og D

Pr M
m
E 4J
ra
CM 201

The Journal of Molecular Diagnostics, Vol. 16, No. 5, September 2014

jmd.amjpathol.org

Sensitive Detection of EGFR Mutations in Cerebrospinal

Fluid from Lung Adenocarcinoma Patients with Brain
Metastases
Haihong Yang,* Linbo Cai,y Yalei Zhang,* Hongyu Tan,z Qiuhua Deng,x Meiling Zhao,* and Xin Xu*

From the Respiratory Oncology Center,* State Key Laboratory of Respiratory Diseases, Department of Internal Neurology,z and Center for Translational
Medicine,x The First Affiliated Hospital, Guangzhou Medical University, Guangzhou; and the Department of Oncology,y Guangdong 999 Brain Hospital,
Guangzhou, China

CME Accreditation Statement: This activity (“JMD 2014 CME Program in Molecular Diagnostics”) has been planned and implemented in accordance with the Essential Areas and
policies of the Accreditation Council for Continuing Medical Education (ACCME) through the joint sponsorship of the American Society for Clinical Pathology (ASCP) and the
American Society for Investigative Pathology (ASIP). ASCP is accredited by the ACCME to provide continuing medical education for physicians.

The ASCP designates this journal-based CME activity (“JMD 2014 CME Program in Molecular Diagnostics”) for a maximum of 48 AMA PRA Category 1 Credit(s)
. Physicians
should only claim credit commensurate with the extent of their participation in the activity.

CME Disclosures: The authors of this article and the planning committee members and staff have no relevant financial relationships with commercial interests to disclose.

Accepted for publication

April 21, 2014.
Address correspondence to
Haihong Yang, M.D., Respira-
tory Oncology Center, the First
Affiliated Hospital of
Guangzhou Medical University,
Yanjiang Road 151, Guangzhou,
Guangdong, China. E-mail:
bjrf2009@yahoo.com.
Epidermal growth factor receptor (EGFR) mutations in cerebrospinal fluid (CSF) might be useful pre-
dictive markers for EGFR tyrosine kinase inhibitor treatment of intracranial metastatic tumors. In this
retrospective study, amplification refractory mutation system (ARMS)-PCR assays were used to inves-
tigate the EGFR gene status in 30 lung adenocarcinoma patients with brain metastases. A total of 16
patients tested positive for EGFR-activating mutations in CSF or tumor tissues. These included L858R
mutation in exon 21 in six CSF samples and exon 19 deletions in seven CSF samples. EGFR mutations
were detected between CSF and primary tumor samples with a 75% positive predictive value (95% CI,
0.45e1.00), 75% negative predictive value (95% CI, 0.51e0.99), 67% sensitivity (95% CI, 0.36e0.97),
and 82% specificity (95% CI, 0.59e1.00). Most of the patients who had EGFR mutations in CSF achieved
good responses with EGFR-tyrosine kinase inhibitor treatment. In conclusion, ARMS-PCR could be a
sensitive method of detecting EGFR mutations in the CSF of patients with lung adenocarcinoma with
brain metastases. As such, ARMS-PCR could play an important role in guiding EGFR-tyrosine kinase in-
hibitor treatments of intracranial tumors and for diagnosing brain metastases in patients with lung
adenocarcinoma. (J Mol Diagn 2014, 16: 558e563; http://dx.doi.org/10.1016/j.jmoldx.2014.04.008)

Brain metastases are a frequent complication of non-small
cell lung cancer (NSCLC), especially in patients with lung
adenocarcinoma. Brain metastases are observed in 30% to
50% of patients at initial diagnosis with more patients
developing metastases during treatment.1 These patients are
unable to undergo surgical resection of primary or cranial
metastatic tumors to provide specimens for histopatholog-
ical or biomarker studies. However, tumor-derived DNA
could be secreted into body fluids surrounding the tumor.
Therefore, the development of methods to identify potential
molecular biomarkers from nonsurgical biopsy samples,
such as cerebrospinal fluid (CSF), may facilitate the iden-
tification of clinically relevant gene signatures in patients
with metastatic brain tumors.
Brain metastases are diagnosed according to clinical pre-
sentation, primary malignant tumor, and radiological imaging.
If the computed tomography or magnetic resonance imaging
(MRI) aspect is atypical, tissue diagnosis, including brain
Supported by the National Natural Science Foundation of China (grant
#81302000).
Disclosures: None declared.

