Supratentorial = Glioblastoma Multiforme: The Role of Surgical Resection Versus Biopsy = Among Older Patients

Neuro-Oncology

Supratentorial = Glioblastoma=20 Multiforme: The Role of Surgical Resection Versus Biopsy Among Older = Patients=20

Kaisorn L. Chaichana¹, Tomas Garzon-Muvdi¹,=20 Scott Parker¹, Jon D. Weingart¹,=20 Alessandro Olivi¹, Richard Bennett¹,=20 Henry Brem¹ and = Alfredo Qui=F1ones-Hinojosa¹

(1)

The Johns Hopkins Neuro-Oncology = Surgical=20 Outcomes Research Laboratory, Departments of Neurosurgery, = Oncology and=20 Medicine, The Johns Hopkins Hospital, Johns Hopkins University = School of=20 Medicine, Baltimore, MD, USA

Kaisorn L. Chaich= ana (Corresponding=20 author)
Email: kaisorn@jhmi.edu

Alfredo Qui=F1ones-Hinojosa (Corresponding=20 author)
Email: aquinon2@jhmi.edu

Received:=20 17 January 2010 Published online:=20 10 August 2010

Abstract

Background

The peak incidence of glioblastoma multiforme (GBM) = occurs in=20 those aged 65 years and older. However, studies on this patient = group=20 remain limited. The goal of this study is to evaluate the efficacy of = surgery=20 versus biopsy for older patients with these lesions.

Methods

133 and 72 consecutive patients aged 65 years = and older=20 who underwent surgery and needle biopsy for intracranial primary (de = novo) GBM=20 between 1997 and 2007 were retrospectively reviewed. Among these = patients, 40=20 who underwent surgical resection were matched with 40 who underwent = needle=20 biopsy alone for factors consistently shown to be associated with = survival [age,=20 Karnofsky Performance Scale (KPS) indexing, eloquent involvement, = radiation,=20 temozolomide]. Survival was expressed as estimated Kaplan=96Meier plots, = and=20 log-rank analysis was used to compare survival curves. =

Results

Mean =B1 standard deviation age was=20 73 =B1 5 years, and median survival was = 4.9 months. There=20 were no significant differences in perioperative outcomes among patients = who=20 underwent surgical resection versus needle biopsy. Patients who = underwent=20 resection had median survival of 5.7 months as compared with=20 4.0 months for patients who underwent needle biopsy=20 (P =3D 0.02). Likewise, for patients aged 70 years = and=20 older, median survival was 4.5 months for 26 patients who underwent = surgical resection as compared with 3.0 months for 26 patients who=20 underwent needle biopsy (P =3D 0.03). =

Conclusion

This study demonstrates that older patients tolerate = aggressive surgery without increased surgery-related morbidity and have=20 prolonged survival as compared with similar patients undergoing needle = biopsy.=20 These findings may help guide treatment decisions for patients, their = families,=20 and their physicians.

Glioblastoma multiforme (GBM) is the most common = primary brain=20 tumor in adults.1=20 Incidence of this malignant tumor increases with increasing age, with = median age=20 at diagnosis of 64 years.2=20 Despite the increasing number of older patients with GBM, the optimal = treatment=20 strategy for older patients remains poorly defined.3,4=20 Treatment can vary from observation to radical resection, but the = effectiveness=20 of different types of surgical intervention remains poorly understood in = this=20 population of patients.3,4=20

The prognostic factors for patients with GBM are well = known.=20 These factors include age, preoperative neurological function, extent of = resection, and eloquent location.5=969=20 These factors, especially age, typically dictate the aggressiveness of = the=20 treatment strategy. Even though it is believed that extent of resection = may be=20 associated with prolonged survival, aggressive surgery is rarely pursued = in=20 older patients.3,4,10,11=20 Older patients are often assumed to have poor prognosis because of their = increased age and number of medical comorbidities, poor physiological = reserves,=20 and inability to tolerate surgery.3,4=20 As a result, these patients often undergo needle biopsy for diagnosis, = followed=20 by adjuvant therapy.3=20 Therefore, the benefit of aggressive resection in older patients remains = poorly=20 understood.

The goal of this retrospective study is to compare = outcomes=20 among patients who underwent aggressive surgical resection versus needle = biopsy=20 to better understand the efficacy of aggressive therapy versus needle = biopsy for=20 older patients with GBM.

