Glioblastoma multiforme (GBM) is the second most common primary brain tumor diagnosed in individuals between 45 and 84 years of age, with an incidence in the United States of 3.17 per 100,000 person years.¹ According to the 2010 statistical analysis by the Central Brain Tumor Registry of the United States, from 2004 to 2006, GBM accounted for 53.8% of all glioma diagnoses, and 17.1% of all primary brain and CNS tumors reported for that period. Outcome has improved over the past three decades with 2- and 5-year survival rates of 3% and 1%, 30 years before, to 26% and 5%, respectively.^1,2,3 Development of neurosurgical navigational systems and pre- and intraoperative neuroimaging, as well as advances in radiation therapy planning have contributed greatly to the progress in survivorship. These improvements allow for greater resection at the time of diagnosis, as well as increasing the success of
subsequent resections at the time of relapsed disease. Chemotherapy has also added to prolonged length of life for patients with malignant brain tumors, but very few agents with sufficient efficacy are available as standard treatment for glioma management.

Nitrosourea, specifically BCNU (carmustine), was tested in a large randomized study conducted by the Brain Tumor Study Group for the up-front treatment of newly diagnosed GBM.⁴ In this trial of over 200 patients, the addition of intravenous (IV) BCNU, administered in 6-week cycles, following the completion of RT, resulted in an increase in 18-month survival rate to 19% when compared to that of 4% for those treated with surgical resection and RT alone. Myelosuppression and pulmonary toxicity limit the use of BCNU over the course of the disease, as does the development of tumor resistance to this agent. Regardless, the modest, but real survival advantage achieved with BCNU became the benchmark to overcome for the success of any new antineoplastic agent for GBM throughout the 1980s and 1990s.

In an effort to improve drug delivery, increase dose exposure, and reduce systemic toxicity, loco-regional chemotherapy administration was accomplished with the development of biodegradable polyanhydride polymers containing BCNU (Carmustine) 7.7 mg, in the form of Gliadel wafers (Eisai Pharmaceuticals, Woodcliff Lake, NJ) implanted directly into the resection cavity at the time of tumor removal. The efficacy of this modality was first tested in patients with recurrent GBM by the Polymer Brain Tumor Treatment Group in a randomized, double-blinded, placebo-controlled study.⁵ In this clinical trial, at the time of disease progression, patients underwent surgical re-resection with implantation of up to eight carmustine polymers (n = 110) or placebo polymers (n = 112). MST (median survival time) gained, measured from the time of re-resection, was an additional 31 weeks of life for the treatment group, compared with only 23 weeks for the placebo group (P = .006), with 6 month survival rates of 56% versus 36%, respectively (P = .01). Based on these data, the FDA approved Gliadel for use in recurrent GBM in June 1996.

Following this initial proof of efficacy in recurrent GBM, two prospective clinical trials were performed using Gliadel wafers for patients with newly diagnosed high-grade glioma, both grade III and grade IV. The first was a small study with plans to treat 100 subjects, but only 32 were enrolled before the project was closed due to a shortage of study drug.⁶ Those patients were randomized to surgical resection with either Gliadel or placebo wafers, followed by standard fractionated RT. In these patients, MST was 53.3 versus 39.9 weeks for the treatment and placebo groups, respectively, with 2-year survival rates of 30% and 6%. Based on these limited but favorable results, a large multi-institutional pivotal trial ensued as a phase III randomized study, double-blinded and placebo-controlled for patients with high-grade glioma at first diagnosis.⁷ In this study, 240 patients underwent surgical resection with implant of either Gliadel (n = 120) or placebo wafers (n = 120), followed by RT. No additional chemotherapy was permitted until disease progression, with the exception of patients found on final path review of permanent sections to have oligodendroglial tumors who were then allowed to be treated with adjuvant PCV (procarbazine, CCNU, vincristine), which was considered standard treatment at that time for oligodendroglioma. The results of this study have been a point of debate over the years, because of the lack of robust survival advantage in treatment group with MST of 13.8 months, versus 11.6 months for the control group, without the GBM subgroup reaching adequate significance. Nonetheless, the FDA approved Gliadel in February 2003, for the up-front treatment of all high-grade malignant glioma, not just GBM, at initial diagnosis. Long-term follow-up data have now provided information about 2- and 3-year survival rates of 15.8% and 9.2% versus 8.3% and 1.7%, respectively. Eleven patients were alive at 56 months, nine of whom were in the Gliadel treatment group, and the survival advantage remained statistically significant at 3 years (P = .01).⁸

