill hospitalized patients. JAMA
1995; 274:15911598
18 Rischer JB, Childress JB. Cancer pain management: pilot
implementation of the AHCPR guideline in Utah. J Qual
Improvement 1996; 22:683700
19 British Thoracic Society. Recommendations for lung cancer
management. Thorax 1998; 53:S1S8
20 Management of lung cancer: national clinical guideline for
use in Scotland. Edinburgh, Scotland: Scottish Intercolle-
giate Guidelines Network, 1998; 154
21 National Health Service Executive Committee. Improving
outcomes in lung cancer: guidance on commissioning cancer
services. London, UK: Department of Health, 1998; 173
22 Holloran S, Starkey G, Burke P, et al. An educational
intervention in the surgical intensive care unit to improve
clinical decisions. Surgery 1995; 118:294299
23 Lamont E, Christakis N. Prognostic disclosure to patients
with cancer near the end of life. Ann Intern Med 2001;
134:11961205
24 Smith T. Tell it like it is. J Clin Oncol 2000; 18:34413445
25 Azoulay E, Chevret S, Leleu G, et al. Half the families of
intensive care unit patients experience inadequate commu-
nication with physicians. Crit Care Med 2000; 8:30443049
26 Silveira MJ, DiPiero A, Gerrity MS, et al. Patients knowl-
edge of options at the end of life: ignorance in the face of
death. JAMA 2000; 284:24832488
27 Steinhauser K, Clipp E, McNeilly M, et al. In search of a
good death: observations of patients, families and providers.
Ann Intern Med 2000; 132:825832
28 Ford S, Fallowfield L, Lewis S. Doctor-patient interactions
in oncology. Soc Sci Med 1996; 11:15111519
29 Maguire P. Improving communication with cancer patients.
Eur J Cancer 1999; 35:14151422
30 Alpert E. Comparing utilization of life-sustaining treatments
with patient and public preferences. J Gen Intern Med
1998; 13:175181
31 SUPPORT. Study to understand prognoses and preferences
for risks of treatment: study design. J Clin Epidemiol 1990;
43(suppl):1S123S
32 Teno JM, Lynn J, Phillips RS, et al. Do formal directives
affect resuscitation decisions and the use of resources for
seriously ill patients? J Clin Ethics 1994; 5:2330
33 Teno J, Hakim RB, Knaus WA, et al. Preferences for
cardiopulmonary resuscitation: physician-patient agreement
and hospital resource use. J Gen Intern Med 1995; 10:179
186
34 Koch KA. Lessons from the SUPPORT study. Clin Pulm
Med 1997; 4:159165
35 Seckler AB, Meier D, Mulvihili M, et al. Substituted
judgment: how accurate are proxy predictions? Ann Intern
Med 1991; 115:9298
36 Tomlinson T, Howe K, Notman M, et al. Empirical study of
proxy consent for elderly persons. Gerontologist 1990; 30:
5464
37 Zeiebel NR, Cassel CK. Treatment choices at end of life:
comparison of decisions by older persons and their physi-
cian-selected proxies. Gerontologist 1989; 29:615621
38 Ouslander JG, Jymchuk AJ, Rahbar B. Health care decisions
among elderly long term residents and their potential
proxies. Arch Intern Med 1989; 149:13671372
39 Uhlmann RF, Pearlman RA, Cain KC. Physicians and
spouses predictions of elderly patients resuscitation prefer-
ences. J Gerontol 1988; 43:115121
40 Everhart MA, Pearlman RA. Stability of patient preferences
regarding life-sustaining treatments. Chest 1990; 97:159
164
41 Callahan D. The troubled dream of life. New York, NY:
Simon and Schuster, 1993
42 Mularski RA, Bascom P, Osborne ML. Educational agendas
for interdisciplinary end-of-life curricula. Crit Care Med
2001; 29(suppl):N16N23
43 Curtis J, Rubenfeld G, eds. Managing death in the ICU: the
transition from cure to comfort. New York, NY: Oxford
University Press, 2001
44 Danis M, Federman D, Fins J. Incorporating palliative care
into critical care education: principles, challenges, and op-
portunities. Crit Care Med 1999; 1999:20052013
45 Steel K, Vitale C, Whang P. Annotated bibliography of
palliative care and end-of-life care. J Am Geriatr Soc 2000;
48:325332
46 Roter D, Hall J, Kern D, et al. Improving physicians
interviewing skills and reducing patients emotional distress:
a randomized clinical trial. Arch Intern Med 1995; 155:
18771884
47 Gleeson K, Wise S. The do not resuscitate order: still too
little too late. Arch Intern Med 1990; 150:10571060
48 Spiro SG. End-of-life care for patients with lung cancer,
personal communication. London, UK: University College
London Hospitals, 2002
49 Larson D, Tobin D. End-of-life conversations: evolving
practice theory. JAMA 2000; 284:15731578
50 Jonsen AR, Siegler M, Winslade WJ. Clinical ethics. 3rd ed.
New York: McGraw Hill, 1992
51 Fogarty L, Curbow B, Wingard J, et al. Can 40 seconds of
compassion reduce patient anxiety? J Clin Oncol 1999;
17:371379
52 Lo B, Quill T, Tulsky J. Discussing palliative care with
patients. Ann Intern Med 1999; 130:744749
53 Schroen AT, Detterbeck FC, Crawford R, et al. Beliefs
among pulmonologists and thoracic surgeons in the thera-
peutic approach to non-small cell lung cancer. Chest 2000;
118:129137
54 The A, Hak T, Koeter G, et al. Collusion in doctor-patient
communications about imminent death: an ethnographic
study. BMJ 2000; 321:13761381
55 Knaus WA, Harrell FE, Lynn J. The SUPPORT prognostic
model: objective estimates of survival for seriously ill hospi-
talized patients. Ann Intern Med 1995; 122:191203
56 Poses RM, McClish DK, Bekes C, et al. Ego bias, reverse
ego bias, and physicians prognostics. Crit Care Med 1991;
19:15331539
57 Allard P, Lamontagne C, Tremblay C. How effective are
supplementary doses of opioids for dyspnea in terminally ill
cancer patients? A randomized continuous sequential clini-
cal trial. J Pain Symptom Manage 1999; 117:256265
58 Quill TE, Brody H. Physician recommendations and patient
autonomy: finding a balance between patient power and
patient choice. Ann Intern Med 1996; 125:763769
59 Baile W, Glober G, Lenzi R, et al. Discussing disease
progression and end-of-life decisions. Oncology 1999; 13:
10211035
60 Steinhauser KE, Christakis NA, Clipp EC, et al. Factors
considered important at the end of life by patients, family,
physicians, and other care providers. JAMA 2000; 284:2476
2482
61 Finucane T. How gravely ill becomes dying: a key to
end-of-life care [editorial]. JAMA 1999; 282:16701672
62 Quill TE. Initiating end-of-life discussions with seriously ill
patients: addressing the elephant in the room. JAMA 2000;
284:25022507
63 Curtis RJ, Patrick DL, Shannon SE, et al. The family
conference as a focus to improve communication about
end-of-life care in the intensive care unit: opportunities for
improvement. Crit Care Med 2001; 29(suppl): N26N33
64 Verhoef MJ, Hagen N, Pelletier G, et al. Alternative therapy
use in neurologic disease: use in brain tumor patients.
Neurology 1999; 52:617622
65 Burstein HJ, Gelber S, Guadagnoli E, et al. Use of alterna-
tive medicine by women with early-stage breast cancer.
N Engl J Med 1999; 340:17331739
66 Holland JC. Use of alternative medicine: a marker for
distress? N Engl J Med 1999; 340:17581759
67 Abraham JL, Callahan J, Rossetti K, et al. The impact of a
hospice consultation team on the care of veterans with
advanced cancer. J Pain Symptom Manage 1996; 12:2331
68 Daalcman TP, VandeCreek L. Placing religion and spiritu-
ality in end-of-life care. JAMA 2000; 284:25142517
69 Doyle D, Hanks GWC, MacDonald N, eds. Oxford textbook
of palliative medicine. 2nd ed. New York, NY: Oxford
70 Von Gunten C, Ferris F, Emanuel L. Ensuring competency
in end-of-life care: communication and relational skills.
JAMA 2000; 284:30513057
71 Faber-Langendoen K. The clinical management of dying
patients receiving mechanical ventilation: a survey of physi-
cians practice. Chest 1994; 106:880888
72 Cruzan v Director, MDH, 497 US 261 (1990)
73 Omnibus Reconciliation Act of., Pub. L. No. 101508
1990:4206
74 Tonelli MR. Pulling the plug on living wills: a critical analysis
of advance directives. Chest 1996; 110:816822
75 Garner BA, ed. Blacks law dictionary. 7th ed. St. Paul, MN:
West Publication, 1999
76 Heffner JE, Barbieri C. End-of-life preferences of patients
enrolled in cardiovascular rehabilitation programs. Chest
2000; 117:14741481
77 Suhl J, Simons P, Reddy T, et al. Myth of substituted
judgment: surrogate decision making regarding life support
is unreliable. Arch Intern Med 1994; 154:9096
78 Hare J, Pratt C, Nelson C. Agreement between patients and
their self-selected surrogates on difficult medical decisions.
Arch Intern Med 1992; 152:10491054
79 Perkins HS. Time to move advance care planning beyond
advance directives. Chest 2000; 117:12281231
80 22A Am Jur. 2d Death 689 (1988)
81 Danis M, Southerland MPH, Garrett J, et al. A prospective
study of advance directives for life-sustaining care. N Engl
J Med 1991; 324:882888
82 Johnson RF, Baranowski-Birkmeier T, ODonnell JB. Ad-
vance directives in the medical intensive care unit of a
community teaching hospital. Chest 1995; 107:752756
83 Robertson JA. Second thoughts on living wills: 21 Hastings
Center Report, 1991; 69
84 Pope TM. The maladaptation of Miranda to advance direc-
tives: a critique of the implementation of the Patient Self
Determination Act. Health Matrix 1999; 9:139202
85 Robb MJR. The right to refuse life sustaining medical
treatment: a right without a remedy? Univ Dayton Law Rev
1997; 23:169188
86 Kutner L. Due process of euthanasia: the living will, a
proposal. Indiana Law J 1969; 44:539554
87 Kutner L. Euthanasia: due process for death with dignity;
the living will. Indiana Law J 1979; 54:201228
88 Kutner L. The living will: coping with the historical event of
death. Baylor Law Rev 1975; 27:3953