Methods

Patient Selection

All patients aged >65 years who underwent = surgery=20 (surgical resection or needle biopsy) for intracranial primary GBM from = 1997 to=20 2007 were identified by review of all consecutive cases performed at a = single=20 institution. Pathology was determined by a neuropathologist in all cases = and was=20 based on World Health Organization (WHO) criteria. Patients with prior=20 resections or needle biopsies, previous adjuvant therapy, tumors = crossing the=20 corpus callosum, and/or infratentorial gliomas were excluded. Patients = with=20 incomplete medical records lacking presenting symptoms, pre- and = postoperative=20 magnetic resonance imaging (MRI), and adjuvant therapies were also = excluded.=20 Patients who were considered not healthy enough to tolerate surgical = resection=20 as well as those who underwent carmustine wafer implantation were = excluded,=20 since wafers are only implanted in patients who undergo resection. = Additionally,=20 patients who were not confirmed as having died were also excluded. =

Recorded Variables

The clinical, operative, and hospital course records = of all=20 patients who met the inclusion and exclusion criteria were abstracted.=20 Information collected from clinical notes included patient demographics, = presenting symptoms, medical comorbidities, neuroimaging = characteristics,=20 perioperative course, and postoperative adjuvant therapy. The Karnofsky=20 Performance Scale (KPS) index was used to classify preoperative = functional=20 status.12=20 KPS index was assigned by a reviewer blinded to patient outcomes based = upon=20 review of the clinic note prior to surgery. MRI images were obtained and = reviewed for each patient. Recorded characteristics included lesion size = (largest diameter based on T1-contrast-weighted images), specific lobe=20 involvement, and involvement of eloquent cortex. Eloquent cortex = included=20 motor/sensory cortex, language cortex, and/or basal ganglia/internal=20 capsule.9=20 Degree of resection was retrospectively classified from dictations of = MRIs=20 obtained <48 h after surgical resection by a blinded = reviewer as:=20 gross total resection (GTR) (>99%) if no residual enhancement was = noted on=20 postoperative MRI, near total resection (NTR) (>95%) if only rim = enhancement=20 of the resection cavity was noted on postoperative MRI, or subtotal = resection=20 (STR) (>90%) if residual nodular enhancement was noted on = postoperative=20 MRI.13=20

Date of death was recorded based on the Social = Security index=20 database.14=20 Perioperative death was defined as death within 30 days of surgery. =

Perioperative Treatment

The decision to pursue surgical resection versus = needle biopsy=20 was determined by a multidisciplinary team including the surgeon, = radiation=20 oncologist, medical oncologist, and the patient. For patients undergoing = surgical resection, the general aim was to achieve GTR when possible. = GTR was=20 typically not achieved when the tumor involved eloquent brain. For = needle=20 biopsies, the general aim was to obtain tissue for diagnosis. Tissue was = typically obtained from the contrast-enhanced portion of the tumor. = Motor and=20 somatosensory evoked potentials were routinely used in the majority of = cases=20 when tumors were near eloquent motor areas, and surgical navigation = [computed=20 tomography (CT) and/or MRI wand] was used in all cases after 2001. Use = of motor=20 mapping or electrocorticography largely depended upon the preference of = the=20 surgeon and the location of the tumor. Use of adjuvant radiation and=20 chemotherapy was also based upon recommendations from a = multidisciplinary team.=20 Following surgery, all patients received physical and occupational = therapy. The=20 decision to refer patients to rehabilitation was based upon = recommendations of=20 physical and occupational therapists.

Statistical Analysis

To compare the efficacy of surgical resection versus = needle=20 biopsy, a case=96control study was performed.15=20 During the review period, 133 and 72 patients underwent surgical = resection and=20 biopsy of primary GBM, respectively. Forty patients who underwent = surgical=20 resection were matched with 40 patients who underwent needle biopsy. The = groups=20 were matched by a reviewer blinded to patient outcomes for factors = consistently=20 shown to be associated with survival, including age = (=B1 3 years), KPS=20 (=B1 0), eloquent involvement (yes/no), radiation (yes/no), and = temozolomide=20 (yes/no).5,8,9=20

Survival as a function of time after surgical = resection or=20 biopsy was expressed as estimated Kaplan=96Meier plots. Parametric data = were=20 expressed as mean =B1 standard deviation (SD). Nonparametric = data were=20 expressed as median [interquartile range (IQR)]. Percentages were = compared via=20 chi-squared test. Continuous variables were compared via Student = t-test=20 or Mann=96Whitney U test where appropriate. Survival between = patients who=20 underwent surgical resection and needle biopsy was compared via log-rank = analysis. Values with P < 0.05 in these analyses = were=20 considered statistically significant.

