The n e w e ng l a n d j o u r na l of
review article
From the Division of Neuro-Oncology,
Department of Neurology, Brigham and
Women’s Hospital; and the Center for
Neuro-Oncology, Dana−Farber Cancer In-
stitute — both in Boston. Address reprint
requests to Dr. Wen at the Center for
Neuro-­Oncology, Dana−Farber/Brigham
and Women’s Cancer Center, SW430D,
44 Binney St., Boston, MA 02115, or at
pwen@partners.org.
N Engl J Med 2008;359:492-507.
Copyright © 2008 Massachusetts Medical Society.
492
Downloaded from www.nejm.org on December 12, 2008 . Copyright © 2008 Massachusetts Medical Society. All rights reserved.
m e dic i n e
Medical Progress
Malignant Gliomas in Adults
Patrick Y. Wen, M.D., and Santosh Kesari, M.D., Ph.D.
M
alignant gliomas account for approximately 70% of the 22,500
new cases of malignant primary brain tumors that are diagnosed in adults
in the United States each year.1-3 Although relatively uncommon, malig-
nant gliomas are associated with disproportionately high morbidity and mortality.
Despite optimal treatment, the median survival is only 12 to 15 months for patients
with glioblastomas and 2 to 5 years for patients with anaplastic gliomas. Recently,
there have been important advances in our understanding of the molecular patho-
genesis of malignant gliomas and progress in treating them. This review summa-
rizes the diagnosis and management of these tumors in adults and highlights some
areas under investigation.
EPIDEMIOL O GIC FE AT UR E S
The annual incidence of malignant gliomas is approximately 5 cases per 100,000
people.1,2 Each year, more than 14,000 new cases are diagnosed in the United
States.1,2 Glioblastomas account for approximately 60 to 70% of malignant gliomas,
anaplastic astrocytomas for 10 to 15%, and anaplastic oligodendrogliomas and
anaplastic oligoastrocytomas for 10%; less common tumors such as anaplastic
ependymomas and anaplastic gangliogliomas account for the rest.1,2 The incidence
of these tumors has increased slightly over the past two decades, especially in the
elderly,4 primarily as a result of improved diagnostic imaging. Malignant gliomas
are 40% more common in men than in women and twice as common in whites as
in blacks.2 The median age of patients at the time of diagnosis is 64 years in the
case of glioblastomas and 45 years in the case of anaplastic gliomas.2,4
No underlying cause has been identified for the majority of malignant gliomas.
The only established risk factor is exposure to ionizing radiation.4 Evidence for an
association with head injury, foods containing N-nitroso compounds, occupational
risk factors, and exposure to electromagnetic fields is inconclusive.4 Although there
has been some concern about an increased risk of gliomas in association with the
use of cellular telephones,5 the largest studies have not demonstrated this.4,6,7 There
is suggestive evidence of an association between immunologic factors and gliomas.
Patients with atopy have a reduced risk of gliomas,8 and patients with glioblasto-
ma who have elevated IgE levels appear to live longer than those with normal levels.9
The importance of these associations is unclear. Gene polymorphisms that affect
detoxification, DNA repair, and cell-cycle regulation have also been implicated in
the development of gliomas.4
Approximately 5% of patients with malignant gliomas have a family history of
gliomas. Some of these familial cases are associated with rare genetic syndromes,
such as neurofibromatosis types 1 and 2, the Li−Fraumeni syndrome (germ-line p53
mutations associated with an increased risk of several cancers), and Turcot’s syn-
drome (intestinal polyposis and brain tumors).10 However, most familial cases have
n engl j med 359;5  www.nejm.org  july 31, 2008
 Medical Progress

no identified genetic cause. Recently, an interna- and 3).18,22 Glioblastomas can be separated into
tional consortium, GLIOGENE, was established two main subtypes on the basis of biologic and
to study the genetic basis of familial gliomas.11 genetic differences.18,22 Primary glioblastomas typ-
ically occur in patients older than 50 years of age
PATHOL O GIC A L FE AT UR E S and are characterized by EGFR amplification and
mutations, loss of heterozygosity of chromosome
Malignant gliomas are histologically heteroge- 10q, deletion of the phosphatase and tensin ho-
neous and invasive tumors that are derived from mologue on chromosome 10 (PTEN), and p16 de-
glia. The World Health Organization (WHO) clas- letion. Secondary glioblastomas are manifested in
sifies astrocytomas on the basis of histologic fea- younger patients as low-grade or anaplastic astro-
tures into four prognostic grades: grade I (pilo- cytomas and transform over a period of several
cytic astrocytoma), grade II (diffuse astrocytoma), years into glioblastomas. These tumors, which are
grade III (anaplastic astrocytoma), and grade IV much less common than primary glioblastomas,
(glioblastoma).1 Grade III and IV tumors are con- are characterized by mutations in the p53 tumor-
sidered malignant gliomas. Anaplastic astrocy- suppressor gene,23 overexpression of the platelet-
tomas are characterized by increased cellularity, derived growth factor receptor (PDGFR), abnor-
nuclear atypia, and mitotic activity. Glioblastomas malities in the p16 and retinoblastoma (Rb)
also contain areas of microvascular proliferation, pathways, and loss of heterozygosity of chromo-
necrosis, or both (Fig. 1 and 2). Uncommon glio- some 10q.18,24 Secondary glioblastomas have tran-
blastoma variants include gliosarcomas, which scriptional patterns and aberrations in the DNA
contain a prominent sarcomatous element; giant- copy number that differ markedly from those of
cell glioblastomas, which have multinucleated primary glioblastomas.18,22 Despite their genetic
giant cells; small-cell glioblastomas, which are differences, primary and secondary glioblastomas
associated with amplification of the epidermal are morphologically indistinguishable and respond
growth factor receptor (EGFR); and glioblasto- similarly to conventional therapy, but they may re-
mas with oligodendroglial features, which may be spond differently to targeted molecular therapies.
associated with a better prognosis than standard High-grade oligodendrogliomas are character-
glioblastomas.1,12 Oligodendrogliomas are divided ized by the loss of chromosomes 1p and 19q (in
by the WHO into two grades: well-differentiated 50 to 90% of patients).1 Progression from low-
oligodendrogli­omas and oligoastrocytomas (WHO grade to anaplastic oligodendroglioma is associ-
grade II), and anaplastic oligodendrogliomas and ated with defects in PTEN, Rb, p53, and cell-
anaplastic oligoastrocytomas (WHO grade III) cycle pathways.1
(Fig. 1). All of these tumors may contain perinu-
clear halos (Fig. 2C) and a delicate network of Deregulated Growth Factor Signaling
branching blood vessels (chicken-wire pattern).1 The most common defects in growth-factor sig-
Malignant gliomas typically contain both neo- naling involve EGFR and PDGFR (Fig. 3).18 Am-
plastic and stromal tissues, which contribute to plification of EGFR occurs almost exclusively in
their histologic heterogeneity and variable out- primary glioblastomas and is seen in approximate-
come. Molecular studies such as gene-expression ly 40 to 50% of patients with that type of tumor.
profiling potentially allow for better classification About half of the tumors with EGFR amplifica-
of these tumors and separation of the tumors tion express a constitutively autophosphorylated
into different prognostic groups.13-17 variant of EGFR, known as EGFRvIII, that lacks
the extracellular ligand-binding domain (exons 2
MOL ECUL A R PATHO GENE SIS through 7).14,18 This characteristic variant has
become an important therapeutic target for ki-
Recently, there has been important progress in our nase inhibitors, immunotoxins, and peptide vac­
understanding of the molecular pathogenesis of cines.3,18 Recently, activating mutations in the ex-
malignant gliomas, and especially the importance tracellular domain of EGFR have been identified.25
of cancer stem cells.18,19 Malignant transforma- PDGF signaling is a key regulator of glial devel­
tion in gliomas results from the sequential accu- opment,26 and both ligand and receptors are fre-
mulation of genetic aberrations and the deregu- quently expressed in gliomas, creating an auto-
lation of growth-factor signaling pathways (Fig. 1 crine loop that stimulates proliferation of the

n engl j med 359;5  www.nejm.org  july 31, 2008 493

Downloaded from www.nejm.org on December 12, 2008 . Copyright © 2008 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

Cell-of-Origin: Differentiated Glial or Stem or Progenitor Cells

Olig2 expression (100%) Olig2 expression (100%) Olig2 expression (100%)

P53 mutated (>65%) EGFR amplified (~40%) LOH 1p, 4q, 19q
PDGFA/PDGFR-α overexpressed (~60%) EGFR overexpressed (~60%) EGFR overexpressed
EGFR mutated (~20–30%) PDGF/PDGFR overexpressed
MDM2 amplified (~10%)
Low-Grade Astrocytoma (5–10 yr)*
MDM2 overexpressed (>50%)
(WHO Grade II) Low-Grade Oligodendroglioma (5–10 yr)*
LOH 10q (~70%)
LOH 19q (~50%) (WHO Grade II)
P16Ink4a/P14ARF loss (~30%)
RB mutated (~25%) P16Ink4a/P14ARF loss
PTEN mutated (~40%)
CDK4 amplified (15%) RB mutated (~65%)
PI3K mutated/amplified (~20%)
MDM2 overexpressed (10%) p53 mutated
RB mutated
P16Ink4a/P14ARF loss (4%) PTEN loss
VEGF overexpressed
LOH 11p (~30%) LOH 9p, 10q
Anaplastic Astrocytoma (2–3 yr)* CDK4/EGFR/MYC amplified
(WHO Grade III) VEGF overexpressed
LOH 10q (~70%)
DCC loss (~50%) Anaplastic Oligodendroglioma (3–5 yr)*
PDGFR-α amplified (~10%)
(WHO Grade III)
PTEN mutated (~10%)
PI3K mutated/amplified (~10%)
VEGF overexpressed
Secondary Glioblastoma (12–15 mo)* Primary Glioblastoma (12–15 mo)*
(WHO Grade IV) (WHO Grade IV)

Figure 1. Pathways in the Development of Malignant Gliomas.