Copyright ª 2014 American Society for Investigative Pathology

and the Association for Molecular Pathology.
Published by Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jmoldx.2014.04.008
 EGFR Mutations in CSF of Brain Metastases

tumor or CSF cytology, is necessary.2 However, in certain
clinical situations, MRI would not be helpful for patients with
leptomeningeal metastases in which positive CSF cytology
results are <40%.3 The detection of oncogenes in CSF might
facilitate the diagnosis of brain metastases in patients with lung
adenocarcinoma, especially if the results are consistent with
the corresponding results from the primary tumor.
Epidermal growth factor receptor (EGFR) tyrosine kinase
inhibitor (TKI) is a small-molecular agent capable of
penetrating brain tissue and has been found to significantly
improve survival rates and tumor responses in lung adeno-
carcinoma patients with metastatic brain tumors that harbor
EGFR-activating mutations.4,5
However, the most common target populations for treat-
ment with EGFR-TKI remain females with adenocarcinomas,
because these patients have a higher rate of EGFR mutation,
and EGFR gene status can only be detected in approximately
10% of patients with advanced NSCLC in China.6 The limited
availability of testing technology and economic factors are the
leading causes of such a low detection rate. The evaluation of
the EGFR status in metastatic intracranial tumors may offer
direct evidence to guide the clinical use of EGFR-TKI in
treating these patients. Because positive rates of CSF cytology
are low and readily available methods may not be suitable, the
detection of EGFR gene status in tumor-derived free DNA in
CSF might be a good clinical option.7
To test this hypothesis, we analyzed the EGFR status of
tumor-derived free DNA in the CSF of lung adenocarci-
noma patients with brain metastases by using amplification
refractory mutation system (ARMS)-PCR assays.
Material and Methods
paraffin-embedded primary tumor tissues were collected in
fine-needle aspiration by bronchial fiberscopic or percutaneous
transthoracic biopsy.
Genomic DNA and cell-free DNA were extracted from
formalin-fixed, paraffin-embedded lung tumor tissues and
CSF samples, respectively, by using a QIAamp DNA FFPE
Tissue Kit or a QIAamp Circulating Nucleic Acid Kit
(Qiagen, Hilden, Germany) as appropriate. EGFR mutations
were detected with the AmoyDx Human EGFR Gene 29
Mutations Detection kit with fluorescence PCR (Amoy
Diagnostics, Xiamen, China), and assays were performed
on an ABI7900 real-time PCR instrument (Applied Bio-
systems, Foster City, CA). Primers were labeled with
6-carboxyfluorescein and HEX/VIC. This EGFR kit detects
29 mutations in exons 18 to 21, including T790M, L858R,
L861Q, S768I, G719S, G719A, and G719C; three insertions
in exon 20; and 19 deletions in exon 19. DNA was amplified
by PCR in a final volume of 25 mL that contained 5 mL of
DNA, 25 mmol/L MgCl2, 25 mmol/L dNTP, 100 mmol/L of
specific forward and reverse primers, 10  Takara buffer,
and 5 U/mL Takara HS-Taq (TaKaRa Biotechnology,
Dalian, China). The first cycle of amplifications was per-
formed with a 5-minute initial denaturation at 95 C, fol-
lowed by 30 cycles of 45 seconds at 95 C, 45 seconds at
54 C, and 1 minute at 72 C, and a 6-minute final extension
at 72 C. Products from the first cycle were amplified in the
secondary cycle by using the same PCR conditions.
Statistical Analysis
Fisher’s exact method was used to compare EGFR status
between paired CSF and primary tumor tissues in patients