Results

Patient Population

Forty older patients who underwent needle biopsy = were matched=20 with 40 patients who underwent surgical resection (Table 1).=20 For all 80 patients, mean =B1 SD age was = 73 =B1 5 years. At=20 presentation, median (IQR) KPS was 80 (80=9680). Seventeen (21%) = patients=20 presented with seizures, 11 (14%) with headaches, 33 (41%) with motor = deficits,=20 10 (13%) with sensory deficits, 21 (26%) with language deficits, 11 = (14%) with=20 gait deficits, and 26 (33%) with mental status changes. Average tumor = size was=20 4.0 =B1 1.5 cm, involving the frontal lobe in 35 (44%), = temporal=20 lobe in 28 (35%), parietal lobe in 14 (18%), and occipital lobe in 3 = (4%). The=20 characteristics of the patients in each group are presented in = Table 1.=20 There were no significant differences in the preoperative = characteristics=20 between patients who underwent needle biopsy versus surgical = resection.

Table 1 Study=20 population

Characteristics	Resection number (%)	Biopsy number (%)	P-value
Gender
Male	20 (50%)	20 (50%)	0.99
Age (years)^a	72.8 =B1 4.8	72.9 =B1 5.1	0.74
Presenting symptoms
KPS index^b	80 (70=9680)	80 (70=9680)	0.32
Symptom duration (months)^a =	1 (0.5=961.5)	1.4 (0.9=963.0)	0.19
Preoperative symptoms
Seizures	6 (15%)	11 (28%)	0.27
Headaches/nausea/vomiting	5 (13%)	6 (15%)	0.99
Motor deficit	15 (38%)	18 (45%)	0.65
Sensory deficit	3 (8%)	7 (18%)	0.32
Language deficit	9 (23%)	12 (27%)	0.61
Visual deficit	4 (10%)	3 (7%)	0.99
Gait deficit	5 (13%)	6 (13%)	0.99
Confusion/memory loss	12 (30%)	14 (35%)	0.81
Radiographics
Tumor size (cm)^a	4.0 =B1 1.6	4.0 =B1 1.4	0.84
Location
Frontal	16 (40%)	19 (48%)	0.65
Temporal	14 (35%)	14 (35%)	0.99
Parietal	8 (20%)	6 (15%)	0.77
Occipital	2 (5%)	1 (3%)	0.99
Hemorrhagic	4 (10%)	3 (7%)	0.99
Cortical	14 (31%)	11 (28%)	0.63
Subcortical	33 (73%)	30 (75%)	0.59
Eloquent cortex	11 (28%)	11 (28%)	0.99
Motor/sensory cortex	4 (10%)	5 (13%)	0.99
Language cortex	4 (10%)	4 (10%)	0.99
Basal ganglia/internal = capsule	3 (7%)	2 (5%)	0.99
Adjuvant therapy
Carmustine wafers	0 (0%)	0 (0%)	0.99
Temozolomide	4 (10%)	4 (10%)	0.99
Radiation therapy	32 (80%)	32 (80%)	0.99

Characteristics of patients aged 65 years and = older with=20 primary glioblastoma multiforme (GBM) who underwent surgical resection = or=20 biopsy. Forty patients who underwent surgical resection were matched = with 40=20 older patients who underwent needle biopsy of GBM. The groups were = matched for=20 age, Karnofsky performance score (KPS), eloquent location, postoperative = radiation, and postoperative temozolomide chemotherapy. Clinical and = treatment=20 variables were similar between patients who underwent surgical resection = and=20 patients who underwent needle biopsy