AUTHOR: Wen RETAKE 1st
Genetic and chromosomal alterations involved ICMin the development of the three main types of malignant gliomas (primary and secondary
FIGURE: 2nd
glioblastomas and anaplastic oligodendroglioma)
REG F are shown.1 of 5
Oligodendrocyte transcription factor
3rd 2 (Olig2) (blue) and vascular endo­
thelial growth factor (VEGF) (red) are expressed
CASE in all high-grade gliomas. Median lengths
Revised of survival (asterisks) are shown. A slash
EMaildeleted in colorectalLine
indicates one or the other or both. DCC denotes 4-C EGFR epidermal
carcinoma, SIZE growth factor receptor, LOH loss of
ARTIST: ts
heterozygosity, MDM2 murine double minuteEnon 2, PDGF platelet-derivedH/Tgrowth H/T
factor, PDGFR platelet-derived growth factor receptor,
Combo 39p6
PI3K phosphatidylinositol 3-kinase, PTEN phosphatase and tensin homologue, and RB retinoblastoma.
AUTHOR, PLEASE NOTE:
Figure has been redrawn and type has been reset.
Please check carefully.

tumor.18 Growth factor–receptor

JOB: 35905 signaling, through
intermediate signal-transduction generators, re-
sults in the activation of transcriptional programs
for survival, proliferation, invasion, and angio-
genesis. Common signal-transduction pathways
activated by these growth factors are the Ras–­
mitogen-activated protein (MAP) kinase path-
way, which is involved in proliferation and cell-
cycle progression, and the phosphatidylinositol
3-kinase (PI3K)–Akt–mammalian target of rapa­
mycin (mTOR) pathways, which are involved in the
inhibition of apoptosis and cellular proliferation
(Fig. 3).18 PTEN, a tumor-suppressor gene that neg-
atively regulates the PI3K pathway, is inactivated
in 40 to 50% of patients with glio­blastomas.18,27
Many of the above pathways lead to the up-
regulation of vascular endothelial growth factor
(VEGF) and angiogenesis.28,29 EGFR, PDGFR, and
VEGF-receptor (VEGFR) pathways also play an
important role in the normal development of the
ISSUE: 7-31-08
nervous system by promoting the proliferation of
multipotent stem cells. Other developmental path-
ways that contribute to the biologic features of
gliomas are those involving sonic hedgehog, wing-
less, Notch, CXC chemokine receptor 4 (CXCR4),
and bone morphogenetic proteins.19 In addition,
oligodendrocyte transcription factor 2 (Olig2),
a developmentally regulated, lineage-restricted
neural transcription factor, is a universal marker
of diffuse gliomas and stem cells that may be
a prerequisite for early transformation.30 Drugs
that target these pathways are under active in-
vestigation as treatment for gliomas.
Role of Stem Cells in Pathogenesis
and Resistance to Therapy
Although the genetic and signaling pathways in-
volved in the development of malignant gliomas

494 n engl j med 359;5  www.nejm.org  july 31, 2008

Downloaded from www.nejm.org on December 12, 2008 . Copyright © 2008 Massachusetts Medical Society. All rights reserved.
 Medical Progress

A B

*
*

C D

E F

Figure 2. Pathological Features of Malignant Gliomas.

Panels A and B show the histologic appearance of a glioblastoma, characterized by nuclear pleomorphism, dense cellu-
larity, and pseudopalisading necrosis (asterisk) (Panel A, hematoxylin and eosin) as well as vascular endothelial prolifera-
tion (asterisk) and mitotic figures (arrows)
ICM (Panel
AUTHORB, hematoxylin
wen and eosin).RETAKE
Panels C and
1st D show the histologic features
of an anaplastic oligodendroglioma, including the typical
REG F FIGURE perinuclear halo (“fried egg”)2nd
2a-f appearance (Panel C, hematoxy-
3rd
lin and eosin) and diffuse Olig2 staining
CASE(Panel D, brown color). The proliferation
TITLE index
Revised
can be quantified by immunohis-
EMail
tochemical analysis with the use of Ki67 staining (Panel E, black color), 4-C
Line and heterogeneous EGFR amplification by colori-
Enon than
metric in situ hybridization showing more two signals
mst (brown SIZE
ARTIST: H/T spots in nuclei)
H/T in almost all tumor cells (Panel F).
FILL
(Courtesy of Ali G. Saad, M.D., Department Combo
of Pathology, Brigham and Women’s33p9Hospital.)
AUTHOR, PLEASE NOTE:
Figure has been redrawn and type has been reset.
Please check carefully.
have been relatively well characterized, the cellu- differentiation into distinctive mature cell types.31
JOB: 35905 ISSUE: 7-31-08
lar origins of these tumors are unknown. The There is increasing evidence that neural stem cells,
adult nervous system harbors neural stem cells or related progenitor cells, can be transformed into
that are capable of self-renewal, proliferation, and cancer stem cells and give rise to malignant gliomas

n engl j med 359;5  www.nejm.org  july 31, 2008 495

Downloaded from www.nejm.org on December 12, 2008 . Copyright © 2008 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

by escaping the mechanisms that control prolif-

Figure 3 (facing page). Major Signaling Pathways
eration and programmed differentiation (Fig. in Malignant Gliomas and the Corresponding Targeted
4).32-36 These stem cells are identified by several Agents in Development for Glioblastoma.
immunocytochemical markers, such as CD133, RTK inhibitors that target epidermal growth factor
a glycoprotein also known as prominin 1.26,32,35,37 (EGF) receptor include gefitinib, erlotinib, lapatinib,
Although stem cells account for only a minority BIBW2992, and vandetanib; those that target platelet-
derived growth factor (PDGF) receptor include ima-
of the cells within malignant gliomas, they ap-
tinib, dasatinib, and tandutinib; those that target vas-
pear to be critical for generating these tumors.36,38 cular endothelial growth factor (VEGF) receptor
Recent studies suggest that glioma stem cells pro- include cediranib, pazopanib, sorafenib, sunitinib, vata-
duce VEGF and promote angiogenesis in the tu- lanib, vandetanib, and XL184. EGF receptor antibodies
mor microenvironment.39 In addition, tumor stem include cetuximab and panitumumab. Farnesyl trans-
ferase inhibitors include lonafarnib and tipifarnib;
cells appear to require a vascular niche for opti-
HDAC inhibitors include depsipeptide, vorinostat, and
mal function.40 These observations raise the pos- LBH589; PI3K inhibitors include BEZ235 and XL765;
sibility that antiangiogenic therapy may inhibit mTOR inhibitors include sirolimus, temsirolimus,
the functioning of glioma stem cells. everolimus, and deforolimus; and VEGF receptor inhib-
There is growing evidence that glioma stem itors include bevacizumab, aflibercept (VEGF-trap),
and CT-322. Growth factor ligands include EGF, PDGF,
cells may contribute to the resistance of malig-
IGF, TGF, HGF/SF, VEGF, and FGF. Stem-cell pathways
nant gliomas to standard treatments (Fig. 4). include SHH, wingless family, and Notch. Akt denotes
Radioresistance in stem cells generally results murine thymoma viral oncogene homologue (also
from the preferential activation of DNA-damage- known as protein kinase B), CDK cyclin-dependent ki-
response pathways,41 whereas chemoresistance re- nase, ERK extracellular signal-regulated kinase, FGF fi-
broblast growth factor, FTI farnesyl transferase inhibi-
sults partly from the overexpression of O6-meth­
tors, GDP guanine diphosphate, Grb 2 growth factor
ylguanine−DNA methyltransferase (MGMT), the receptor-bound protein 2, GTP guanine triphosphate,
up-regulation of multidrug resistance genes, and HDAC histone deacetylase, HGF/SF hepatocyte growth
the inhibition of apoptosis.42-44 Therapeutic strat- factor/scatter factor, IGF insulin-like growth factor,
egies that effectively target stem cells and over- MEK mitogen-activated protein kinase kinase, mTOR
mammalian target of rapamycin, NF1 neurofibromin 1,
come their resistance to treatment will be neces-
PIP2 phosphatidylinositol (4,5) biphosphate, PIP3
sary if malignant gliomas are to be completely phosphatidylinositol 3,4,5-triphosphate, PI3K phos-
eradicated (Fig. 4). A better understanding of the phatidylinositol 3-kinase, PKC protein kinase C, PLC
biologic differences between normal and cancer phospholipase C, PTEN phosphatase and tensin ho-
stem cells will be required to develop selective mologue, RAF v-raf 1 murine leukemia viral oncogene
homologue 1, RAS rat sarcoma viral oncogene homo-
therapies that spare normal brain cells.
logue, RTK receptor tyrosine kinase inhibitor, SHH
sonic hedgehog, SOS son of sevenless, Src sarcoma
DI AGNOSIS (Schmidt-Ruppin A-2) viral oncogene homologue, TGF
transforming growth factor family, and TSC1 and 2 tu-
Clinical Presentation berous sclerosis gene 1 and 2. Red text denotes inhibi-
tors. Data are from Sathornsumetee et al.,3 Furnari
Patients with malignant gliomas may present with
et al.,18 Chi and Wen,20 and Sathornsumetee et al.21
a variety of symptoms, including headaches, sei-
zures, focal neurologic deficits, confusion, mem-
typically show a heterogeneously enhancing mass
ory loss, and personality changes. Although the
with surrounding edema. Glioblastomas frequent-
classic headaches that are suggestive of increased
ly have central areas of necrosis and more exten-
intracranial pressure are most severe in the morn-
ing and may wake the patient from sleep, many sive peritumoral edema than that associated with
anaplastic gliomas.45 Functional MRI may help
patients experience headaches that are indistin-
guishable from tension headaches. When severe,define the relationship of speech and motor ar-
eas to the tumor and aid in the planning of sur-
the headaches may be associated with nausea and
vomiting. gery. Diffusion-weighted imaging, diffusion ten-
sor imaging, dynamic contrast-enhanced MRI to
Imaging measure vessel permeability, and perfusion im-
The diagnosis of malignant gliomas is usually aging to measure relative cerebral blood volume
suggested by magnetic resonance imaging (MRI) are increasingly used as diagnostic aids and as a
or computed tomography. These imaging studies means of monitoring the response to therapy.46