Patient Selection
Table 1 Clinical Characteristics of Lung Adenocarcinoma
Thirty patients with pathologically confirmed diagnoses of Patients with Brain Metastases (n Z 30)
lung adenocarcinoma with brain metastases were enrolled in Characteristic n (%)
this retrospective study. The patients had all been admitted to Age (years)
the First Affiliated Hospital of Guangzhou Medical University 60 10 (33.3)
and Guangdong 999 Brain Hospital between November 2011 <60 20 (66.6)
and December 2012. The inclusion criteria were as follows: Sex
they had not received prior brain radiotherapy, and patients Male 16 (53.3)
with posterior fossa lesions or intracranial hypertension were Female 14 (46.7)
excluded. Tumor responses were evaluated by radiological Smoking status
computed tomography imaging or MRI according to guide- No 17 (56.7)
Yes 13 (43.3)
lines of Response Evaluation Criteria in Solid Tumors version
Brain metastases
1.1.8 Written informed consent was obtained from all partici-
Only metastatic tumors 23 (76.7)
pants before the study. This study was approved by the insti- With meningeal lesions 7 (23.3)
tutional review board of the First Affiliated Hospital, Prior treatment
Guangzhou Medical University (Guangzhou, China). Only chemotherapy 3 (10)
Only gefitinib 4 (13.3)
CSF Sample, Primary Tissue Collection, and Genotyping Gefitinib and chemotherapy 7 (23.3)
of EGFR Mutations No treatment 16 (53.3)
CSF EGFR mutation
A lumbar puncture was performed on each patient during Positive 13 (43)
Negative 17 (57)
which time 5 mL of CSF was aspirated. Formalin-fixed,