^aMean =B1 standard deviation =

^bMedian (interquartile range)=20

Perioperative Characteristics and Long-Term = Outcomes

Among the 40 older patients who underwent surgical = resection,=20 GTR was achieved in 15 (38%) patients, NTR in 25 (62%), and STR in 0 = (0%)=20 patients. All of the patients who underwent surgical resection = had >95%=20 of their tumor resected. Perioperatively, seven (9%) patients incurred a = motor=20 deficit and two (3%) a language deficit. Twenty-eight (35%) required = inpatient=20 rehabilitation, and median (IQR) hospital length of stay was 5 (2.8=968) = days. At=20 last follow-up, temozolomide was used in 8 (10%) patients and radiation = therapy=20 in 64 (80%) patients. There were no significant differences in = perioperative=20 characteristics between patients who underwent needle biopsy and = surgical=20 resection (Table 2).=20 At last follow-up, all patients (100%) had died. Median (IQR) survival = (40=20 biopsy, 40 surgical resection) of all patients in this study was=20 4.9 (2.9=9610.4) months (Fig. 1).

Table 2 Peri- and=20 postoperative outcomes

Characteristics	Resection number (%)	Biopsy number (%)	P-Value
Perioperative outcomes
New motor deficit	5 (11%)	2 (5%)	0.43
New language deficit	1 (3%)	1 (3%)	0.99
Surgical-site infection	1 (3%)	1 (3%)	0.99
Meningitis	0 (0%)	0 (0%)	0.99
Mortality	0 (0%)	1 (3%)	0.99
Discharge
Discharge to home	25 (63%)	27 (68%)	0.81
Discharge to rehabilitation	15 (37%)	13 (32%)	0.81
Hospital stay (days)^a	6 (4=968.3)	3 (1=966.5)	0.08
Survival
Died at last follow-up	40 (100%)	40 (100%)	0.99
Median survival (months)	5.7	4.0	0.02
3-Month survival rate	70%	53%
6-Month survival rate	40%	25%
9-Month survival rate	25%	18%
12-Month survival rate	15%	12%

Incidence of 30-day perioperative morbidity and = overall=20 survival in adults aged 65 years and older undergoing resection or = biopsy=20 of glioblastoma multiforme (GBM). Forty patients who underwent surgical=20 resection were matched with 40 patients who underwent surgical biopsy. = The=20 groups were matched for age, Karnofsky performance score (KPS index), = tumor=20 size, postoperative radiation, and postoperative temozolomide = chemotherapy.=20 Incidence of perioperative morbidity did not differ between the two = cohorts.=20 However, older patients who underwent resection experienced prolonged = survival=20 as compared with patients who underwent biopsy

^aMedian (interquartile range) =

Fig. 1 Kaplan=96Meier=20 plots of survival for older patients (aged 65 years and older) with = glioblastoma multiforme (GBM) who underwent surgical resection or needle = biopsy.=20 Median survival was 4.9 months, and 3-, 6-, 9-, and 12-month = survival rates=20 were 63%, 33%, 21%, and 14%, respectively

Patients Who Underwent Needle Biopsy versus = Surgical=20 Resection

Despite similarities in preoperative clinical status = and=20 treatment regimens (Tables 1=20 and 2),=20 patients who underwent surgical resection demonstrated prolonged = survival as=20 compared with patients who only underwent needle biopsy (Fig. 2).=20 Median survival for patients who underwent surgical resection was=20 5.7 months, while median survival for patients who underwent needle = biopsy=20 was 4.0 months (P =3D 0.02). This was = significantly=20 different on log-rank analysis (P =3D 0.02).=20

Fig. 2 Kaplan=96Meier=20 plots of survival for older patients (aged 65 years and older) who=20 underwent surgical resection versus biopsy of a primary glioblastoma = multiforme=20 (GBM). Median survival was significantly longer for patients who = underwent=20 aggressive resection as compared with patients who underwent biopsy (5.7 = versus=20 4.0 months, respectively, P =3D 0.02). The groups = were=20 matched for age, Karnofsky performance score (KPS index), eloquent = location,=20 postoperative radiation, and postoperative temozolomide chemotherapy =

Additionally, a subgroup analysis was performed on = patients=20 aged 70 years and older. As shown in Fig. 3,=20 similar to the overall analysis, the group who underwent surgical = resection had=20 significantly longer survival times as compared with those who underwent = needle=20 biopsy (median survival 4.5 months versus 4.0 months,=20 P =3D 0.03). These differences were significantly = different on=20 log-rank analysis (P =3D 0.03). There were not enough = patients to=20 compare the efficacy of surgical resection versus needle biopsy for = patients=20 older than 75 years of age. Among those patients who underwent = surgical=20 resection, 15 and 25 patients who had GTR and NTR of their tumor had = median=20 survival times of 5.8 and 5.4 months, respectively=20 (P =3D 0.79).=20