496 n engl j med 359;5  www.nejm.org  july 31, 2008

Downloaded from www.nejm.org on December 12, 2008 . Copyright © 2008 Massachusetts Medical Society. All rights reserved.
 Medical Progress

Proton magnetic resonance spectroscopy detects
the levels of metabolites and may help differenti-
ate a tumor from necrosis or benign lesions. In
patients with malignant gliomas, this imaging
technique typically shows an increase in the cho-
line peak (reflecting increased membrane turn-
over) and a decrease in the N-acetyl aspartate peak
(reflecting decreased neuronal cellularity), as com-
pared with the findings in unaffected areas of the
brain.45,46 Positron-emission tomography that uses

n engl j med 359;5  www.nejm.org  july 31, 2008 497

Downloaded from www.nejm.org on December 12, 2008 . Copyright © 2008 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

Figure 4. Resistance Mechanisms in Glioma Cells.

Normal neural stem cells self-renew and give rise to multipotential progenitor cells that form neurons, oligodendro-
glia, and astrocytes. Glioma stem cells arise from the transformation of either neural stem cells or progenitor cells
(red) or, less likely, from differentiation of a oligodendrocytes or astrocytes (thin red arrows) and lead to malignant
gliomas. Glioma stem cells are relatively resistant to standard treatments such as radiation and chemotherapy and
lead to regrowth of the tumor after treatment. Therapies directed at stem cells can deplete these cells and poten­
tially lead to more durable tumor regression (blue).

isotopes such as 18F-fluorodeoxyglucose, 18F-flu-
oro-l-thymidine, 11C-methionine, and 3,4-dihy-
droxy-6-18F-fluoro-l-phenylalanine is being eval-
uated for its usefulness in diagnosis and in mon-
­itoring the response to therapy.47
In up to 40% of cases, the MRI studies that
are performed in the first month after radiothera-
py show increased enhancement.48 In 50% of these
cases, the increased enhancement reflects a tran-
sient increase in vessel permeability as a result of
radiotherapy, a phenomenon termed “pseudopro-
gression,” which improves with time.48 Differen-

498 n engl j med 359;5  www.nejm.org  july 31, 2008

Downloaded from www.nejm.org on December 12, 2008 . Copyright © 2008 Massachusetts Medical Society. All rights reserved.
 Medical Progress

tiating this transient effect from true progres- of 20 to 30%.49,53 The risk of intratumoral hem-
sion of the cancer can be challenging initially, orrhage associated with anticoagulation therapy
even with advanced imaging techniques. in patients with gliomas who have venous throm-
boembolism is low,49,54 whereas inferior vena cava
T R E ATMEN T filters are associated with high complication
rates.55 Unless a patient with malignant glioma
General Medical Management and venous thromboembolism has an intracere-
Much of the care of patients with malignant bral hemorrhage or other contraindications, it is
gliomas involves general medical management. generally safe to provide anticoagulation therapy
The most common problems include seizures, per- for the venous thromboembolism. Low-molecular-
itumoral edema, venous thromboembolism, fa- weight heparin may be more effective and safer
tigue, and cognitive dysfunction.49 Patients who than warfarin.56
present with seizures should be treated with an- Patients with malignant gliomas frequently ex­
tiepileptic drugs. Since antiepileptic drugs that perience fatigue and may benefit from treatment
induce hepatic cytochrome P-450 enzymes, such with modafinil or methylphenidate.57 Methyl­
as phenytoin and carbamazepine, increase the me- phenidate may also help abulia, and donepezil58
tabolism of many chemotherapeutic agents, anti- and memantine may reduce memory loss, although
epileptic drugs that do not induce these enzymes, evidence supporting these approaches remains
such as levetiracetam, are generally preferred. The limited. Depression is underdiagnosed in patients
use of prophylactic antiepileptic drugs in patients with malignant gliomas, and antidepressants and
with malignant gliomas who have never had a psychiatric support are often invaluable.59
seizure is controversial. The American Academy of
Neurology issued a practice guideline indicating Specific Therapy for Newly Diagnosed
that there is no evidence that prophylactic anti- Malignant Gliomas
epileptic drugs are beneficial and advises against The standard therapy for newly diagnosed malig-
the routine use of antiepileptic drugs in patients nant gliomas involves surgical resection when fea-
with brain tumors who have not had seizures.50 sible, radiotherapy, and chemotherapy (Table 1).
Corticosteroids such as dexamethasone are fre- Malignant gliomas cannot be completely eliminat-
quently used to treat peritumoral edema. Cush- ed surgically because of their infiltrative nature,
ing’s syndrome and corticosteroid myopathy may but patients should undergo maximal surgical
develop in patients who require prolonged treat- resection whenever possible. Surgical debulking
ment with high doses of corticosteroids. Patients reduces the symptoms from mass effect and pro-
with brain tumors who receive corticosteroids are vides tissue for histologic diagnosis and molecu-
at increased risk for Pneumocystis jiroveci pneumoni- lar studies. Advances such as MRI-guided neuro-
tis, and prophylactic antibiotic therapy should be navigation, intraoperative MRI, functional MRI,
considered,49 although a recent meta-analysis did intraoperative mapping,60 and fluorescence-guid-
not show a benefit from this approach.51 As the ed surgery61 have improved the safety of surgery
rate of survival among patients with malignant and increased the extent of resection that can be
glioma improves, long-term complications from achieved. The value of surgery in prolonging sur-
treatment with corticosteroids, including osteo- vival is controversial, but patients who undergo
porosis and compression fractures, are becoming extensive resection probably have a modest sur-
increasingly evident, and preventive measures, vival advantage.60-62 Stereotactic biopsies should
such as treatment with vitamin D, calcium sup- be performed only in patients who have inoper-
plements, and bisphosphonates, should be con- able tumors that are located in critical areas.
sidered. Novel therapies such as corticotropin- Radiotherapy is the mainstay of treatment for
releasing factor, bevacizumab (a humanized VEGF malignant gliomas. The addition of radiotherapy
monoclonal antibody), and VEGFR inhibitors de- to surgery increases survival among patients with
crease peritumoral edema and may reduce the glioblastomas from a range of 3 to 4 months to
need for corticosteroids.49,52 a range of 7 to 12 months.63,64 Conventional ra-
Patients with malignant gliomas are at in- diotherapy consists of 60 Gy of partial-field exter-
creased risk for venous thromboembolism from nal-beam irradiation delivered 5 days per week
leg and pelvic veins, with a cumulative incidence in fractions of 1.8 to 2.0 Gy. After standard radio-

n engl j med 359;5  www.nejm.org  july 31, 2008 499

Downloaded from www.nejm.org on December 12, 2008 . Copyright © 2008 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

Table 1. Summary of Current Treatments for Malignant Gliomas.*

Type of Tumor Therapy

Newly diagnosed tumors
Glioblastomas (WHO grade IV)
Anaplastic astrocytomas (WHO
grade III)
Anaplastic oligodendrogliomas
and anaplastic oligoastrocy-
tomas (WHO grade III)
Recurrent tumors
Maximal surgical resection, plus radiotherapy, plus concomitant and adjuvant
TMZ or carmustine wafers (Gliadel)†
Maximal surgical resection, with the following options after surgery (no ac-
cepted standard treatment): radiotherapy, plus concomitant and adjuvant
TMZ or adjuvant TMZ alone†
Maximal surgical resection, with the following options after surgery (no ac-
cepted standard treatment): radiotherapy alone, TMZ or PCV with or with-
out radiotherapy afterward, radiotherapy plus concomitant and adjuvant
TMZ, or radiotherapy plus adjuvant TMZ†‡
Reoperation in selected patients, carmustine wafers (Gliadel), conventional
chemotherapy (e.g., lomustine, carmustine, PCV, carboplatin, irinotecan,
etoposide), bevacizumab plus irinotecan, experimental therapies‡