In the 1990s, temozolomide (TMZ) (Temodar Schering-Plough, Kenilworth, NJ; Temodar Schering-Plough, Houten, The Netherlands), a novel alkylating agent, was introduced for the treatment of malignant gliomas. TMZ is a novel, second generation alkylating agent that is an imidazotetrazine derivative, orally absorbed, that crosses the blood-brain barrier with ease. Its method of action is by undergoing transformation to an active metabolite MTIC (monomethyl triazenoimidazole carboxamide). Its therapeutic benefit is achieved by methylation of several DNA sites, including O⁶ guanine and O⁶ methylguanine. This agent gained immediate attention and popularity for several reasons: low side effect profile, no permanent end organ toxicity, oral preparation, and 100% bioavailability with CNS penetration. A series of phase II trials were conducted to establish safety and efficacy profiles for anaplastic glioma and GBM. Yung et al.⁹ published the results of a phase II study in which patients (n = 111) with anaplastic astrocytoma (AA) and mixed anaplastic glioma were treated at relapse following RT (with or without prior chemotherapy) with adjuvant cycles of TMZ, dosed for 5 consecutive days every 28 days. Survival was extended a median of an additional 13.6 months, with 6- and 12-month survival rates of 75% and 56%, respectively. In addition, there were objective responses noted with 8% complete response, 27% partial response (PR), and 26% with stable disease, with a median duration of 4.4 months. The follow-up phase II trial was conducted as a randomized, open label study, in which 225 patients with recurrent GBM received either TMZ standard 5 days every 28 days schedule or procarbazine monotherapy daily for 28 continuous days, repeated every 56 days.¹⁰ Many patients had been previously treated with PCV. Progression-free survival (PFS) rate at 6 months was 21%, for the TMZ group compared with 8% for the procarbazine group (P = .008), and overall 6-month survival rates were 60% versus 44% (P = .019). TMZ was FDA approved for the treatment of recurrent AA or oligoastrocytoma after failure of RT and nitrosourea or procarbazine chemotherapy.

TMZ was not approved for newly diagnosed GBM in the United States until May 2005, following completion and publication of a large randomized phase III clinical trial conducted in Canada and Europe between 2000 and 2002.³ In this study, patients with newly diagnosed GBM were treated with surgery followed by RT alone (n = 286), or with the addition of TMZ
administered concurrently (75 mg/m²/day) with RT, followed by six cycles of adjuvant treatment (150 to 200 mg/m²/day for 5 days every 28 days) after completion of chemoradiotherapy (n = 287). MST was 12.1 months for the control group and 14.6 months for the treatment group (P < .001). Final results and 5-year analysis have now been published.¹¹ Of those treated with RT alone, 97% are deceased and 89% of patients treated with chemoradiotherapy (RT/TMZ) have also died. Survival rates at 2 and 5 years for the RT only and the RT/TMZ groups were 10.9% and 1.9% versus 27.2% and 9.8%, respectively.

Analysis of the methylation status of the methylguanine methyl transferase (MGMT) gene was determined from the tumor tissue of 206 patients on this trial.¹² Production of the DNA repair enzyme O⁶MGMT is limited by the methylation of the promoter of the MGMT gene. Analysis of tissue samples from patients enrolled in a prospective clinical trial of chemoradiotherapy (RT/TMZ) demonstrated that the subgroup of patients with methylated MGMT reached a median survival of 23.4 months compared only 12.6 months for those with unmethylated MGMT, so that MGMT methylation became an independent predictor of survival.^11,12