Fig. 3 Kaplan=96Meier=20 plots of survival for older patients (aged 70 years and older) who=20 underwent surgical resection versus biopsy of a primary glioblastoma = multiforme=20 (GBM). Median survival was significantly longer for patients who = underwent=20 surgical resection as compared with needle biopsy (4.5 versus = 3.0 months,=20 respectively, P =3D 0.03). The groups were matched for = age,=20 Karnofsky performance score (KPS), eloquent location, postoperative = radiation,=20 and postoperative temozolomide chemotherapy

Discussion

In this GBM study, 40 older patients who underwent = aggressive=20 surgical resection were matched with 40 older patients who underwent = needle=20 biopsy. These two cohorts were matched for factors consistently shown to = be=20 associated with survival, including age, KPS index, eloquent = involvement,=20 radiation, and temozolomide.5,8,9=20 Importantly, patients who underwent surgical resection had median = survival of=20 5.7 months, while patients who underwent needle biopsy without = resection=20 had median survival of 4.0 months. Likewise, when looking at = individuals=20 over the age of 70 years, patients who underwent surgical resection = had=20 median survival of 4.5 months, while individuals who underwent = needle=20 biopsy had median survival of 3.0 months. Although modest in = absolute=20 terms, older patients did have significantly prolonged survival with = aggressive=20 resection as compared with needle biopsy, without an increase in = surgical=20 morbidity or mortality. These findings can inform discussions with = patients=20 about potential risks and benefits from surgery, and should prove = helpful to=20 patients, families, and their physicians who are called upon to help in = these=20 difficult decisions.

GBM are characterized by their invasive and = infiltrative nature,=20 making curative resection difficult.16=20 In fact, in the 1930 s, Walter Dandy performed hemispherectomies to = attempt=20 to achieve curative resection, but survival remained less than=20 2 years.17=20 In recent years, with the advent of new surgical technologies, there has = been=20 renewed interest in understanding the effects of extensive resection on = survival=20 for patients with GBM.18=9622=20 These previous studies, however, are primarily limited to younger=20 patients.18=9622=20 Lacroix et al. performed a multivariate analysis on 416 patients with = GBM (mean=20 age 53 years) and found that patients who underwent >98% = resection=20 had improved survival over patients who underwent less extensive=20 resection.10=20 Similarly, Laws and colleagues compared the effects of surgery versus = biopsy for=20 patients (mean age 54 years; 20% under 40 years) with = malignant=20 astrocytomas (GBM and anaplastic astrocytomas) and found that surgery = was=20 associated with prolonged survival over biopsy.23=20 Neither study included enough older patients to define the risks and = benefits in=20 those aged over 65 years. Besides age, previous outcome studies = have been=20 limited by not controlling for pre- and postoperative confounding = variables,=20 including pathology, adjuvant treatment, and follow-up time.18=9621=20 As a result, the effects of extensive resection for older patients with = GBM have=20 remained poorly understood.

Older patients with GBM are rarely offered aggressive=20 surgery.3,4=20 This is presumably due to the assumption that older patients have poor=20 survival.24=9626=20 Median survival for most older patients with GBM is approximately = 4 months=20 as compared with 14 months for the general population.24=9626=20 Older patients tend to have an increased number of medical = comorbidities, which=20 presumably makes it difficult for older patients to tolerate surgical = resection=20 because of their reduced physiological reserves.27=9629=20 In the postoperative period, older patients are relatively = immunosuppressed as=20 compared with younger individuals, which may make them more susceptible = to=20 concomitant infections and prolonged hospital courses.30=20 These features have all been used as arguments for not offering older = patients=20 aggressive therapies for their disease.3,30,31=20 These patients instead are offered needle biopsy for diagnosis with or = without=20 subsequent adjuvant therapy.3,4=20

This study demonstrates that older patients undergoing = aggressive resection have similar surgery-related morbidity to older = patients=20 undergoing needle biopsy. The incidence of perioperative morbidity among = older=20 patients undergoing surgical resection is also similar to the = complication rates=20 seen in younger patients undergoing surgical resection in previous=20 studies.6,22,32=20 While the majority of older patients undergo needle biopsy for = diagnosis, older=20 patients may be able to tolerate aggressive surgery without = significantly added=20 inherent risks of surgery.3,4=20 Interestingly, the number of patients requiring rehabilitation placement = was=20 similar between the two cohorts. Not surprisingly, median length of = hospital=20 stay was longer for those patients undergoing surgical resection than = for those=20 undergoing needle biopsy, which trended towards but did not reach = statistical=20 significance.