* Data are from Sathornsumetee et al.,3 Furnari et al.,18 Chi and Wen,20 and Sathornsumetee et al.21 PCV denotes procar-
bazine, lomustine (CCNU), and vincristine, and TMZ temozolomide.
† Radiotherapy is administered at a dose of 60 Gy given in 30 fractions over a period of 6 weeks. Concomitant TMZ is ad-
ministered at a dose of 75 mg per square meter of body-surface area per day for 42 days with radiotherapy. Beginning
4 weeks after radiotherapy, adjuvant TMZ is administered at a dose of 150 mg per square meter per day on days 1 to
5 of the first 28-day cycle, followed by 200 mg per square meter per day on days 1 to 5 of each subsequent 28-day cycle,
if the first cycle was well tolerated.
‡ PCV therapy consists of lomustine (CCNU), 110 mg per square meter, on day 1; procarbazine, 60 mg per square meter,
on days 8 to 21; and vincristine, 1.5 mg per square meter (maximum dose, 2 mg), on days 8 and 29.

therapy, 90% of the tumors recur at the original
site.65 Strategies to increase the radiation dose to
the tumor with the use of brachytherapy 66 and
stereotactic radiosurgery 67,68 have failed to im-
prove survival. Newer chemotherapeutic agents,69
targeted molecular agents,20 and antiangiogenic
agents70 may enhance the effectiveness of radio-
therapy.
Patients who are older than 70 years of age have
a worse prognosis than younger patients and rep-
resent a particular challenge. Among these pa-
tients, radiotherapy produces a modest benefit in
median survival (29.1 weeks) as compared with
supportive care (16.9 weeks).71 Since older pa-
tients often tolerate radiotherapy less well than
younger patients, an abbreviated course of radio-
therapy (40 Gy in 15 fractions over a period of
3 weeks)72 or chemotherapy with temozolomide
(an oral alkylating agent with good penetration
of the blood–brain barrier) alone73 may be con-
sidered, since the outcomes with these approach-
es are similar to the outcomes with conventional
radiotherapy regimens.
Chemotherapy is assuming an increasingly im-
portant role in the treatment of malignant gliomas.
Although early studies of adjuvant chemotherapy
for malignant gliomas with the use of nitroso­
ureas failed to show a benefit,63,74 two meta-
analyses have suggested that adjuvant chemother­
apy results in a modest increase in survival (a 6 to
10% increase in the 1-year survival rate).75,76
The European Organisation for Research and
Treatment of Cancer (EORTC) and the National
Cancer Institute of Canada (NCIC) conducted a
phase III trial comparing radiotherapy alone (60 Gy
over a period of 6 weeks) with radiotherapy and
concomitant treatment with temozolomide (75 mg
per square meter of body-surface area per day
for 6 weeks), followed by adjuvant temozolomide
therapy (150 to 200 mg per square meter per day
for 5 days every 28 days for 6 cycles), in patients
with newly diagnosed glioblastomas.64 As reported
by Stupp et al., the combination of radiotherapy
and temozolomide had an acceptable side-effect
profile and, as compared with radiotherapy alone,
increased the median survival (14.6 months vs.
12.1 months, P<0.001).64 In addition, the survival
rate at 2 years among the patients who received
radiotherapy and temozolomide was significantly
greater than the rate among the patients who
received radiotherapy alone (26.5% vs. 10.4%),64
establishing radiotherapy with concomitant and
adjuvant temozolomide as a useful combination
for newly diagnosed glioblastomas.
MGMT is an important repair enzyme that con-
tributes to resistance to temozolomide. In a com-
panion study to the EORTC–NCIC study reported
by Stupp et al., tumor specimens from the pa-

500 n engl j med 359;5  www.nejm.org  july 31, 2008

Downloaded from www.nejm.org on December 12, 2008 . Copyright © 2008 Massachusetts Medical Society. All rights reserved.
 Medical Progress
tients were examined for epigenetic silencing of
the MGMT gene.15 MGMT promoter methylation
silences the gene, thus decreasing DNA repair ac-
tivity and increasing the susceptibility of the tu-
mor cells to temozolomide. Patients with glioblas-
toma and MGMT promoter methylation (45% of the
total) who were treated with temozolomide had
a median survival of 21.7 months and a 2-year
survival rate of 46%. In contrast, patients without
MGMT promoter methylation who were treated
with temozolomide had a significantly shorter
median survival of only 12.7 months and a 2-year
survival rate of 13.8%.15 Currently, temozolomide
is used in the treatment of glioblastomas regard-
less of MGMT promoter methylation status. How-
ever, if the importance of MGMT promoter methy-
lation is confirmed by the results of an ongoing
study by the Radiation Therapy Oncology Group
(RTOG 0525), patients with unfavorable MGMT
methylation status may be selected for other
treatments in future investigations. Studies of
dose-intensive temozolomide regimens to deplete
MGMT and of combinations of temozolomide
with inhibitors of MGMT, such as O6-benzylgua-
nine, and inhibitors of other repair enzymes, such
as poly-(ADP-ribose)-polymerase, are in progress.
Another chemotherapeutic approach involves
the implantation of biodegradable polymers con-
taining carmustine (Gliadel Wafers, MGI Pharma)
into the tumor bed after resection of the tumor.
The aim of treatment with these polymers, which
release carmustine gradually over the course of
several weeks, is to kill residual tumor cells. In
a randomized, placebo-controlled trial that inves-
tigated the use of these polymers in patients with
newly diagnosed malignant gliomas, median sur-
vival increased from 11.6 months to 13.9 months
(P = 0.03).77 This survival advantage was main-
tained at 2 and 3 years.78
Therapy for Anaplastic Gliomas
Anaplastic astrocytomas are treated with radio-
therapy and either concurrent and adjuvant temo-
zolomide (as for glioblastomas) or adjuvant temo-
zolomide alone. Currently, there are no findings
from controlled trials that support the use of con-
current temozolomide in patients with anaplastic
astrocytomas.
Anaplastic oligodendrogliomas and anaplas-
tic oligoastrocytomas are an important subgroup
of malignant gliomas that are generally more re-
sponsive to therapy than are pure astrocytic tu-
mors.79 A codeletion of chromosomes 1p and
n engl j med 359;5  www.nejm.org  july 31, 2008
Downloaded from www.nejm.org on December 12, 2008 . Copyright © 2008 Massachusetts Medical Society. All rights reserved.
19q,79 mediated by an unbalanced translocation
of 19p to 1q,80 occurs in 61 to 89% of patients with
anaplastic oligodendrogliomas and 14 to 20% of
patients with anaplastic oligoastrocytomas. Tu-
mors in patients with the 1p and 19q codeletion
are particularly sensitive to chemotherapy with
PCV — procarbazine, lomustine (CCNU), and vin-
cristine — with response rates of up to 100%, as
compared with response rates of 23 to 31% among
patients without the deletion of chromosomes 1p
and 19q.81,82 The reason for the increased chemo-
sensitivity of tumors in patients with the 1p and
19q codeletion is unclear. One study suggested
that 1p loss is associated with decreased levels of
stathmin and an increased sensitivity to nitroso­
ureas.83 The status of chromosomes 1p and 19q,
rather than standard histologic assessment, is now
used as an eligibility criterion in studies involv-
ing patients with anaplastic oligodendrogliomas
and anaplastic oligoastrocytomas, reflecting a par-
adigm shift in the design of clinical trials for pa-
tients with these tumors.
Two large phase III studies of PCV chemother-
apy with radiotherapy, as compared with radio-
therapy alone, in patients with newly diagnosed
anaplastic oligodendrogliomas or anaplastic oli-
goastrocytomas, have been reported.84,85 In both
studies, the addition of chemotherapy to radio-
therapy increased the time to tumor progression
by 10 to 12 months but did not improve overall
survival (median, 3.4 and 4.9 years).84,85 The fail-
ure of chemotherapy to increase survival may be
partly explained by the fact that patients who ini-
tially received radiotherapy alone subsequently re-
ceived chemotherapy when they had a relapse, so
that most patients in both groups eventually re-
ceived chemotherapy. In both studies, patients
with the codeletion of 1p and 19q had improved
survival as compared with those without the code-
letion of 1p and 19q. Although most studies
involving patients with anaplastic oligodendro-
gliomas or anaplastic oligoastrocytomas were con-
ducted with PCV chemotherapy, temozolomide is
likely to have similar activity and less toxicity79;
however, studies directly comparing the two regi-
mens have not been performed.
Therapy for Recurrent Malignant Gliomas
Despite optimal treatment, nearly all malignant
gliomas eventually recur. For glioblastomas, the
median time to progression after treatment with
radiotherapy and temozolomide is 6.9 months.64
If the tumor is symptomatic from mass effect, re-
501
 The n e w e ng l a n d j o u r na l
operation may be indicated (Table 1). However, sur-
gery performed in selected patients results in only
limited prolongation of survival.86
The usefulness of radiotherapy for recurrent
malignant gliomas is controversial.87 Although
some reports have suggested that fractionated ste-
reotactic reirradiation88 and stereotactic radiosur-
gery68 may be beneficial, selection bias may have
influenced these results.
The value of conventional chemotherapy for re-
current malignant gliomas is modest. In general,
chemotherapy is more effective for anaplastic
gliomas than for glioblastomas.79,87 Temozolo-
mide was evaluated in a phase II study involving
patients with recurrent anaplastic gliomas who
had previously been treated with nitrosoureas; the
study showed a 35% response rate. The 6-month
rate of progression-free survival was 46%,89 com-
paring favorably with the 31% rate of progression-
free survival at 6 months for therapies that were
reported to be ineffective.90 In contrast, temozolo-
mide has only limited activity in patients with
recurrent glioblastomas (response rate, 5.4%;
6-month rate of progression-free survival, 21%).91
Other chemotherapeutic agents that are used for
recurrent gliomas include nitrosoureas, carbo-
platin, procarbazine, irinotecan, and etoposide.
Carmustine wafers have modest activity, increas-
ing the median survival by approximately 8 weeks
in patients with recurrent glioblastomas.92
IN V E S T IG AT IONA L THER A PIE S
Targeted Molecular Therapies
The improved understanding of the molecular
pathogenesis of malignant gliomas has allowed
a more rational use of targeted molecular thera-
pies (Fig. 3).18,20,21 Particular interest has focused
on inhibitors that target receptor tyrosine kinases
such as EGFR,93 PDGFR,94 and VEGFR,52 as well
as on signal-transduction inhibitors targeting
mTOR,95,96 farnesyltransferase,97 and PI3K (Ta-
ble 2). Single agents have only modest activity,
with response rates of 0 to 15% and no prolonga-
tion of 6-month progression-free survival.3,20,21
These disappointing results are due to several fac-
tors. Most malignant gliomas have coactivation of
multiple tyrosine kinases,98 as well as redundant
signaling pathways, thus limiting the activity of
single agents. In addition, many of these agents
502 n engl j med 359;5  www.nejm.org  july 31, 2008
Downloaded from www.nejm.org on December 12, 2008 . Copyright © 2008 Massachusetts Medical Society. All rights reserved.
of m e dic i n e
have poor penetration across the blood−tumor bar-
rier. There has been considerable interest in iden-
tifying molecular features of the tumor that pre-
dict a response, so that patients who are most
likely to benefit can be selected for a particular
treatment. EGFR inhibitors appear to be more ef-
fective in patients who have tumors with EGFRvIII
mutations and intact PTEN than in patients who
do not have these molecular changes99; patients
who have tumors with increased activity of the
PI3K–Akt pathway, as indicated by an increase in
phosphorylated Akt, generally do not have a re-
sponse.100 Current experimental strategies to in-
crease the effectiveness of targeted molecular ther-
apies include the use of a single agent targeted
against several kinases, combinations of agents
that inhibit complementary targets such as EGFR
and mTOR (Table 2 and Fig. 5A through 5D), and
targeted agents combined with radiotherapy and
chemotherapy.3,18,20,21
Antiangiogenic Agents
Malignant gliomas are among the most vascular
of human tumors,18 making them especially at-
tractive targets for angiogenesis inhibitors.29 Al-
though older antiangiogenic agents such as tha-
lidomide had only modest activity,101 newer and
more potent angiogenesis inhibitors show prom-
ising activity. In preliminary studies, treatment
with the combination of bevacizumab and irino-
tecan was associated with a low incidence of hem-
orrhage and response rates of 57 to 63% among
patients with malignant gliomas (Fig. 5E through
5H).102,103 Some of the improvement that is seen
on radiographic images may be artifactual, caused
by reduced vascular permeability and decreased
contrast enhancement as a result of the inhibi-
tion of VEGF. However, this regimen also has
antitumor activity, as evidenced by the fact that it
increased the 6-month rate of progression-free
survival to 46% among patients with recurrent
glioblastomas,102,103 as compared with a 6-month
rate of progression-free survival of 21% for pa-
tients who were receiving treatment with temozo-
lomide.91 Recently, a large, randomized phase II
trial of bevacizumab alone and bevacizumab with
irinotecan was completed. Preliminary results con-
firmed the safety of bevacizumab and showed an
increase in the 6-month rate of progression-free
survival to 35.1% for patients receiving bevacizu­
 Medical Progress