89 Bok S. Personal directions for care at the end of life. N Engl
J Med 1976; 295:367369
90 Danis M. Following advance directives. 24 Hastings Center
Report. 1994; S21S23
91 Gunter-Hunt G, Mahoney JC, and Steger CE. A comparison
of state advance directive documents. Gerontologist 2002;
42:5160
92 Meisel A. The right to die. New York, NY: John Wiley and
Sons, 1989
93 Committee ethical problems. Deciding to forego life-sus-
taining treatment: The Presidents Commission for the
Study of Ethical Problems in Medicine and Biomedical and
Behavioral Research. Washington, DC: US Government
Printing Office, 1983; 160170
94 Joint Commission Accreditation Healthcare. Comprehen-
sive accreditation manual for hospitals. Oakbrook Terrace,
IL: Joint Commission on Accreditation of Health Care
Organizations, 1992
95 Youngner S, et al. A national survey of hospital ethics
committees. Crit Care Med 1983; 11:902905
96 McGee G, Caplan AL, Spanogle JP, et al. A national study of
ethics committees. Am J Bioethics 2001; 1:6064
97 Aulisio MP, Arnold RM, Younger SJ. Health care ethics
consultation: nature, goals, and competencies. Ann Intern
Med 2000; 133:5969
98 Fletcher JC, Siegler M. What are the goals of ethics
consultation? A consensus statement. J Clin Ethics 1996;
7:122126
99 Ackerman TF. The role of an ethicist in health care. In:
Anderson GR, Glesnes-Anderson VA, eds. Health care
ethics: a guide for decision makers. Rockville, MD: Aspen
Publishers, 1987; 309320
100 LaPuma J, Stocking CB, Silverstein MD, et al. An ethics
consultation service in a teaching hospital: utilization and
evaluation. JAMA 1988; 260:808811
101 Orr RD, Moon E. Effectiveness of an ethics consultation
service. J Fam Practice 1993; 36:4953
102 LaPuma J, Stocking CB, Darling CM, et al. Community
hospital ethics consultation: evaluation and comparison with
a university hospital service. Am J Med 1992; 92:346351
103 Fox E, Arnold RM. Evaluating outcomes in ethics consul-
tation research. J Med Ethics 1996; 7:127138
104 Faden RR, Beauchamp TL. A history and theory of in-
formed consent. New York, NY: Oxford University Press,
1986
105 King JMP. Making sense of advance directives. Dordrecht,
Holland: Kluwer Academic Publishers, 1991
106 Buchanan AE, Brock DW. Deciding for others: the ethics of
surrogate decision making. New York, NY: Cambridge
107 Day JR, Smith ML, Erenberg G, et al. An assessment of a
formal ethics committee consultation process. HEC Forum
1994; 6:1830
108 McClung JA, Kamer RS, DeLuca M, et al. Evaluation of
medical ethics consultation: opinions of patients and health
care providers. Am J Med 1996; 100:456460
109 Ackerman TF. Conceptualizing the role of the ethics con-
sultant: some theoretical issues. In: Fletcher JC, Quist N,
Jonsen AR, eds. Ethics consultation in health care. Ann
Arbor, MI: Health Administration Press, 1989; 3752
110 Angus DC, Kelley MA, Schmitz RJ, et al. Caring for the
critically ill patient: current and projected workforce re-
quirements for care of the critically ill and patients with
pulmonary disease: can we meet the requirements of an
aging population? JAMA 2000; 284:27622770
111 Groeger JS, Guntupalli KK, Strosberg M, et al. Descriptive
analysis of critical care units in the United States: patient
characteristics and intensive care unit utilization. Crit Care
Med 1993; 21:279291
112 Groeger JS, White P Jr, Nierman DM, et al. Outcome for
cancer patients requiring mechanical ventilation. J Clin
Oncol 1999; 17:991997
113 Kress JP, Christenson J, Pohlman AS, et al. Outcomes of
critically ill cancer patients in a university hospital setting.
114 Groeger JS, Lemeshow S, Price K, et al. Multicenter
outcome study of cancer patients admitted to the intensive
care unit: a probability of mortality model. J Clin Oncol
1998; 16:761770
115 Marsh HM, Krishan I, Naessens JM, et al. Assessment of
prediction of mortality by using the APACHE II scoring
system in intensive-care units. Mayo Clin Proc 1990; 65:
15491557
116 Knaus WA, Wagner DP, Lynn J. Short term mortality
predictions for critically ill hospitalized adults: science and
ethics. Science 1991; 254:389394
117 Boussat S, Elrini T, Dubiez A, et al. Predictive factors of
death in primary lung cancer patients on admission to the
intensive care unit. Intensive Care Med 2000; 26:18111816
118 Ewer MS, Ali MK, Atta MS, et al. Outcome of lung cancer
patients requiring mechanical ventilation for pulmonary
failure. JAMA 1986; 256:33643366
119 Schapira DV, Studnicki J, Bradham DD, et al. Intensive
care, survival, and expense of treating critically ill cancer
patients. JAMA 1993; 269:783786
120 Nelson J, Meier D, Oei E, et al. Self-reported symptom
experience of critically ill cancer patients receiving intensive
care. Crit Care Med 2001; 29:277282
121 Claessens MT, Lynn J, Zhong Z, et al. Dying with lung
cancer or chronic obstructive pulmonary disease: insights
from SUPPORT. J Am Geriatr 2000; 48:S146S153
122 Chalfin DB, Carlon GC. Age and utilization of intensive care
unit resources of critically ill cancer patients. Crit Care Med
1990; 18:694698
123 Prendergast TJ, Luce JM. Increasing incidence of withhold-
ing and withdrawal of life support from the critically ill. Am J
124 Danis M, Gerrity MS, Southerland LI, et al. A comparison
of patient, family, and physician assessments of the value of
medical intensive care. Crit Care Med 1988; 16:594600
125 Teno J, Fisher E, Hamel M, et al. Decision-making and
outcomes of prolonged ICU stays in seriously ill patients.