Once Gliadel and TMZ were approved for treatment independently, the obvious question was if even greater survival could be reached by exploiting any synergy of using combined modality treatment. Investigators from Johns Hopkins, the original developers of Gliadel reported on the combined use of Gliadel in newly diagnosed patients who then underwent chemoradiotherapy with RT and concurrent, followed by maintenance, TMZ cycles.¹³ This is a treatment strategy, which has been popular in the community setting but has not been formally studied in a prospective manner. Despite the limits of this retrospective review, the data are from the single institution with the most experience applying this modality. For a 10-year period (1997 to 2006), all cases undergoing resection of newly diagnosed GBM, with or without Gliadel, were reviewed, including cases from 2004, when TMZ became the new standard of treatment at Johns Hopkins. From these patient cohorts, 33 patients with newly diagnosed GBM were treated with Gliadel followed by RT with concurrent TMZ followed by adjuvant cycles of TMZ. The MST for these patients was 20.7 months, with a 2-year survival rate of 36%. These patients were then compared to the cohort of 78 patients treated at the same institution with Gliadel and RT only, before the addition of TMZ, for whom median survival was only 12.4 months with 2-year survival rate of 18%. There was no additive toxicity noted from combining these therapies. These findings are superior to the use of RT and TMZ established by the EORTC (European Organisation for Research and the Treatment of Cancer) study which achieved a median survival of 14.6 months with a 2-year survival rate of 26%.³ This updated information on the safety and efficacy of combining Gliadel wafers with RT and TMZ is promising but should be more definitively established by a prospective analysis.

With the widespread use of TMZ for newly diagnosed GBM, efforts have been underway to study alternative dosing schedules to optimize efficacy beyond what was initially achieved with concurrent daily chemoradiotherapy followed by 5 days of TMZ every 28 days. Clarke et al.¹⁴ conducted a phase II trial of concurrent RT plus TMZ followed by either (1) a dose-dense (7-day on, 7-day off administration schedule of 150 mg per m²) or a metronomic daily dosing schedule of 50 mg per m² continuously administered. This study enrolled 85 patients randomized between dose-dense schedule (n = 42 with 31 treated) and metronomic schedule (n = 43 with 28 treated). Treatment continued for six adjuvant cycles followed by maintenance treatment with cis-retinoic acid until tumor progression. The overall survival and 2-year survival rates for the dose-dense and metronomic treatment groups were 17.1 versus 15.1 months and 34.8% and 28%, respectively. MGMT analysis was attempted on tissue from 68 patients finding 39 unmethylated, 9 methylated, and 20 samples inadequate for testing. MST for patients with methylated MGMT was 28.1 months. Toxicity was not unexpected, but more frequent for patients on the dose-dense schedule. Modest improvements were noted over standard treatment schedules, but overall, this study was too small to influence a universal change of TMZ dosing in the up-front setting.

The large phase III randomized trial, intergroup study (RTOG [Radiation Therapy Oncology Group] 0525/EORTC /NCCTG [North Central Cancer Treatment Group]) has enrolled over 1,100 subjects with newly diagnosed GBM who were randomized between Arm A: standard concurrent treatment RT plus TMZ 75 mg/m²/day followed by 5-day cycles (200 mg/m²/day), and Arm B: concurrent treatment followed by 21 day cycles (100 mg/m²/day). Final analysis and results are forthcoming but special attention will be paid to the incidence of any prolonged myelosuppression on the extended schedule.

A special subset of patients with GBM is the elderly population. This group has been identified as having possible beneficial effects from being treated with either hypofractionated RT scheduling (3 to 4 weeks of treatment rather than 6 weeks) or chemotherapy alone as first-line treatment. A retrospective analysis reviewed patients with GBM over the age of 70 years at diagnosis treated with TMZ on the standard 5-day every 28 days schedule, but without RT.¹⁵ Patients received a median of 5 of 12 planned cycles. There was an objective response rate of 28%, and stable disease in 31%. Progressive disease occurred in 33%. Median progression free and overall survival times were 20 and 36 weeks, respectively. MGMT expression did not correlate with outcome in this group. Treatment was well tolerated and was discontinued only because of disease progression, with the exception of two patients who discontinued because of toxicity. There were no treatment-related deaths. Alternative approaches for the elderly are attractive given the changing demographics of the population in general.

Two large prospective randomized trials are underway to address these treatment issues. First, the Nordic Clinical Brain Tumor Study Group completed enrollment of 342 patients aged > 60 with GBM who were randomized among three treatment arms: (1) standard fractionated RT over 6 weeks; (2) hypofractionated RT over 3 weeks; (3) six cycles of standard 5 day every 27 days dosing of TMZ without RT. Second, EORTC 26062-26061, in conjunction with Canada’s NCIC (National Canadian Institute of Cancer), is enrolling
560 patients aged > 65, who will be randomized to either shortcourse RT alone or short-course RT with concurrent, followed by 12 cycles of standard, TMZ. Results of both of these trials are pending and may set a new standard for seniors with malignant brain tumors.