The Journal of Molecular Diagnostics - jmd.amjpathol.org 559

Yang et al

with lung adenocarcinoma with brain metastases. Differ- Results

ences were considered to be statistically significant when
P < 0.05. EGFR status in primary tumor tissue is Patient Characteristics
considered to be a standard test method and was evaluated
in CSF by calculating the positive predictive value (PPV), Thirty patients with lung adenocarcinoma metastatic to the
negative predictive value (NPV), sensitivity, and specificity brain were enrolled in this retrospective study. They included
as follows: i) PPV Z number of true positives/(number of 16 males and 14 females with a median age of 55 years (range,
true positives þ number of false positives); ii) NPV Z 28 to 82 years). Of these, 16 patients had received no prior
number of true negatives/(number of true negatives þ treatment, 11 had received gefitinib with 8 showing good
number of false negatives); iii) sensitivity Z number of responses, and 7 patients had leptomeningeal metastases. The
true positives/(number of true positives þ number of false patient characteristics are summarized in Tables 1 and 2.
negatives); and iv) specificity Z number of true negatives/
(number of true negatives þ number of false positives). EGFR Status in CSF or Tumor Tissues
SPSS version 13.0 (SPSS Inc., Chicago, IL) was used for
the statistical analyses. Binomial confidence intervals were EGFR mutations were detected in either the CSF or primary
calculated with Stata version 10.0 (StataCorp LP, College tissues of 16 of 30 patients (53%) by ARMS-PCR assays
Station, TX). (Table 2).
Table 2 Patient Characteristics and EGFR Mutations
Following
Previous Primary Tumor TKI
Age, Neurological gefitinib CSF EGFR tissues EGFR Following response
Patient Sex years Brain metastasis symptoms Smoking treatment mutation collection mutation treatment (intracranial)
1 F 80 Single lesion No No PR Wild-type NA NA Gefitinib PD
2 F 74 Single lesion Yes No PD Wild-type NA NA Gefitinib PD
3 F 71 Multiple lesions Yes No NA L858R B L858R Gefitinib* PR
4 F 61 Single lesion No No NA L858R P Wild-type Gefitinib* PR
5 F 59 Multiple, meningeal No No PR 19-del B 19-del Erlotinib CR
lesions
6 F 58 Multiple lesions No No PR 19-del NA NA Erlotinib PR
7 F 55 Multiple lesions No No PD Wild-type P Wild-type Erlotinib PD
8 F 53 Multiple lesions Yes No NA Wild-type NA NA Gefitinib PD
9 F 52 Meningeal lesions Yes No PR L858R P L858R Erlotinib CR
10 F 52 Multiple lesions Yes No NA 19-del B Wild-type Gefitinib* PR
11 F 37 Multiple lesions No No NA Wild-type B Wild-type Erlotinib SD
12 F 34 Multiple lesions No No PR Wild-type NA NA Erlotinib PR
13 F 32 Multiple, meningeal Yes No PR Wild-type NA NA Erlotinib CR
lesions
14 F 28 Single lesion No No NA Wild-type P Wild-type Chemotherapy SD
15 M 82 Single lesion No No PD Wild-type B Wild-type Erlotinib PD
16 M 76 Single lesion No No SD Wild-type P 19-del Erlotinib PD
17 M 75 Meningeal lesions Yes No NA 19-del NA NA Chemotherapy NA
18 M 68 Meningeal lesions Yes Yes NA Wild-type P Wild-type Erlotinib PD
19 M 65 Meningeal lesions Yes Yes NA L858R P L858R Erlotinib SD
20 M 60 Multiple lesions No Yes NA 19-del NA NA Chemotherapy SD
21 M 58 Multiple lesions Yes No PR 19-del B 19-del NA NA
22 M 57 Multiple lesions Yes Yes NA Wild-type P Wild-type Brain radiotherapy SD
23 M 55 Multiple lesions Yes Yes NA Wild-type P Wild-type Brain radiotherapy NA
24 M 54 Single lesion No Yes NA Wild-type P Wild-type Erlotinib PD
25 M 53 Multiple lesions No Yes NA Wild-type B 19-del Erlotinib CR
26 M 49 Multiple lesions No No NA 19-del NA NA Gefitinib SD
27 M 49 Single lesion No Yes PD Wild-type B Wild-type Brain radiotherapy PD
28 M 47 Single lesion No Yes NA L858R NA NA Erlotinib PR
29 M 42 Multiple lesions Yes Yes NA Wild-type P L858R Brain radiotherapy NA
30 M 36 Meningeal lesions Yes No PR L858R P L858R Erlotinib* SD
*The following therapy was TKI combined with brain radiotherapy.
F, female; M, male; B, fine needle aspiration by bronchial fiberscope; CR, complete remission; CSF, cerebrospinal fluid; EGFR, epidermal growth factor
receptor; NA, not applicable; P, percutaneous transthoracic biopsy; PD, progressive disease; PR, partial remission; SD, stable disease; TKI, tyrosine kinase
inhibitor.