J Am Geriatr Soc 2000; 48:S70S74
126 Hammes BJ, Rooney BL. Death and end-of-life planning in
one midwestern community. Arch Intern Med 1998; 158:
383390
127 Cook DJ, Guyatt GH, Rocker G, et al. Cardiopulmonary
resuscitation directives on admission to intensive-care unit:
an international observational study. Lancet 2001; 358:
19411945
128 Faber-Langendoen K. A multi-institutional study of care
given to patients dying in hospitals: ethical and practice
implications. Arch Intern Med 1996; 156:21302136
129 Lynn J, Harrell F Jr, Cohn F, et al. Prognoses of seriously ill
hospitalized patients on the days before death: implications
for patient care and public policy. New Horiz 1997; 5:5661
130 Nelson J, Danis M. End-of-life care in the intensive care
unit: where are we now? Crit Care Med 2001; 29:N2N9
131 Rubenfeld GD, Randall CJ. End-of-life care in the intensive
care unit: a research agenda. Crit Care Med 2001; 29:2001
2006
132 Ries LG, Pollack ES, Young JL. Cancer patient survival:
surveillance, epidemiology, and end results program. J Natl
Cancer Inst 1983; 70:693709
133 American Society Clinical Oncology. Cancer care during the
last phase of life. J Clin Oncol 1998; 16:19861996
134 Higginson J, McCarthy M. Measuring symptoms in terminal
cancer: are pain and dyspnea controlled? J R Soc Med 1989;
82:264267
135 Donnelly S, Walsh D, Rybicky L. The symptoms of ad-
vanced cancer: identification of clinical and research prior-
ities by assessment of prevalence and severity. J Palliat Care
1995; 11:2732
136 Medical Research Council. Lung cancer: a medical research
council randomized trial of palliative radiation therapy with
two fractions or ten fractions. Br J Cancer 1991; 65:934941
137 Bruera E, MacEachern T, Ripmonti C, et al. Subcutaneous
morphine for dyspnea in patients. Ann Intern Med 1993;
119:906907
138 Ferrell BR, Wisdom C, Wenzl C. Quality of life as an
outcome variable in the management of cancer pain. Cancer
1989; 63(11 Suppl):23212327
139 Foley KM. Advances in cancer pain. Arch Neurol 1999;
56:413417
140 World Health Organization. Cancer pain relief. 2nd ed.
Geneva, Switzerland: World Health Organization, 1996
141 Levy MH. Pharmacologic treatment of pain. N Engl J Med
1996; 335:11241132
142 Jacox A, Carr DB, Payne R. New clinical practice guidelines
for the management of pain in patients with cancer. N Engl
J Med 1994; 330:651655
143 Wachtel T, Allen-Masterson S, Goldberg R, et al. The end
stage cancer patients: terminal common pathway. Hospice J
1988; 4:4380
144 Bayer R, Feldman E. Hospice under the Medicare wing. 12
Hastings Center Report, 1982; 56
145 Christakis NA, Escarce JJ. Survival of Medicare patients
after enrollment in hospice programs. N Engl J Med 1996;
3355:172178
146 Mor V, Hiris J. Determinants of site of death among hospice
cancer patients. J Health Soc Behav 1983; 24:375385
147 Greer DS, Mor V, Morris JN, et al. An alternative in
terminal care: results of national hospice study. J Chronic
Dis 1986; 39:926
148 Kane RL, Klein SJ, Rothenberg R, et al. Hospice role in
alleviating the emotional stress of terminal patients and their
families. Med Care 1985; 23:189197
149 Mor V, Kidder D. Final results of the national hospice study.
Health Serv Res 1985; 20:407422
150 Hill TP. Treating the dying patient: the challenge for
medical education. Arch Intern Med 1995; 155:12651269
151 Merman AC, Gune DB, Dickinson GE. Learning to care for
the dying: a survey of medical schools can form a model
course. Acad Med 1991; 66:3538
152 Brady MJ, Peterman ANM, Fithchett B, et al. A case for
including spirituality in quality of life measurement in
oncology. Psychooncology 1999; 8:417428
2003;123;312-331 Chest
Ackerman, Melinda Thompson and F. Hammond Cole, Jr
John P. Griffin, Judith E. Nelson, Kathryn A. Koch, Harvey B. Niell, Terrence F.
End-of-Life Care in Patients With Lung Cancer
& Services
Updated Information
S
References
S#BIBL
free at:
Citations
S#otherarticles
Reprints
2003;123;332-337 Chest
H. Williams, Jr
W. Michael Alberts, Gerold Bepler, Todd Hazelton, John C. Ruckdeschel and James
Practice Organization
ISSN: 0012-3692.
W. Michael Alberts, MD, MBA, FCCP, Gerold Bepler, MD; Todd Hazelton, MD;
John C. Ruckdeschel, MD, FCCP; and James H. Williams, Jr, MD, FCCP
The management of patients with suspected or known lung cancer is becoming increasingly
complex. State-of-the-art care often requires input from many sources, including pulmonology,
thoracic surgery, medical oncology, radiation oncology, pathology, and radiology. Valuable
contributions to care also come from nursing, social work, psychology, psychiatry, pastoral care,
and palliative care, among others. As a result, multidisciplinary input into care is vital. Patients
with suspected lung cancer should be expeditiously evaluated and referred for management.