Alternate dosing schedules of TMZ remain an attractive option for patients at relapse and are discussed below.

A universally accepted standard for the treatment of newly diagnosed anaplastic gliomas has not been established with the same rigor as for GBM. Surgery followed by RT has been the basis of treatment. However, there has been widespread application of the chemoradiotherapy (RT/TMZ) regimen of concurrent fractionated RT with concurrent TMZ followed by adjuvant TMZ approved for GBM, known as the “Stupp protocol” outlined above.³ Officially, this is considered “offlabel” use of TMZ for up-front treatment of patients with newly diagnosed anaplastic tumors. Currently, large phase III, randomized clinical trials testing the efficacy of TMZ for newly diagnosed AA, anaplastic oligodendroglioma (AO), and anaplastic oligoastrocytoma (AOA) tumors are ongoing. Specifically, two large intergroup clinical trials are open for enrollment for WHO grade III gliomas. These trials have a novel approach to histologic diagnosis and rather than classifying patients as AA, AO, or AOA, patients are enrolled based on status of 1p/19q allelic co-deletion rather than histologic diagnosis. The significance of this chromosome signature was brought to attention in a study by Cairncross, in which patients with AO and AOA tumors were evaluated for both overall survival and responsiveness to chemotherapy, specifically multiagent PCV, which was the treatment regimen of choice for all malignant gliomas at the time of that investigation.¹⁶ In this study of 34 patients with anaplastic glioma, there was a difference in overall survival, favoring patients whose tumors were found to have deletion of the short arm of chromosome 1, the long arm of chromosome 19, or both. This finding has been replicated by many other investigators in large treatment series as well as pathology series. Jenkins et al.¹⁷ have further attributed the mechanism of this co-deletion as a chromosome translocation. The RTOG along with the EORTC, NCIC-CTG (Clinical Trials Group), and NCCTG consortia are conducting these large randomized, multi-institutional trials to test a variety of combinations of fractionated RT in combination with TMZ, concurrently or sequentially, versus RT alone, or TMZ alone. Together, these two trials will enroll over 1,200 subjects and should firmly establish the best chemoradiotherapy combination for the treatment of anaplastic glioma, as well as redefining the neuropathologic approach to categorizing these tumors.

Based largely on Cairncross’ observation that anaplastic oligodendroglial tumors are chemosensitive, there have been several chemotherapy-based clinical trials for anaplastic glioma. First, an intergroup study conducted for AA has closed to accrual and is undergoing data analysis. In this trial, over 200 patients were randomized to receive RT with either oral TMZ or IV BCNU. Results are forthcoming.

Vogelbaum et al.¹⁸ published the results of RTOG BR0131, a phase II study of preirradiation TMZ followed by concurrent RT plus TMZ in patients with newly diagnosed AO and AOA. This was a small, nonrandomized study that enrolled 42 patients for treatment, of which only 28 were evaluable for response. In this treatment schema, the preirradiation TMZ was a dose-dense schedule of 150 mg/m²/day for 7 days on alternating with 7 days off, for six cycles. Response was evaluated prior to initiating RT plus concurrent TMZ (75 mg/m²/day). No further adjuvant cycles were administered. The overall objective response rate (complete and partial radiographic responses) was 32%. Co-deletion of 1p 19q was found in 17 of 28 patients (60.7%), all of whom were progressionfree at 6 months. Methylation of the MGMT promoter was found in 16 (80%) of 20 patients whose tumors were tested, all of whom were also progression-free at 6 months. Thirteen of the 16 patients with methylated MGMT also had 1p and 19q co-deletion (81%). Co-deletion of 1p/19q was associated with a reduced risk of progression (P = .01) and prolonged survival (P = .04); but, methylated MGMT did not correlate significantly with prolonged survival. The overall and PFS rates at 30 months were 81% and 64%, respectively. Thirteen patients experienced progressive disease and went on to salvage treatment with surgery, chemotherapy, or stereotactic radiosurgery. For the 22 patients who completed the prescribed therapy, more than half-experienced grade III or IV hematologic toxicity during the dose-dense preirradiation treatment, and 36% experienced grade III toxicity during the concurrent RT/TMZ treatment. The results of this clinical trial did not establish dose-dense preirradiation TMZ as the new standard treatment for these tumors but did demonstrate sufficient efficacy to launch the ongoing clinical trials outlined above.