560 jmd.amjpathol.org - The Journal of Molecular Diagnostics

 EGFR Mutations in CSF of Brain Metastases

Table 3 EGFR Mutations in CSF or Tumor Tissues
EGFR Mutation Tumor EGFR mutation (n Z 20)
CSF EGFR mutation (n Z 20) þ 
þ 6 2 detected in both the CSF and primary tumor tissue, and he
 3 9 eventually received a diagnosis of leptomeningeal metasta-
P value 0.065
PPV (95% CI) 0.75 (0.45e1.00)
NPV (95% CI) 0.75 (0.51e0.99)
Sensitivity (95% CI) 0.67 (0.36e0.97)
Specificity (95% CI) 0.82 (0.59e1.00)
CSF, cerebrospinal fluid; EGFR, epidermal growth factor receptor; NPV,
negative predictive value; PPV, positive predictive value.
The CSF samples of all of the patients were analyzed, and
EGFR mutations were detected in the CSF of 13 of 30
patients (43%). The mutations included L858R in exon 21
in 6 of 30 CSF samples (20%) and exon 19 deletions in 7 of
30 samples (23%). Metastatic tumors in the brain paren-
chyma were diagnosed in eight of these patients, including
two with a single intracranial lesion.
The primary lung adenocarcinoma tissue samples of 20
patients were analyzed. L858R mutations were identified in
the tissue samples from 5 of 20 patients (25%), and exon 19
deletions were detected in the samples from 4 of 20 patients
(20%).
Comparison of EGFR Mutations between CSF and
Primary Tumors
EGFR status was compared between CSF samples from
patients with brain metastases and paired primary lung
tumor samples (Table 3). The PPV and NPV of the EGFR
mutation test in CSF samples versus primary tumor sam-
ples were 6/8 (0.75, 95% CI: 0.45 to 1.00) and 9/12 (0.75;
95% CI, 0.51e0.99), respectively. The sensitivity and
specificity of the CSF test were 6/9 (0.67; 95% CI,
0.36e0.97) and 9/11 (0.82; 95% CI, 0.59e1.00), respec-
tively. We did not find a difference in EGFR status be-
tween CSF and primary tumor tissues (P Z 0.065,
k Z 0.49; 95% CI, 0.11e0.87).
instability, and intellectual impairment after 2 years. A
cranial MRI showed ventricle hydrops (Figure 1), and the
CSF cytology was negative. However, L858R mutation was
ses. After receiving treatment with erlotinib there was rapid
remission of the neurological symptoms.
Discussion
Traces of tumor-derived free DNA were extracted from the
CSF of 30 patients with lung adenocarcinoma with brain
metastases. We analyzed the traces of DNA by ARMS-
PCR, one of the real-time PCR methods, which is more
sensitive than Sanger DNA sequencing methods for low
abundance of DNA.9,10 The use of ARMS-PCR was
necessary because DNA levels in CSF are low and not
solely derived from metastatic tumors. Our results indi-
cate the ARMS-PCR method may be suitable for patients
with a low abundance of mutant DNA in their CSF. With
the use of ARMS-PCR, we detected EGFR mutations
in the CSF of 53% of patients and achieved a 75% PPV
(95% CI, 0.45e1.00) and 75% NPV (95% CI,
0.51e0.99), 67% sensitivity (95% CI, 0.36e0.97) and
82% specificity (95% CI, 0.59e1.00) correlation between
CSF and primary tumor tissue samples despite no sta-
tistical difference between CSF and primary tumor tissue
samples because of too small samples (P Z 0.065).
Indeed, between the paired primary tumor and tumor-
derived DNA in CSF, 75% of samples had a concor-
dant EGFR status. This is in close agreement with pre-
vious studies that reported EGFR mutation status in 70%

Clinical Response and EGFR Status in CSF

Of the 13 patients with EGFR mutations in their CSF, 6 only
received EGFR-TKI treatment after a diagnosis of brain
metastases. Four patients achieved a complete response or
partial response of their intracranial tumor after erlotinib
treatment, including three patients who showed an initial
good response to gefitinib. Of these, two patients had
L858R mutations and the other two had exon 19 deletions.
The median overall survival after the beginning of treatment
of brain metastases was 11.5 months (range, 4 to 21
months). Figure 1 The cranial MRI of patient 19 with right lung cancer. The
A patient (patient 19) who had right lung cancer under- arrow shows lateral ventricles, an enlarged third ventricle, and associated
went radical resection and developed severe headaches, gait hydrocephalus.