Clear and understandable information on the diagnosis, treatment options, and possible out-
comes should be provided. Treatment recommendations should be based on locally agreed-on
adaptations of clinical practice guidelines. Provisions for ongoing care should be apparent to all
concerned (CHEST 2003; 123:332S337S)
Key words: communication; coordinated care; multidisciplinary care; practice organization
T
he management of patients with suspected or
known lung cancer is becoming increasingly
complex. State-of-the-art care often requires input
from many sources, including pulmonology, thoracic
surgery, medical oncology, radiation oncology, pa-
thology, and radiology. Valuable contributions to
care also come from nursing, social work, psychol-
ogy, psychiatry, pastoral care, and palliative care,
among others. As a result, multidisciplinary input
into care is vital.
Patients with lung cancer may follow a variety of
routes within the health-care system. Traditionally,
consultation and management have been achieved
with referrals to specialists occurring in a sequential
fashion. This process may be slow, and delays are
common. Numerous referrals to a variety of special-
ists may result in fragmented, poorly coordinated,
and even inappropriate care, especially where treat-
ment requires the coordination of multiple special-
ties. Establishing a seamless coordinated approach to
the management of the patient with lung cancer is
critical for state-of-the-art diagnosis, treatment, and
care.
Recommendations on practice organization are
infrequently included in lung cancer guidelines pub-
lished by other organizations or groups.
15
In gen-
eral, these guidelines advocate for multidisciplinary
care but cite the paucity of objective data.
Multidisciplinary Approach
Many groups have attempted to coordinate care
through multidisciplinary management conferences
and multidisciplinary clinics. In the latter, physicians
from several specialties conduct clinics in the same
location at the same time. In addition to streamlining
the workup and treatment planning, multidisci-
plinary conferences and clinics provide a forum for
collegial exchange of professional opinions. The
team approach and consensus development enables
the consulting physician to convey a clear and con-
sistent management opinion to the patient.
The feasibility of multiple specialty evaluations is
affected by many factors, including the availability of
physicians with appropriate expertise and willingness
to interact in this manner, referral patterns of the
community of primary care physicians, and many
economic factors. Economic factors often have broad
impact, affecting motivation of patients, participating
physicians, and referring physicians. Success has
often depended on the support of third-party payers.
Compensation for the time spent by multiple physi-
cians in a single encounter may prove an obstacle in
some settings. These issues need to be considered
when forming multidisciplinary teams, as the dura-
bility of the program will often hinge on how effec-
tively these issues are addressed and managed.
Although widely advocated, there is sparse objec-
tive evidence that a multidisciplinary approach to
treating lung cancer is more effective than traditional
care. There is evidence that patients referred to a
respiratory specialist receive more expeditious and
more appropriate care
6
and that a fast-track system
Tampa, FL.
Correspondence to: W. Michael Alberts, MD, MBA, FCCP,
Associate Center Director for Clinical Affairs, H. Lee Moffitt
Cancer Center and Research Institute, 12902 Magnolia Dr,
Tampa, FL 33612; e-mail: alberts@moffitt.usf.edu
of diagnosis and staging can increase the proportion
of patients reaching surgery.
7
Studies of multidisci-
plinary breast cancer clinics have shown an increase
in patient satisfaction and a shorter time from diag-
nosis to treatment.
8
Nevertheless, a multidisciplinary
team or clinic approach to the management of the
patient with suspected or known lung cancer seems
appropriate, where feasible.
Regardless of the presence or absence of a multi-
specialty clinic, all cancer units, treatment facilities,
and centers should have a multidisciplinary lung
cancer conference where new cases may be dis-
cussed in a prospective fashion. This conference
should occur on a regular and continuing basis. The
composition of the multidisciplinary thoracic oncol-
ogy team will vary depending on availability and local
interest but should, where possible, include repre-
sentatives from pulmonology, thoracic surgery, med-
ical oncology, radiation oncology, radiology, and
pathology. A physician with expertise and an interest
in palliative care would be a valuable asset to the
team. A lead (or coordinating) clinician should be
identified, and sufficient clerical support should be
provided. Scheduled meeting times with adequate
time allotted for all case presentations should be
identified. This forum may serve to assist in the
development of the initial treatment plan and in the
design of adjuvant, second-line, and palliative care
plans.
Medical economic forces vary between and within
regions. While ideally these factors should not affect
the care of patients with lung cancer, their impact
needs to be considered. The perspective of primary
care physicians and specialists in managed-care set-
tings may differ and may therefore lead to varying
use of resources. While the impact of these forces on
the treatment of lung cancer is incompletely docu-
mented, there is no reason to assume the forces
differ with this disease. Finally, personal and finan-
cial issues of the patient may affect their desire to
seek intervention, including the costs of medical
care, availability of family support, and the potential
impact of a prolonged period of diminished capacity
of the patient. However, awareness of these factors
should not inhibit education of patients and their
families regarding the potential options for diagnosis,
treatment, and palliation.
Recommendations
1. All cancer units, treatment facilities, and cen-
ters should have a multidisciplinary lung cancer
conference that meets on a regular and con-
tinuing basis. Quality of evidence: poor; net
benefit: substantial; strength of recommenda-
tion: C
2. Multidisciplinary lung cancer teams should
consider establishing a multispecialty lung
cancer clinic. Quality of evidence: poor; net
benefit: substantial; strength of recommen-
dation: C
Referral Pattern
Patients with suspected lung cancer are often
identified by their primary care physician. Patients
should be expeditiously referred to a physician with
experience in the management of lung cancer, often
a pulmonologist or a thoracic surgeon. Either may be
responsible for the diagnostic workup and the pre-
treatment functional assessment. If available, the
patient may be referred to a multidisciplinary lung
cancer clinic where new referrals are triaged to the
most appropriate specialist for their initial evaluation
and subsequent diagnostic and staging procedures.