Table 2. Selected Investigational Treatments for Malignant Gliomas.*

Type of Treatment Example

Convection-enhanced surgical delivery of pharmacologic Cintredekin besudotox
agent
Drugs to overcome resistance to TMZ
Dose-dense TMZ
MGMT inhibitors O6-benzylguanine
PARP inhibitors BSI-201, ABT-888
New chemotherapies RTA744, ANG1005
Antiangiogenic therapies
Anti-αvβ5 integrins Cilengitide
Anti-hepatocyte growth factor AMG-102
Anti-VEGF Bevacizumab, aflibercept (VEGF-Trap)
Anti-VEGFR Cediranib, pazopanib sorafenib, sunitinib, vandetinib,
vatalanib, XL184, CT-322
Other agents Thalidomide
Targeted molecular therapies
Akt Perifosine
EGFR inhibitors Erlotinib, gefitinib, lapatinib, BIBW2992, nimotuzumab,
cetuximab
FTI inhibitors Tipifarnib, lonafarnib
HDAC inhibitors Vorinostat, depsipeptide, LBH589
HSP90 inhibitors ATI3387
Met XL184
mTOR inhibitors Everolimus, sirolimus, temsirolimus, deforolimus
PI3K inhibitors BEZ235, XL765
PKCβ Enzastaurin
PDGFR inhibitors Dasatinib, imatinib, tandutinib
Proteasome Bortezomib
Raf Sorafenib
Src Dasatinib
TGF-β AP12009
Combination therapies Erlotinib plus temsirolimus, gefitinib plus everolimus,
gefitinib plus sirolimus, sorafenib plus temsirolimus,
erlotinib, or tipifarnib, pazopanib plus lapatinib
Immunotherapies
Dendritic cell and EGFRvIII peptide vaccines DCVax, CDX-110
131I-anti-tenascin
Monoclonal antibodies antibody
Gene therapy
131I-TM-601
Other therapies

* Data are from Sathornsumetee et al.,3 Furnari et al.,18 Chi and Wen,20 and Sathornsumetee et al.21 EGFR denotes
­epidermal growth factor receptor, FTI farnesyltransferase, HDAC histone deacetylase, HSP90 heat-shock protein 90,
MGMT O6-methylguanine−DNA methyltransferase, mTOR mammalian target of rapamycin, PARP poly (ADP-ribose)
polymerase, PDGFR platelet-derived growth factor receptor, PI3K phosphatidylinositol 3-kinase, PKCβ protein kinase C β,
TGF transforming growth factor, TMZ temozolomide, and VEGFR vascular endothelial growth factor receptor.

n engl j med 359;5  www.nejm.org  july 31, 2008 503

Downloaded from www.nejm.org on December 12, 2008 . Copyright © 2008 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

A B E F

C D G H

Figure 5. MRI Scans Showing Responses to Targeted Agents.

Panels A through D show MRI scans in a patient with a recurrent malignant glioma who was treated with a combi-
nation of erlotinib (an inhibitor of epidermal growth factor receptor [EGFR]) and sirolimus (an inhibitor of the mam-
malian target of rapamycin [mTOR]). ICM AUTHORimages
T1-weighted wen obtained after the administration
RETAKE 1st of gadolinium show a re-
duction in the size of the enhancingREG tumor pretreatment image (Panel A) 2nd
from the 5a-h
F FIGURE to the image obtained 2 months
CASE 3rd
after treatment (Panel B), with fluid-attenuated
TITLE inversion recovery (FLAIR) Revised
studies showing a reduction of edema
EMail
from the pretreatment image (Panel C) to the post-treatment Lineimage4-C (Panel D). Panels E through H show MRI scans
Enon whoARTIST: mst with SIZE
of a recurrent glioblastoma in a patient was treated H/Ta combination
H/T of bevacizumab and irinotecan. T1-
FILL Combo 33p9
weighted images obtained after the administration of gadolinium show a reduction in the size of the enhancing tu-
mor from the image obtained before treatment AUTHOR,(Panel E)PLEASE NOTE: obtained 7 months after treatment (Panel F)
to the image
and an associated reduction of edemaFigure from has
thebeen redrawn andFLAIR
pretreatment type has been (Panel
image reset. G) to the post-treatment FLAIR
Please check carefully.
image (Panel H).
JOB: 35905 ISSUE: 7-31-08

mab alone and 50.2% for patients receiving the cross the blood−tumor barrier more effectively,
combination of bevacizumab and irinotecan.104 gene therapy,105 peptide and dendritic-cell vac­
A phase II trial of the pan-VEGFR inhibitor ce­ cines,106 radiolabeled monoclonal antibodies
dira­nib in patients with recurrent glioblastomas against the extracellular matrix protein tena­
showed response rates in excess of 50% and pro- scin,107 synthetic chlorotoxins (131I-TM-601),108
longation of the 6-month rate of progression-free and infusion of radiolabeled drugs and target­
survival to approximately 26%.52 These agents also ed toxins into the tumor and surrounding brain
decrease peritumoral edema, potentially allowing by means of convection-enhanced delivery (Ta-
for a reduction in corticosteroid requirements. ble 2).109
Since antiangiogenic agents may have synergistic
activity with radiotherapy, there is increasing in- PRO GNOS T IC FAC T OR S
terest in combining them with radiotherapy and
temozolomide in patients with newly diagnosed The most important adverse prognostic factors
glioblastomas.29,70 As noted previously, glioma
in patients with malignant gliomas are advanced
stem cells produce VEGF39 and require a vascular
age, histologic features of glioblastoma, poor Kar-
niche for optimal function.40 Antiangiogenic agents
nofsky performance status, and unresectable tu-
may therefore also target glioma stem cells. mor.110 There are ongoing efforts to identify bio-
logic and genetic alterations in the tumors that
Other Therapies may provide additional prognostic information, as
Other investigational therapies for malignant well as guidance in making decisions about opti-
gliomas include chemotherapeutic agents that mal therapy.15,16,82