The Journal of Molecular Diagnostics - jmd.amjpathol.org 561

Yang et al
to 80% of the metastatic primary tumors of NSCLC pa-
tients.11,12 Moreover, false-negative results in the CSF
may stem from pretreatment with gefitinib, resulting in
low tumor burden, and patients without gefitinib pre-
treatment might be expected to have increased amounts
of mutated DNA in the CSF. In contrast, a study on the
EGFR status in patients with neoplastic meningitis re-
ported EGFR mutations by DNA sequencing in 45% of
the patients with positive CSF cytology and 30% of the
patients in CSF with negative CSF cytology.3
Because of the difficulty in obtaining intracranial tumors,
it was not possible for us to compare EGFR status between
CSF samples and intracranial tumors.
Although the EGFR status in CSF from patients with
leptomeningeal metastases is relatively easy to detect
because the tumor-derived DNA is secreted directly into the
CSF from the tumor cells, there are few reports on EGFR
status in tumor-derived DNA in CSF from patients with
metastatic tumors in the brain parenchyma. In our study,
most of the patients with EGFR mutations in CSF had
metastatic tumors in the brain parenchyma, including two
patients with a single intracranial lesion. EGFR-TKI has
been found to significantly improve the response rates of
metastatic brain tumors and survival rates of lung adeno-
carcinoma patients with asymptomatic synchronous brain
metastasis, particularly in patients with EGFR-activating
mutations in exons 19 or 21.4,5 It has been reported that
[11C]-erlotinib positron emission tomography/computed to-
mography accumulates in brain metastases,13 and this was
also found in brain metastases from patients with NSCLC.
Therefore, we exploited the sensitivity of ARMS-PCR to
determine the status of the EGFR oncogene in brain me-
tastases. In our study, four of the six patients with EGFR-
activating mutations in CSF, including three patients who
had undergone pretreatment with gefitinib, achieved good
intracranial tumor responses after receiving erlotinib for
brain metastases. This result is in agreement with a report by
Wu et al14 on the treatment of asymptomatic brain metas-
tases in NSCLC patients with EGFR mutations. We propose
that a further study should be performed to observe the
progress of intracranial diseases by following the EGFR
gene mutation status in their CSF.
However, the current methods of diagnosing brain me-
tastases are limited by suboptimal sensitivity or specificity,
especially in patients with neoplastic meningeal or meta-
static tumors that are difficult to distinguish from mani-
festations of the primary neurological diseases only by
brain imaging.15 Therefore, an alternative, more sensitive
method of diagnosing brain metastases is required for
clinical applications. Oncogenes are tumor specific and are
not present in normal tissues; therefore, they can be
detected in tumor-derived DNA in malignant pleural fluid
or plasma from patients with lung cancer.16,17 This sug-
gests that oncogene detection in body fluids could
contribute to the diagnosis of specific metastatic lesions. In
the case of the patient 19 in our study, CSF examination
562
was helpful for the diagnosis and treatment of his lep-
tomeningeal metastases.
In conclusion, ARMS-PCR could be a sensitive assay
for detecting EGFR mutations in the CSF of lung
adenocarcinoma patients with brain metastases. As such,
it could be used as a guide in clinical EGFR-TKI thera-
pies for patients with intracranial tumors, improve pre-
dictions of EGFR-TKI efficacy, and contribute to the
diagnosis of brain metastases.
References
1. Hazard LJ, Jensen RL, Shrieve DC: Role of stereotactic radio-
surgery in the treatment of brain metastases. Am J Clin Oncol 2005,
28:403e410
2. Soffietti R, Cornu P, Delattre JY, Grant R, Graus F, Grisold W,
Heimans J, Hildebrand J, Hoskin P, Kalljo M, Krauseneck P,
Marosi C, Siegal T, Vecht C: EFNS Guidelines on diagnosis and
treatment of brain metastases: report of EFNS Task Force. Eur J Neurol
2006, 13:674e681
3. Shingyoji M, Kageyama H, Sakaida T, Nakajima T, Matsui Y,
Itakura M, Iuchi T, Yokoi S, Kimura H, Iizasa T: Detection of
epithelial growth factor receptor mutations in cerebrospinal fluid from
patients with lung adenocarcinoma suspected of neoplastic meningitis.
J Thorac Oncol 2011, 6:1215e1220
4. Kim JE, Lee DH, Choi Y, Yoon DH, Kim SW, Suh C, Lee JS:
Epidermal growth factor receptor tyrosine kinase inhibitors as a first-
line therapy for never-smokers with adenocarcinoma of the lung
having asymptomatic synchronous brain metastasis. Lung Cancer
2009, 65:351e354
5. Porta R, Sánchez-Torres JM, Paz-Ares L, Massutí B, Reguart N,
Mayo C, Lianes P, Queralt C, Guillem V, Salinas P, Catot S, Isla D,
Pradas A, Gúrpide A, de Castro J, Polo E, Puig T, Tarón M,
Colomer R, Rosell R: Brain metastases from lung cancer responding to
erlotinib: the importance of EGFR mutation. Eur Respir J 2011, 37:
624e631
6. Xue C, Hu Z, Jiang W, Zhao Y, Xu F, Huang Y, Wu J, Zhang Y,
Zhao L, Zhang J, Chen L, Zhang L: National survey of the medical
treatment status for non-small cell lung cancer (NSCLC) in China.
Lung Cancer 2012, 77:371e375
7. Rhodes CH, Honsinger C, Sorenson GD: Detection of tumor-derived
DNA in cerebrospinal fluid. J Neuropathol Exp Neurol 1994, 53:
364e368
8. Eisenhauer EA, Therasse P, Bogaerts J, Schwartz LH, Sargent D,
Ford R, Dancey J, Arbuck S, Gwyther S, Mooney M, Rubinstein L,
Shankar L, Dodd L, Kaplan R, Lacombe D, Verweij J: New response
evaluation criteria in solid tumours: revised RECIST guideline (version
1.1). Eur J Cancer 2009, 45:228e247
9. Ellison G, Donald E, McWalter G, Knight L, Fletcher L, Sherwood J,
Cantarini M, Orr M, Speake G: A comparison of ARMS and DNA
sequencing for mutation analysis in clinical biopsy samples. J Exp Clin
Cancer Res 2010, 29:132
10. Angulo B, Conde E, Suárez-Gauthier A, Plaza C, Martínez R,
Redondo P, Izquierdo E, Rubio-Viqueira B, Paz-Ares L, Hidalgo M,
López-Ríos F: A comparison of EGFR mutation testing methods in
lung carcinoma: direct sequencing, real-time PCR and immunohisto-
chemistry. PLos One 2012, 7:e43842
11. Monaco SE, Nikiforova MN, Cieply K, Teot LA, Khalbuss WE,
Dacic S: A comparison of EGFR and KRAS status in primary
lung carcinoma and matched metastases. Hum Pathol 2010, 41:
94e102
12. Han C, Ma J, Zhao J, Zhou Y, Jing W, Zou H: EGFR mutations,
gene amplification, and protein expression and KRAS mutations in
primary and metastatic tumors of nonsmall cell lung cancers and
jmd.amjpathol.org - The Journal of Molecular Diagnostics