Alternatively, the primary care physician may con-
duct the initial evaluation and, based on the results,
refer the patient to the appropriate specialist. Pa-
tients with presumed stage I and stage II lung cancer
may be referred directly to a thoracic surgeon. The
management of a high-risk surgical candidate, how-
ever, may require the input of a pulmonologist,
cardiologist, or other subspecialists. For patients
considered to have stage III tumors where more than
one treatment modality is usually considered, a
multidisciplinary evaluation should be performed.
This may be coordinated by the primary care physi-
cian, a pulmonologist, or by referral to a multidisci-
plinary lung cancer clinic. Should neoadjuvant che-
motherapy and/or radiation therapy prove efficacious
in patients with stage I and II, the indication for a
multidisciplinary evaluation would significantly
broaden. While other treatment modalities may be
employed and other specialists consulted, patients
with documented stage IV disease may be referred
directly to a medical oncologist.
All patients should have equal access to lung
cancer services regardless of age, gender, race, or
socioeconomic status. While age predicts the pres-
ence of other medical conditions that may affect the
management recommendation, age alone is not a
reliable predictor of the success of interventions.
9
However, effective multidisciplinary management
must tailor diagnostic, therapeutic, and palliative
approaches to accommodate the particular medical,
social, moral, ethical, and economic issues of each
patient.
Recommendations
cancer should be referred to a multidisciplinary
team of physicians or a physician with experi-
ence in the management of lung cancer. Qual-
ity of evidence: poor; net benefit: substantial;
strength of recommendation: C
4. For patients in whom tissue diagnosis or staging
remains incomplete, referral should be to a
specialist with expertise in these areas. When
completed, the choice of referral may vary with
the interventions(s) proposed. Quality of evi-
dence: poor; net benefit: moderate; strength of
recommendation: C
5. A multidisciplinary group is particularly valu-
able for management of patients who may be
offered multimodality therapy. Quality of evi-
dence: poor; net benefit: substantial; strength
Management Decisions
Management decisions emanating from the mul-
tidisciplinary conference should be guided by locally
agreed-on adaptations of clinical practice guidelines
or other evidence. There is evidence that use of
clinical practice guidelines modified to fit local prac-
tice patterns may improve both the process and the
outcome of care.
10
Documentation of recommenda-
tions made by the management team should be
periodically reviewed for adherence and resultant
patient outcomes. All patients should be evaluated as
potential candidates for clinical trials, and enroll-
ment should be encouraged. Ideally, these trials
should include opportunities to evaluate the full
range of approaches, including interventions with
curative and palliative intent, as well as studies of
comfort care near the end of life.
Management input from nursing, social work,
psychology, and palliative care is advised. A func-
tional and emotional assessment of the patient
should guide the planning and implementation of
nursing care. Nursing care should be viewed as
collaborative within the wider care team and should
focus on family-centered care.
4
Locally agreed-on
standards of nursing care, incorporating current
research and best practices, should be implemented.
There is good evidence that specialist nurses working
as a part of a multidisciplinary team can improve
patient outcomes in terms of relief of symptoms and
time spent at home.
4
Ideally, patients and family
should have easy access to an appropriately trained
specialist nurse throughout the illness.
A specific physician coordinator of care should be
identified to the patient and his or her caregivers.
This individual should serve as the point of first
contact and the orchestrator of the diagnostic, stag-
ing, and treatment plan. In many cases, this may be
the primary care physician who wants to remain
involved in the whole diagnostic and treatment
process. Alternatively, it may be the physician with
the most contact with the patient during treatment.
Recommendations
6. Management decisions emanating from the
multidisciplinary conference should be guided
by locally agreed-on adaptations of clinical
practice guidelines or other evidence. Quality
of evidence: fair; net benefit: substantial;
strength of recommendation: B
7. All patients should be evaluated as potential
candidates for clinical trials, and enrollment
should be encouraged. Quality of evidence:
poor; net benefit: none/negative; strength of
recommendation: I
8. A specific coordinator of care should be iden-
tified to the patient and caregivers. Quality
of evidence: poor; net benefit: substantial;
Timetable
Although there is no direct evidence that acceler-
ated diagnostic testing reduces mortality, individuals
with known or suspected lung cancer and their
families are understandably anxious. As a result,
rapid evaluation, diagnosis, staging, and treatment
planning is important, even if only for the manage-
ment of psychological stress. Specialists and primary
care physicians should collaborate to organize an
intake process that produces minimal delays. Coor-
dinated care, such as provided by multidisciplinary
teams, should help to minimize delay in the delivery
of care for every stage of the disease.
Guidelines developed by the British Thoracic So-
ciety recommend specific time frames for evaluation,
diagnosis, and treatment.
3
Although local situations
vary greatly and timelines may need to be adapted to
local standards, these recommendations may serve as
a general guide. All patients should be seen for an
initial evaluation within 1 week of referral by the
primary care physician. Diagnostic tests should be
performed within 2 weeks of the decision to undergo
the test, and the results should be communicated to
the patient as soon as they are available. Chemother-
apy should be initiated within 7 working days of the
decision to use a particular protocol. Radiation ther-
apy should be initiated within 2 working days for
urgent cases, within 4 weeks for radical treatment
where complex treatment planning is needed, and
within 2 weeks for palliative purposes. Surgery
should occur within 4 to 8 weeks of referral to a
management team and within 4 weeks of a surgical
evaluation, or later if purposefully delayed for initial
neoadjuvant therapy.