504 n engl j med 359;5  www.nejm.org  july 31, 2008

Downloaded from www.nejm.org on December 12, 2008 . Copyright © 2008 Massachusetts Medical Society. All rights reserved.
 Medical Progress

Supported by grants from the National Institutes of Health

SUM M A R Y
Recently, there has been important progress in
the treatment of malignant gliomas111 and in our
understanding of the molecular pathogenesis of
these tumors and the critical role that stem cells
play in their development and resistance to treat-
ment. As our understanding of the molecular cor-
relates of response improves, it may be possible
to select the most appropriate therapies on the
basis of the patient’s tumor genotype. These ad-
vances provide real opportunities for the develop-
ment of effective therapies for malignant gliomas.
(U01 CA062407, to Dr. Wen; and KO8 CA1240804, to Dr. Kesari),
a Sontag Foundation Distinguished Scientist Grant (to Dr. Kesari),
and research support from the Elizabeth Atkins and Will Kraft
Brain Tumor Research Funds (to Dr. Wen) and the John Kenney
Brain Tumor Research Fund (to Dr. Kesari).
Dr. Wen reports receiving speaking fees from Schering-
Plough, serving as a consultant for AngioChem, and receiving
research funding from Exelixis, Schering-Plough, Genentech,
GlaxoSmithKline, Amgen, AstraZeneca, and Celgene. Dr. Kesari
reports receiving consulting fees from Bristol–Myers Squibb,
speaking fees from Enzon, and research funding from Adnexus.
No other potential conflict of interest relevant to this article was
reported.
We thank Drs. Andrew Norden and Elizabeth Claus for help-
ful comments.
This article is dedicated to the memories of Elizabeth Atkins,
Will Kraft, and John Kenney.

References
1. Louis DN, Ohgaki H, Wiestler OD, et
al. The 2007 WHO classification of tu-
mors of the central nervous system. Lyon,
France: IARC Press, 2007.
2. CBTRUS 2008 statistical report: pri-
mary brain tumors in the United States,
1998-2002. Central Brain Tumor Registry
of the United States, 2000-2004. (Accessed
July 7, 2008, at http://www.cbtrus.org/
reports/2007-2008/2007report.pdf.)
3. Sathornsumetee S, Rich JN, Reardon
DA. Diagnosis and treatment of high-
grade astrocytoma. Neurol Clin 2007;25:
1111-39.
4. Fisher JL, Schwartzbaum JA, Wrensch
M, Wiemels JL. Epidemiology of brain tu-
mors. Neurol Clin 2007;25:867-90.
5. Hardell L, Carlberg M, Soderqvist F,
Mild KH, Morgan LL. Long-term use of
cellular phones and brain tumours: in-
creased risk associated with use for > or =10
years. Occup Environ Med 2007;64:626-32.
6. Lahkola A, Auvinen A, Raitanen J, et
al. Mobile phone use and risk of glioma in
5 North European countries. Int J Cancer
2007;120:1769-75.
7. Inskip PD, Tarone RE, Hatch EE, et al.
Cellular-telephone use and brain tumors.
N Engl J Med 2001;344:79-86.
8. Linos E, Raine T, Alonso A, Michaud
D. Atopy and risk of brain tumors: a meta-
analysis. J Natl Cancer Inst 2007;99:1544-
50.
9. Wrensch M, Wiencke JK, Wiemels J, et
al. Serum IgE, tumor epidermal growth
factor receptor expression, and inherited
polymorphisms associated with glioma
survival. Cancer Res 2006;66:4531-41.
10. Farrell CJ, Plotkin SR. Genetic causes
of brain tumors: neurofibromatosis, tu-
berous sclerosis, von Hippel-Lindau, and
other syndromes. Neurol Clin 2007;25:
925-46.
11. Malmer B, Adatto P, Armstrong G, et
al. GLIOGENE, an international consor-
tium to understand familial glioma. Can-
cer Epidemiol Biomarkers Prev 2007;16:
1730-4.
12. Louis DN, Ohgaki H, Wiestler OD, et
al. The 2007 WHO classification of tu-
mours of the central nervous system. Acta
Neuropathol 2007;114:97-109. [Erratum,
Acta Neuropathol 2007;114:547.]
13. Nutt CL, Mani DR, Betensky RA, et al.
Gene expression-based classification of
malignant gliomas correlates better with
survival than histological classification.
Cancer Res 2003;63:1602-7.
14. Pelloski CE, Ballman KV, Furth AF, et
al. Epidermal growth factor receptor vari-
ant III status defines clinically distinct
subtypes of glioblastoma. J Clin Oncol
2007;25:2288-94.
15. Hegi ME, Diserens AC, Gorlia T, et al.
MGMT gene silencing and benefit from
temozolomide in glioblastoma. N Engl J
Med 2005;352:997-1003.
16. Phillips HS, Kharbanda S, Chen R, et
al. Molecular subclasses of high-grade
glioma predict prognosis, delineate a pat-
tern of disease progression, and resemble
stages in neurogenesis. Cancer Cell 2006;9:
157-73.
17. Collins VP. Mechanisms of disease:
genetic predictors of response to treat-
ment in brain tumors. Nat Clin Pract On-
col 2007;4:362-74.
18. Furnari FB, Fenton T, Bachoo RM, et
al. Malignant astrocytic glioma: genetics,
biology, and paths to treatment. Genes
Dev 2007;21:2683-710.
19. Lee da Y, Gutmann DH. Cancer stem
cells and brain tumors: uprooting the bad
seeds. Expert Rev Anticancer Ther 2007;7:
1581-90.
20. Chi AS, Wen PY. Inhibiting kinases in
malignant gliomas. Expert Opin Ther
Targets 2007;11:473-96.
21. Sathornsumetee S, Reardon DA, Des-
jardins A, Quinn JA, Vredenburgh JJ, Rich
JN. Molecularly targeted therapy for ma-
lignant glioma. Cancer 2007;110:13-24.
22. Ohgaki H, Kleihues P. Genetic path-
ways to primary and secondary glioblas-
toma. Am J Pathol 2007;170:1445-53.
23. Watanabe K, Tachibana O, Sata K,
Yonekawa Y, Kleihues P, Ohgaki H. Over-
expression of the EGF receptor and p53
mutations are mutually exclusive in the
evolution of primary and secondary
­ lioblastomas. Brain Pathol 1996;6:217-
g
23.
24. Ueki K, Ono Y, Henson JW, Efird JT,
von Deimling A, Louis DN. CDKN2/p16 or
RB alterations occur in the majority of
glioblastomas and are inversely correlat-
ed. Cancer Res 1996;56:150-3.
25. Lee JC, Vivanco I, Beroukhim R, et al.
Epidermal growth factor receptor activa-
tion in glioblastoma through novel mis-
sense mutations in the extracellular do-
main. PLoS Med 2006;3(12):e485.
26. Kesari S, Stiles CD. The bad seed:
PDGF receptors link adult neural progeni-
tors to glioma stem cells. Neuron 2006;
51:151-3.
27. Steck PA, Perhouse MA, Jasser SA, et
al. Identification of a candidate tumour
suppressor gene, MMAC1, at chromosome
10q23.3 that is mutated in multiple ad-
vanced cancers. Nat Genet 1997;15:356-62.
28. Guo P, Hu B, Gu W, et al. Platelet-
derived growth factor-B enhances glioma
angiogenesis by stimulating vascular en-
dothelial growth factor expression in tu-
mor endothelia and by promoting pericyte
recruitment. Am J Pathol 2003;162:1083-
93.
29. Jain RK, di Tomaso E, Duda DG, Loef-
fler JS, Sorensen AG, Batchelor TT. Angio-
genesis in brain tumours. Nat Rev Neuro-
sci 2007;8:610-22.
30. Ligon KL, Huillard E, Mehta S, et al.
Olig2-regulated lineage-restricted path-
way controls replication competence in
neural stem cells and malignant glioma.
Neuron 2007;53:503-17.
31. Reynolds BA, Weiss S. Generation of
neurons and astrocytes from isolated
cells of the adult mammalian central ner-
vous system. Science 1992;255:1707-10.
32. Assanah M, Lochhead R, Ogden A,
Bruce J, Goldman J, Canoll P. Glial pro-
genitors in adult white matter are driven
to form malignant gliomas by platelet-
derived growth factor-expressing retrovi-
ruses. J Neurosci 2006;26:6781-90.
33. Sanai N, Alvarez-Buylla A, Berger MS.
Neural stem cells and the origin of
gliomas. N Engl J Med 2005;353:811-22.