their clinical implications: a meta-analysis. Cancer Invest 2011, 29:
626e634
13. Weber B, Winterdahl M, Memon A, Sorensen BS, Keiding S,
Sorensen L, Nexo E, Meldgaard P: Erlotinib accumulation in brain
metastases from non-small cell lung cancer: visualization by posi-
tron emission tomography in a patient harboring a mutation in the
epidermal growth factor receptor. J Thorac Oncol 2011, 6:
1287e1289
14. Wu YL, Zhou C, Cheng Y, Lu S, Chen GY, Huang C, Huang YS,
Yan HH, Ren S, Liu Y, Yang JJ: Erlotinib as second-line treatment in
patients with advanced non-small-cell lung cancer and asymptomatic
The Journal of Molecular Diagnostics - jmd.amjpathol.org
EGFR Mutations in CSF of Brain Metastases
brain metastases: a phase II study (CTONG-0803). Ann Oncol 2013,
24:993e999
15. Chamberlain MC: Neoplastic meningitis. Neurologist 2006, 12:179e187
16. Brevet M, Johnson ML, Azzoli CG, Ladanyi M: Detection of EGFR
mutations in plasma DNA from lung cancer patients by mass spec-
trometry genotyping is predictive of tumor EGFR status and response
to EGFR inhibitors. Lung Cancer 2011, 73:96e102
17. Zhang X, Zhao Y, Wang M, Yap WS, Chang AY: Detection and
comparison of epidermal growth factor receptor mutations in cells and
fluid of malignant pleural effusion in non-small cell lung cancer. Lung
Cancer 2008, 60:175e182
563