This study also found that older patients undergoing = surgery=20 have increased survival times as compared with those undergoing needle = biopsy.=20 Patients who underwent surgery experienced an approximate 2-month = increase in=20 survival, or a 40% increase in survival time. This is substantial given = that the=20 survival time for older patients is approximately 4 months in = previous=20 studies.24=9626=20 While survival may seem dismal regardless of surgery or biopsy, older = patients=20 can benefit from surgery. Surgical resection, besides prolonged = survival, may=20 also provide additional benefits beyond needle biopsy, including = decreasing mass=20 effect, allowing for improved efficacy of radiation and chemotherapy, = and=20 possibility of implanting carmustine wafers or brachytherapy.5,33,34=20 Especially with advances in surgical technology including navigational = systems,=20 motor and language mapping, and radiology techniques including=20 diffusion-tensor-imaging-based fiber tracking, older patients may = benefit from=20 aggressive surgical resection as seen in younger patients without an = increase in=20 surgery-related morbidity.35,36=20 The 3-day increase in median hospital stay for older patients undergoing = surgical resection appears to be justified by the increase in survival = time.=20

This study demonstrates that older patients tolerate = aggressive=20 resection without increased morbidity and have prolonged survival, but = it does=20 have limitations. This was a retrospective case=96control analysis, not = a=20 prospective randomized trial. However, we tried to create a uniform = population=20 of older patients undergoing surgery for primary GBM by using strict=20 inclusion/exclusion criteria. We included only patients=20 aged >65 years at time of surgery and only those who = underwent=20 surgical resection or biopsy of supratentorial lesions. Furthermore, we=20 controlled for factors known to be associated with survival (age, KPS, = tumor=20 location, radiation, and temozolomide) and only included patients with = long-term=20 follow-up. With the use of these strict statistical controls and a = relatively=20 precise outcome measure, we believe that our findings offer useful = insights for=20 the management of older patients with GBM. Second, the decision to = pursue=20 aggressive surgery may have been biased to healthier patients with = better=20 preoperative neurological function or tumor location. However, this bias = was=20 mitigated by matching KPS index and tumor location. Because this was a=20 retrospective study, data indicative of quality of life and differences = in=20 quality of life between patients undergoing surgical resection versus = needle=20 biopsy were not ascertainable, and future prospective studies should = consider=20 data collection in this regard. Furthermore, this study did not use = volumetric=20 analysis of tumor, which may be a more accurate means of analyzing tumor = sizes.=20 The patients in this study also had aggressive (estimated to = be >95%)=20 resection of their tumors, and the findings of this study may therefore = not=20 apply to patients who underwent STR (<95%) of their tumor. = Ultimately,=20 larger, prospective studies are needed to confirm the role of aggressive = surgery=20 and extent of resection for older patients with GBM.

Conclusions

Incidence of GBM in older patients is expected to = increase in=20 coming decades, reflecting national and world demographics. Currently, = older=20 patients are typically not offered aggressive surgery for their tumors = because=20 of their poor prognosis, and concomitant age-related morbidity. This=20 case=96control study demonstrates that older patients undergoing = aggressive=20 resection did not have an increase in surgical morbidity, and had = prolonged=20 survival as compared with matched older patients undergoing needle = biopsy. Older=20 patients, as with younger patients, may therefore benefit from = aggressive=20 surgical resection, and aggressive resection should be considered for = older=20 patients who present with GBM.

References

1.	Wrensch M, Minn Y, Chew T, Bondy M, Berger MS. = Epidemiology of primary brain tumors: current concepts and review = of the=20 literature. Neuro Oncol. 2002;4:278=9699.

2.	Hegi ME, Diserens AC, Gorlia T, Hamou MF, de = Tribolet=20 N, Weller M, et al. MGMT gene silencing and benefit from = temozolomide in=20 glioblastoma. N Engl J Med. 2005;10:997=961003.