n engl j med 359;5  www.nejm.org  july 31, 2008 505

Downloaded from www.nejm.org on December 12, 2008 . Copyright © 2008 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l
34. Stiles CD, Rowitch DH. Glioma stem
cells: a midterm exam. Neuron 2008;58:
832-46.
35. Singh SK, Hawkins C, Clarke ID, et al.
Identification of human brain tumour ini-
tiating cells. Nature 2004;432:396-401.
36. Dirks PB. Brain tumor stem cells:
bringing order to the chaos of brain can-
cer. J Clin Oncol 2008;26:2916-24.
37. Beier D, Hau P, Proescholdt M, et al.
CD133(+) and CD133(-) glioblastoma-
derived cancer stem cells show differen-
tial growth characteristics and molecular
profiles. Cancer Res 2007;67:4010-5.
38. Vescovi AL, Galli R, Reynolds BA.
Brain tumour stem cells. Nat Rev Cancer
2006;6:425-36.
39. Bao S, Wu Q, Sathornsumetee S, et al.
Stem cell-like glioma cells promote tumor
angiogenesis through vascular endotheli-
al growth factor. Cancer Res 2006;66:
7843-8.
40. Calabrese C, Poppleton H, Kocak M,
et al. A perivascular niche for brain tumor
stem cells. Cancer Cell 2007;11:69-82.
41. Bao S, Wu Q, McLendon RE, et al.
Glioma stem cells promote radioresistance
by preferential activation of the DNA dam-
age response. Nature 2006;444:756-60.
42. Liu G, Yuan X, Zeng Z, et al. Analysis
of gene expression and chemoresistance
of CD133+ cancer stem cells in glioblas-
toma. Mol Cancer 2006;5:67.
43. Salmaggi A, Boiardi A, Gelati M, et al.
Glioblastoma-derived tumorospheres iden-
tify a population of tumor stem-like cells
with angiogenic potential and enhanced
multidrug resistance phenotype. Glia 2006;
54:850-60.
44. Dean M, Fojo T, Bates S. Tumour stem
cells and drug resistance. Nat Rev Cancer
2005;5:275-84.
45. Cha S. Update on brain tumor imag-
ing: from anatomy to physiology. AJNR
Am J Neuroradiol 2006;27:475-87.
46. Young GS. Advanced MRI of adult
brain tumors. Neurol Clin 2007;25:947-73.
47. Chen W. Clinical applications of PET
in brain tumors. J Nucl Med 2007;48:1468-
81.
48. Brandsma D, Stalpers L, Taal W,
Sminia P, van den Bent MJ. Clinical fea-
tures, mechanisms, and management of
pseudoprogression in malignant gliomas.
Lancet Oncol 2008;9:453-61.
49. Wen PY, Schiff D, Kesari S, Drappatz J,
Gigas D, Doherty L. Medical management
of patients with brain tumors. J Neuroon-
col 2006;80:313-32.
50. Glantz MJ, Cole BF, Forsyth PA, et al.
Practice parameter: anticonvulsant pro-
phylaxis in patients with newly diagnosed
brain tumors: report of the Quality Stan-
dards Subcommittee of the American Acad-
emy of Neurology. Neurology 2000;54:
1886-93.
51. Green H, Paul M, Vidal L, Leibovici L.
Prophylaxis of Pneumocystis pneumonia
in immunocompromised non-HIV-infect-
506 n engl j med 359;5  www.nejm.org  july 31, 2008
Downloaded from www.nejm.org on December 12, 2008 . Copyright © 2008 Massachusetts Medical Society. All rights reserved.
of m e dic i n e
ed patients: systematic review and meta-
analysis of randomized controlled trials.
Mayo Clin Proc 2007;82:1052-9.
52. Batchelor TT, Sorensen AG, di Tomaso
E, et al. AZD2171, a pan-VEGF receptor
tyrosine kinase inhibitor, normalizes tu-
mor vasculature and alleviates edema in
glioblastoma patients. Cancer Cell 2007;
11:83-95.
53. Gerber DE, Grossman SA, Streiff MB.
Management of venous thromboembo-
lism in patients with primary and meta-
static brain tumors. J Clin Oncol 2006;24:
1310-8.
54. Ruff RL, Posner JB. Incidence and
treatment of peripheral venous thrombo-
sis in patients with glioma. Ann Neurol
1983;13:334-6.
55. Levin JM, Schiff D, Loeffler JS, Fine
HA, Black PM, Wen PY. Complications of
therapy for venous thromboembolic dis-
ease in patients with brain tumors. Neu-
rology 1993;43:1111-4.
56. Lee AY, Levine MN, Baker RI, et al.
Low-molecular-weight heparin versus a
coumarin for the prevention of recurrent
venous thromboembolism in patients with
cancer. N Engl J Med 2003;349:146-53.
57. Meyers CA, Weitzner MA, Valentine
AD, Levin VA. Methylphenidate therapy
improves cognition, mood, and function
of brain tumor patients. J Clin Oncol
1998;16:2522-7.
58. Shaw EG, Rosdhal R, D’Agostino RB
Jr, et al. Phase II study of donepezil in ir-
radiated brain tumor patients: effect on
cognitive function, mood, and quality of
life. J Clin Oncol 2006;24:1415-20.
59. Litofsky NS, Farace E, Anderson F Jr,
Meyers CA, Huang W, Laws ER Jr. Depres-
sion in patients with high-grade glioma:
results of the Glioma Outcomes Project.
Neurosurgery 2004;54:358-66.
60. Asthagiri AR, Pouratian N, Sherman
J, Ahmed G, Shaffrey ME. Advances in
brain tumor surgery. Neurol Clin 2007;
25:975-1003.
61. Stummer W, Pichlmeier U, Meinel T,
Wiestler OD, Zanella F, Reulen HJ. Fluo-
rescence-guided surgery with 5-aminole-
vulinic acid for resection of malignant
glioma: a randomised controlled multi-
centre phase III trial. Lancet Oncol 2006;
7:392-401.
62. Lacroix M, Abi-Said D, Fourney D, et
al. A multivariate analysis of 416 patients
with glioblastoma multiforme: prognosis,
extent of resection, and survival. J Neuro-
surg 2001;95:190-8.
63. Walker MD, Alexander E Jr, Hunt WE,
et al. Evaluation of BCNU and/or radio-
therapy in the treatment of anaplastic
gliomas: a cooperative clinical trial. J Neu-
rosurg 1978;49:333-43.
64. Stupp R, Mason WP, van den Bent MJ,
et al. Radiotherapy plus concomitant and
adjuvant temozolomide for glioblastoma.
N Engl J Med 2005;352:987-96.
65. Hochberg FH, Pruitt A. Assumptions
in the radiotherapy of glioblastoma. Neu-
rology 1980;30:907-11.
66. Selker RG, Shapiro WR, Burger P, et
al. The Brain Tumor Cooperative Group
NIH Trial 87-01: a randomized compari-
son of surgery, external radiotherapy, and
carmustine versus surgery, interstitial ra-
diotherapy boost, external radiation ther-
apy, and carmustine. Neurosurgery 2002;
51:343-55.
67. Souhami L, Seiferheld W, Brachman D,
et al. Randomized comparison of stereo­
tactic radiosurgery followed by conven-
tional radiotherapy with carmustine to
conventional radiotherapy with carmus-
tine for patients with glioblastoma multi-
forme: report of Radiation Therapy Oncol-
ogy Group 93-05 protocol. Int J Radiat
Oncol Biol Phys 2004;60:853-60.
68. Tsao MN, Mehta MP, Whelan TJ, et al.
The American Society for Therapeutic Ra-
diology and Oncology (ASTRO) evidence-
based review of the role of radiosurgery
for malignant glioma. Int J Radiat Oncol
Biol Phys 2005;63:47-55.
69. Stupp R, Hegi ME, Gilbert MR, Chakra-
varti A. Chemoradiotherapy in malignant
glioma: standard of care and future direc-
tions. J Clin Oncol 2007;25:4127-36.
70. Duda DG, Jain RK, Willett CG. Antian-
giogenics: the potential role of integrating
this novel treatment modality with chemo-
radiation for solid cancers. J Clin Oncol
2007;25:4033-42.
71. Keime-Guibert F, Chinot O, Taillan-
dier L, et al. Radiotherapy for glioblasto-
ma in the elderly. N Engl J Med 2007;
356:1527-35.
72. Roa W, Brasher PM, Bauman G, et al.
Abbreviated course of radiation therapy in
older patients with glioblastoma multi-
forme: a prospective randomized clinical
trial. J Clin Oncol 2004;22:1583-8.
73. Glantz M, Chamberlain M, Liu Q,
Litofsky NS, Recht LD. Temozolomide as
an alternative to irradiation for elderly pa-
tients with newly diagnosed malignant
gliomas. Cancer 2003;97:2262-6.
74. Randomized trial of procarbazine,
lomustine, and vincristine in the adju-
vant treatment of high-grade astrocyto-
ma: a Medical Research Council trial.
J Clin Oncol 2001;19:509-18.
75. Fine HA, Dear KB, Loeffler JS, Black
PM, Canellos GP. Meta-analysis of radia-
tion therapy with and without adjuvant
chemotherapy for malignant gliomas in
adults. Cancer 1993;71:2585-97.
76. Stewart LA. Chemotherapy in adult
high-grade glioma: a systematic review
and meta-analysis of individual patient
data from 12 randomised trials. Lancet
2002;359:1011-8.
77. Westphal M, Hilt D, Bortey E, et al.
A phase 3 trial of local chemotherapy
with biodegradable carmustine (BCNU)
wafers (Gliadel wafers) in patients with
primary malignant glioma. Neuro Oncol
2003;5:79-88.