Recommendation
9. For patients with suspected lung cancer, eval-
uation, diagnosis, and treatment planning
should be expedited. Quality of evidence: fair;
net benefit: substantial; strength of recommen-
dation: B
Communication
Patients with lung cancer have reported a need to
have clear information about their diagnosis, treat-
ment, and possible outcomes.
11
At every stage of
treatment, patients and their families should be
offered clear, full, and prompt information in both
verbal and written form. Lack of information can
increase anxiety, uncertainty, distress, and dissatis-
faction.
11
Communication should include informa-
tion about the disease, diagnostic procedures, and
the aims and likely effects of treatment (including
potential adverse effects). Patients should be given a
clear indication of the expected start date and dura-
tion of the course of treatment. All health profes-
sionals involved in the care of the patient should
receive communication as to the clinical staging,
what the patient has been told, and the proposed
treatment plan. Information about available support
groups and organizations should be made available to
the patient and family. The patient should have
access to written information, appropriate to his or
her case, to supplement the verbal communication.
Although some patients do not wish to take an active
part in decision making, there is strong evidence that
patients value accurate information and many feel
that they are not provided sufficient information.
12
Where additional testing or consultation is deemed
necessary, this should be arranged as expeditiously as
possible.
Communicating the diagnosis and the treatment
should be undertaken with great care. Bad news may
be best communicated in person by an individual
with experience and training. If possible, the patient
should be accompanied by a relative or friend. A
quiet room is essential and the presence of a nurse
for support both during and after the meeting is
advisable. Telephone (followed by written documen-
tation) or written communication with the primary
care physician should occur within 2 days. It is
essential that there are clear and efficient channels of
communication within the lung cancer team and
with the primary care physician, nurses, and other
health-care professionals.
1
Recommendations
10. Patients with lung cancer should have clear
understandable information about their diag-
nosis, treatment, and possible outcomes. Pa-
tients and their families should be offered
clear, full, prompt, and culturally appropriate
information, preferably in both verbal and
written form. Quality of evidence: fair; net
tion: B
11. All health professionals involved in the care of
the patient should be aware of the manage-
ment plan. This communication should in-
clude the clinical staging, what the patient has
been told, and the proposed treatment plan.
Quality of evidence: poor; net benefit: sub-
stantial; strength of rcommendation: C
Ongoing Care
Patients with lung cancer are rarely cured, and
those treated with curative intent are at high risk for
relapse or second primary lung cancers. As a result,
continued follow-up by a physician or a team of
physicians with expertise in lung cancer is vital (the
reader is referred to the chapter Follow-up and
Surveillance of the Lung Cancer Patient earlier in
this issue). Explicit guidelines for follow-up and
surveillance after the initial treatment should be
developed. This plan should be appropriate to local
needs and capabilities. Any new findings or develop-
ments may be presented for opinion at the multidis-
ciplinary conference. It should be clear to the patient
who will be supervising their ongoing care and
surveillance. Patients should have clear instructions
regarding who to contact for urgent problems.
It is the responsibility of the communicating phy-
sician to provide accurate information to the patient
so that the patient and his or her family can be
meaningfully involved in treatment decisions. The
patients wishes should be sought and must be
honored. It is the physicians duty, however, to
provide adequate and appropriate information on
which the patients decision may be based. Where
major decisions are to be made concerning a change
in treatment, the decision should be made after
verbal consultation with the primary care physician
and the appropriate management team members,
including the nurses.
Unfortunately, the majority of patients with lung
cancer will ultimately succumb to their disease. As a
result, access to ongoing care, including primarily
palliative care, is vital. A system should be estab-
lished to ensure that patients receive optimal symp-
tom control, together with psychological, social, and
piritual care throughout their illness.
1
Such support
should also be available to caregivers, both during
the patients lifetime and during bereavement.
5
Ad-
mission to the hospital, where appropriate, for symp-
tom control should be expedited.
When death occurs, the primary care physician
and all management team members should be ad-
vised within 1 working day. Likewise, the primary
care physician should notify the team and all inter-
ested parties should the patient die at home. Be-
reavement information detailing the availability of
support should be provided where appropriate.
Recommendations
12. For all patients with lung cancer, explicit
guidelines for follow-up and surveillance after
the initial treatment should be developed. It
should be clear to the patient who will be
supervising their ongoing care and surveil-
lance. Patients should be aware of who and
how to access assistance for urgent problems.
stantial; strength of recommendation: C
13. For patients with lung cancer in whom death
or a significant change in clinical status occurs,
the primary care physician and all manage-
ment team members should be advised. Like-
wise, the primary care physician should notify
the management team and all interested par-
ties if a change in clinical status of the patient
should occur at home. Quality of evidence:
poor; net benefit: substantial; strength of
recommendation: C
Conclusion
State-of-the-art management of lung cancer usu-
ally requires input from and participation by a
number of specialties and disciplines. Optimal care is
therefore dependent on a coordinated series of
events from identification of the patient with possi-
ble lung cancer, to diagnosis of malignancy, to
evaluation for potential treatment options, to actual
management, and finally to palliative care. Although
objective proof is wanting, this coordination is con-
ceptually best delivered in a multidisciplinary fash-
ion, whether that be via joint conferences, virtual
networks, or actual multidisciplinary clinics. The
structure of this delivery system is dependent on
local circumstances and capabilities but would
nevertheless appear to be a laudable goal worth
pursuing.