3.	Barnholtz-Sloan JS, Williams VL, Maldonado JL, = Shahani=20 D, Stockwell HG, Chamberlain M, et al. Patterns of care and = outcomes among=20 elderly individuals with primary malignant astrocytoma. J=20 Neurosurg. 2008;4:642=968.

4.	Iwamoto FM, Cooper AR, Reiner AS, Nayak L, = Abrey LE.=20 Glioblastoma in the elderly: the Memorial Sloan=96Kettering Cancer = Center=20 experience (1997=962007). Cancer. 2009;16:3758=9666.

5.	Brem H, Piantadosi S, Burger PC, Walker M, = Selker R,=20 Vick NA, et al. Placebo-controlled trial of safety and efficacy of = intraoperative controlled delivery by biodegradable polymers of=20 chemotherapy for recurrent gliomas. The Polymer-brain Tumor = Treatment=20 Group. Lancet. 1995;8956:1008=9612.

6.	Chang EL, Yi W, Allen PK, Levin VA, Sawaya RE, = Maor=20 MH. Hypofractionated radiotherapy for elderly or younger = low-performance=20 status glioblastoma patients: outcome and prognostic factors. = Int J=20 Radiat Oncol Biol Phys. 2003;2:519=9628.

7.	Keles GE, Lundin DA, Lamborn KR, Chang EF, = Ojemann G,=20 Berger MS. Intraoperative subcortical stimulation mapping for=20 hemispherical perirolandic gliomas located within or adjacent to = the=20 descending motor pathways: evaluation of morbidity and assessment = of=20 functional outcome in 294 patients. J Neurosurg. = 2004;3:369=9675.=20

8.	Chang SM, Parney IF, McDermott M, Barker FG, = II,=20 Schmidt MH, Huang W, et al. Perioperative complications and = neurological=20 outcomes of first and second craniotomies among patients enrolled = in the=20 Glioma Outcome Project. J Neurosurg. 2003;6:1175=9681. =

9.	Chaichana K, Parker S, Olivi A, = Quinones-Hinojosa A.=20 A proposed classification system that projects outcomes based on=20 preoperative variables for adult patients with glioblastoma = multiforme.=20 J Neurosurg. 2010;112:997=961004.

10.	Lacroix M, Abi-Said D, Fourney DR, Gokaslan = ZL, Shi=20 W, DeMonte F, et al. A multivariate analysis of 416 patients with=20 glioblastoma multiforme: prognosis, extent of resection, and = survival.=20 J Neurosurg. 2001;2:190=968.

11.	Nitta T, Sato K. Prognostic implications of = the=20 extent of surgical resection in patients with intracranial = malignant=20 gliomas. Cancer. 1995;11:2727=9631.

12.	Dutta D, Vanere P, Gupta T, Munshi A, Jalali = R.=20 Factors influencing activities of daily living using FIM-FAM = scoring=20 system before starting adjuvant treatment in patients with brain = tumors:=20 results from a prospective study. J Neurooncol. = 2009;94:103=9610.=20

13.	McGirt MJ, Chaichana KL, Gathinji M, = Attenello FJ,=20 Than K, Olivi A, et al. Independent association of extent of = resection=20 with survival in patients with malignant brain astrocytoma. J=20 Neurosurg. 2009;1:156=9662.

14.	Social Security Death Index Database, vol = 2009,=20 Rootsweb. http://ssdi.rootsweb.ancestry= .com/.=20

15.	Altman DG. Practical statistics for medical = research.=20 New York: Chapman & Hall/CRC; 1991.

16.	Claes A, Idema AJ, Wesseling P. Diffuse = glioma=20 growth: a guerilla war. Acta Neuropathol (Berl).=20 2007;5:443=9658.

17.	Dandy WE. Removal of right cerebral = hemisphere for=20 certain tumors with hemiplegia. JAMA. 1928;90:823=965. =

18.	Grant R, Metcalfe SE. Biopsy versus resection = for=20 malignant glioma. Cochrane Database Syst Rev. = 2005;CD002034.

19.	Hess KR. Extent of resection as a prognostic = variable=20 in the treatment of gliomas. J Neurooncol. = 1999;3:227=9631.

20.	Nazzaro JM, Neuwelt EA. The role of surgery = in the=20 management of supratentorial intermediate and high-grade = astrocytomas in=20 adults. J Neurosurg. 1990;3:331=9644.