78. Westphal M, Ram Z, Riddle V, Hilt D,
Bortey E. Gliadel wafer in initial surgery
for malignant glioma: long-term follow-
up of a multicenter controlled trial. Acta
Neurochir (Wien) 2006;148:269-75.
79. van den Bent MJ. Anaplastic oligoden-
droglioma and oligoastrocytoma. Neurol
Clin 2007;25:1089-109.
80. Jenkins RB, Blair H, Ballman KV, et
al. A t(1;19)(q10;p10) mediates the com-
bined deletions of 1p and 19q and pre-
dicts a better prognosis of patients with
oligodendroglioma. Cancer Res 2006;66:
9852-61.
81. Cairncross JG, Ueki K, Zlatescu MC,
et al. Specific genetic predictors of che-
motherapeutic response and survival in
patients with anaplastic oligodendro-
gliomas. J Natl Cancer Inst 1998;90:
1473-9.
82. Ino Y, Betensky RA, Zlatescu MC, et
al. Molecular subtypes of anaplastic oli-
godendroglioma: implications for patient
management at diagnosis. Clin Cancer
Res 2001;7:839-45.
83. Ngo TT, Peng T, Liang XJ, et al. The
1p-encoded protein stathmin and resis-
tance of malignant gliomas to nitroso­
ureas. J Natl Cancer Inst 2007;99:639-
52.
84. Cairncross G, Berkey B, Shaw E, et al.
Phase III trial of chemotherapy plus ra-
diotherapy compared with radiotherapy
alone for pure and mixed anaplastic oligo-
dendroglioma: Intergroup Radiation Ther-
apy Oncology Group Trial 9402. J Clin
Oncol 2006;24:2707-14.
85. van den Bent MJ, Carpentier AF, Bran-
des AA, et al. Adjuvant procarbazine, lo-
mustine, and vincristine improves pro-
gression-free survival but not overall
survival in newly diagnosed anaplastic oli-
godendrogliomas and oligoastrocytomas:
a randomized European Organisation for
Research and Treatment of Cancer phase
III trial. J Clin Oncol 2006;24:2715-22.
86. Keles GE, Lamborn KR, Chang SM,
Prados MD, Berger MS. Volume of residu-
al disease as a predictor of outcome in
adult patients with recurrent supratento-
rial glioblastomas multiforme who are
undergoing chemotherapy. J Neurosurg
2004;100:41-6.
87. Butowski NA, Sneed PK, Chang SM.
Diagnosis and treatment of recurrent high-
grade astrocytoma. J Clin Oncol 2006;24:
1273-80.
88. Combs SE, Thilmann C, Edler L, De-
bus J, Schulz-Ertner D. Efficacy of frac-
tionated stereotactic reirradiation in re-
current gliomas: long-term results in 172
n engl j med 359;5  www.nejm.org  july 31, 2008
Downloaded from www.nejm.org on December 12, 2008 . Copyright © 2008 Massachusetts Medical Society. All rights reserved.
Medical Progress
patients treated in a single institution.
J Clin Oncol 2005;23:8863-9.
89. Yung WK, Prados MD, Yaya-Tur R, et
al. Multicenter phase II trial of temozolo-
mide in patients with anaplastic astrocy-
toma or anaplastic oligoastrocytoma at
first relapse: Temodal Brain Tumor Group.
J Clin Oncol 1999;17:2762-71. [Erratum,
J Clin Oncol 1999;17:3693.]
90. Wong ET, Hess KR, Gleason MJ, et
al. Outcomes and prognostic factors in
recurrent glioma patients enrolled onto
phase II clinical trials. J Clin Oncol 1999;
17:2572-8.
91. Yung WK, Albright RE, Olson J, et al.
A phase II study of temozolomide vs. pro-
carbazine in patients with glioblastoma
multiforme at first relapse. Br J Cancer
2000;83:588-93.
92. Brem H, Piantadosi S, Burger PC, et
al. Placebo-controlled trial of safety and
efficacy of intraoperative controlled deliv-
ery by biodegradable polymers of chemo-
therapy for recurrent malignant gliomas:
the Polymer-brain Tumor Treatment Group.
Lancet 1999;345:1008-12.
93. Rich JN, Reardon DA, Peery T, et al.
Phase II trial of gefitinib in recurrent
glioblastoma. J Clin Oncol 2004;22:133-
42.
94. Wen PY, Yung WK, Lamborn KR, et al.
Phase I/II study of imatinib mesylate for
recurrent malignant gliomas: North Amer-
ican Brain Tumor Consortium Study 99-08.
Clin Cancer Res 2006;12:4899-907.
95. Galanis E, Buckner JC, Maurer MJ, et
al. Phase II trial of temsirolimus (CCI-779)
in recurrent glioblastoma multiforme:
a North Central Cancer Treatment Group
Study. J Clin Oncol 2005;23:5294-304.
96. Chang SM, Wen P, Cloughesy T, et al.
Phase II study of CCI-779 in patients with
recurrent glioblastoma multiforme. In-
vest New Drugs 2005;23:357-61.
97. Cloughesy TF, Wen PY, Robins HI, et
al. Phase II trial of tipifarnib in patients
with recurrent malignant glioma either
receiving or not receiving enzyme-induc-
ing antiepileptic drugs: a North American
Brain Tumor Consortium Study. J Clin
Oncol 2006;24:3651-6.
98. Stommel JM, Kimmelman AC, Ying
H, et al. Coactivation of receptor tyrosine
kinases affects the response of tumor cells
to targeted therapies. Science 2007;318:
287-90.
99. Mellinghoff IK, Wang MY, Vivanco I, et
al. Molecular determinants of the response
of glioblastomas to EGFR kinase inhibi-
tors. N Engl J Med 2005;353:2012-24. [Er-
ratum, N Engl J Med 2006;354:884.]
100. Haas-Kogan DA, Prados MD, Lam-
born KR, Tihan T, Berger MS, Stokoe D.
Biomarkers to predict response to epider-
mal growth factor receptor inhibitors.
Cell Cycle 2005;4:1369-72.
101. Fine HA, Wen PY, Maher EA, et al.
Phase II trial of thalidomide and carmus-
tine for patients with recurrent high-
grade gliomas. J Clin Oncol 2003;21:2299-
304.
102. Vredenburgh JJ, Desjardins A, Hern-
don JE II, et al. Phase II trial of bevaci-
zumab and irinotecan in recurrent malig-
nant glioma. Clin Cancer Res 2007;13:
1253-9.
103. Vredenburgh JJ, Desjardins A, Hern-
don JE II, et al. Bevacizumab plus irinote-
can in recurrent glioblastoma multiforme.
J Clin Oncol 2007;25:4722-9.
104. Cloughesy TF, Prados MD, Wen PY,
et al. A phase II, randomized, non-com-
parative clinical trial of the effect of beva-
cizumab (BV) alone or in combination
with irinotecan (CPT) on 6-month pro-
gression free survival (PFS6) in recurrent,
treatment-refractory glioblastoma (GBM).
J Clin Oncol 2008;26:Suppl:91s. abstract.
105. Fulci G, Chiocca EA. The status of
gene therapy for brain tumors. Expert
Opin Biol Ther 2007;7:197-208.
106. Sampson JH, Archer GE, Mitchell DA,
Heimberger AB, Bigner DD. Tumor-specif-
ic immunotherapy targeting the EGFRvIII
mutation in patients with malignant glio-
ma. Semin Immunol (in press).
107. Reardon DA, Akabani G, Coleman
RE, et al. Salvage radioimmunotherapy
with murine iodine-131-labeled antitenas-
cin monoclonal antibody 81C6 for patients
with recurrent primary and metastatic ma-
lignant brain tumors: phase II study re-
sults. J Clin Oncol 2006;24:115-22.
108. Mamelak AN, Rosenfeld S, Bucholz
R, et al. Phase I single-dose study of intra-
cavitary-administered iodine-131-TM-601
in adults with recurrent high-grade glio-
ma. J Clin Oncol 2006;24:3644-50.
109. Ferguson S, Lesniak MS. Convection
enhanced drug delivery of novel therapeu-
tic agents to malignant brain tumors.
Curr Drug Deliv 2007;4:169-80.
110. Curran WJ Jr, Scott CB, Horton J, et
al. Recursive partitioning analysis of prog-
nostic factors in three Radiation Therapy
Oncology Group malignant glioma trials.
J Natl Cancer Inst 1993;85:704-10.
111. Brem SS, Bierman PJ, Black PH, et al.
Central nervous system cancers. J Natl
Compr Canc Netw 2008;6:456-504.
Copyright © 2008 Massachusetts Medical Society.
507
 New England Journal of Medicine

CORRECTION

Malignant Gliomas in Adults

Malignant Gliomas in Adults . The institution in Dr. Wen’s address

(page 492) should have read ``Dana–Farber/Brigham and Women’s
Cancer Center.´´ In Table 2 (page 503), the ``Dendritic cell and
EGFRvIII peptide vaccines´´ row should have included the following
entry: DCVax, CDX-110. The article has been corrected on the Jour-
nal’s Web site at www.nejm.org.

N Engl J Med 2008;359:877-a

Downloaded from www.nejm.org on December 12, 2008 . Copyright © 2008 Massachusetts Medical Society. All rights reserved.