Multidisciplinary Approach
1. All cancer units, treatment facilities, and cen-
ters should have a multidisciplinary lung cancer
conference that meets on a regular and con-
tinuing basis. Quality of evidence: poor; net
tion: C
2. Multidisciplinary lung cancer teams should
consider establishing a multispecialty lung
cancer clinic. Quality of evidence: poor;
net benefit: substantial; strength of recom-
mendation: C
Referral Pattern
cancer should be referred to a multidisciplinary
team of physicians or a physician with experi-
ence in the management of lung cancer. Qual-
ity of evidence: poor; net benefit: substantial;
4. For patients in whom tissue diagnosis or staging
remains incomplete, referral should be to a
specialist with expertise in these areas. When
completed, the choice of referral may vary with
the interventions(s) proposed. Quality of evi-
dence: poor; net benefit: moderate; strength of
recommendation: C
5. A multidisciplinary group is particularly valu-
able for management of patients who may be
offered multimodality therapy. Quality of evi-
dence: poor; net benefit: substantial; strength
Management Decisions
6. Management decisions emanating from the
multidisciplinary conference should be guided
by locally agreed-on adaptations of clinical
practice guidelines or other evidence. Quality
of evidence: fair; net benefit: substantial;
strength of recommendation: B
7. All patients should be evaluated as potential
candidates for clinical trials and enrollment
should be encouraged. Quality of evidence:
poor; net benefit: none/negative; strength of
recommendation: I
8. A specific coordinator of care should be iden-
tified to the patient and caregivers. Quality
of evidence: poor; net benefit: substantial;
Timetable
9. For patients with suspected lung cancer, eval-
uation, diagnosis, and treatment planning
should be expedited. Quality of evidence: fair;
net benefit: substantial; strength of recommen-
dation: B
Communication
10. Parients with lung cancer should have clear
understandable information about their diag-
nosis, treatment, and possible outcomes. Pa-
tients and their families should be offered
clear, full, prompt, and culturally appropriate
information, preferably in both verbal and
written form. Quality of evidence: fair; net
tion: B
11. All health professionals involved in the care of
the patient should be aware of the manage-
ment plan. This communication should in-
clude the clinical staging, what the patient has
been told, and the proposed treatment plan.
Ongoing Care
12. For all patients with lung cancer, explicit
guidelines for follow-up and surveillance after
the initial treatment should be developed. It
should be clear to the patient who will be
supervising their ongoing care and surveil-
lance. Patients should be aware of who and
how to access assistance for urgent problems.
13. For patients with lung cancer in whom death
or a significant change in clinical status occurs,
the primary care physician and all manage-
ment team members should be advised. Like-
wise, the primary care physician should notify
the management team and all interested par-
ties if a change in clinical status of the patient
should occur at home. Quality of evidence:
poor; net benefit: substantial; strength of rec-
ommendation: C
References
1 The Royal College of Radiologists Clinical Oncology Infor-
mation Network. Guidelines on the non-surgical management
of lung cancer. Clin Oncol 1999; 11:S1S53
practice guidelines. Oncology 1997; 11:889895
3 British Thoracic Society. BTS Recommendations to respira-
tory physicians for organising the care of patients with lung
cancer. Thorax 1998; 53(suppl1):S1S8
4 Scottish Cancer Therapy Network (SCTN)/Scottish Cancer
Coordinating, and Advisory Committee (SCCAC). Summary
of draft guidelines on best current management for lung
cancer. Edinburgh, Scotland: Lung Cancer Focus Group,
1996
5 National Health Service Guidance on Commissioning Cancer
Services. Improving outcomes in lung cancer. London, UK:
Department of Health, June 1998
6 Fergusson RJ, Gregor A, Dodds R, et al. Management of lung
cancer in South East Scotland. Thorax 1996; 51:569745
7 Billing JS, Wells FC. Delays in the diagnosis and surgical
treatment of lung cancer. Thorax 1996: 51:903906
8 Gabel M, Hilton NE, Nathanson SD. Multidisciplinary breast
cancer clinics: do they work? Cancer 1997; 79:23802384
9 Massard G, Muug R, Wihlm JM, et al. Bronchogenic cancer
in the elderly: operative risk and long term prognosis. Thorac
10 Smith TJ, Hillner BE. Ensuring quality cancer care by the use
of clinical practice guidelines and critical pathways. J Clin
Oncol 2001; 19:28862897
11 Sell L, Devlin B, Bourke SJ, et al. Communicating the
diagnosis of lung cancer. Respir Med 1993; 87:6163
12 Meredith C, Symonds P, Webster L, et al. Information needs
of cancer patients in the West of Scotland: cross sectional
survey of patients views. BMJ 1996; 313:724726
2003;123;332-337 Chest
H. Williams, Jr
W. Michael Alberts, Gerold Bepler, Todd Hazelton, John C. Ruckdeschel and James
& Services
Updated Information
S
References
S#BIBL
at:
Reprints

(Ghiduri) (Cancer) ACCP Diagnosis and Management of Lung Cancer - Evidence-Based Guidelines

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

(Ghiduri) (Cancer) ACCP Diagnosis and Management of Lung Cancer - Evidence-Based Guidelines

Hochgeladen von

Copyright:

Verfügbare Formate

DOI:10.1378/chest.123.1_suppl.

Das könnte Ihnen auch gefallen