21.	Quigley MR, Maroon JC. The relationship = between=20 survival and the extent of the resection in patients with = supratentorial=20 malignant gliomas. Neurosurgery. 1991;3:385=968; discussion = 388=969.=20

22.	McGirt MJ, Chaichana KL, Gathinji M, = Attenello FJ,=20 Than K, Olivi A, et al. Independent association of extent of = resection=20 with survival in patients with malignant brain astrocytoma. J=20 Neurosurg. 2009;110:156=9662.

23.	Laws ER, Parney IF, Huang W, Anderson F, = Morris AM,=20 Asher A, et al. Survival following surgery and prognostic factors = for=20 recently diagnosed malignant glioma: data from the Glioma Outcomes = Project. J Neurosurg. 2003;3:467=9673.

24.	Brandes AA, Franceschi E, Tosoni A, Benevento = F,=20 Scopece L, Mazzocchi V, et al. Temozolomide concomitant and = adjuvant to=20 radiotherapy in elderly patients with glioblastoma: correlation = with MGMT=20 promoter methylation status. Cancer. = 2009;15:3512=968.

25.	Iwamoto FM, Reiner AS, Nayak L, Panageas KS, = Elkin=20 EB, Abrey LE. Prognosis and patterns of care in elderly patients = with=20 glioma. Cancer. 2009;23:5534=9640.

26.	Kimple RJ, Grabowski S, Papez M, Collichio F, = Ewend=20 MG, Morris DE. Concurrent temozolomide and radiation, a reasonable = option=20 for elderly patients with glioblastoma multiforme? Am J Clin = Oncol.=20 2010;33:265=9670.

27.	Alonso M, Hamelin R, Kim M, Porwancher K, = Sung T,=20 Parhar P, et al. Microsatellite instability occurs in distinct = subtypes of=20 pediatric but not adult central nervous system tumors. Cancer = Res.=20 2001;5:2124=968.

28.	Rickert CH, Strater R, Kaatsch P, Wassmann H, = Jurgens=20 H, Dockhorn-Dworniczak B, et al. Pediatric high-grade astrocytomas = show=20 chromosomal imbalances distinct from adult cases. Am J = Pathol.=20 2001;4:1525=9632.

29.	Sung T, Miller DC, Hayes RL, Alonso M, Yee H, = Newcomb=20 EW. Preferential inactivation of the p53 tumor suppressor pathway = and lack=20 of EGFR amplification distinguish de novo high grade pediatric=20 astrocytomas from de novo adult astrocytomas. Brain Pathol. = 2000;2:249=9659.

30.	Brandes A, Fiorentino MV. Treatment of = high-grade=20 gliomas in the elderly. Oncology. 1998;1:1=966.

31.	Lowry JK, Snyder JJ, Lowry PW. Brain tumors = in the=20 elderly: recent trends in a Minnesota cohort study. Arch = Neurol.=20 1998;7:922=968.

32.	McGirt MJ, Mukherjee D, Chaichana KL, Than = KD,=20 Weingart JD, Quinones-Hinojosa A. Association of surgically = acquired motor=20 and language deficits on overall survival after resection of = glioblastoma=20 multiforme. Neurosurgery. 2009;3:463=969; discussion = 469=9670.

33.	Wu GN, Ford JM, Alger JR. MRI measurement of = the=20 uptake and retention of motexafin gadolinium in glioblastoma = multiforme=20 and uninvolved normal human brain. J Neurooncol.=20 2006;1:95=96103.

34.	McGirt MJ, Than KD, Weingart JD, Chaichana = KL,=20 Attenello FJ, Olivi A, et al. Gliadel (BCNU) wafer plus = concomitant=20 temozolomide therapy after primary resection of glioblastoma = multiforme.=20 J Neurosurg. 2009;3:583=968.

35.	Nimsky C, Ganslandt O, Buchfelder M, = Fahlbusch R.=20 Intraoperative visualization for resection of gliomas: the role of = functional neuronavigation and intraoperative 1.5 T MRI. Neurol = Res. 2006;5:482=967.

36.	Sanai N, Mirzadeh Z, Berger MS. Functional = outcome=20 after language mapping for glioma resection. N Engl J Med.=20 2008;1:18=9627.