CHICAGO – Dose-dense temozolomide was no more effective than standard adjuvant temozolomide for patients with newly diagnosed glioblastoma in the randomized, phase III Radiation Treatment Oncology Group 0525 study.
Among 833 patients, the primary end point of median overall survival was 16.6 months with standard temozolomide (Temodar) vs. 14.9 months with dose-dense temozolomide (P = .63; hazard ratio 0.87).
Median progression-free survival was also similar at 5.5 months vs. 6.7 months (P = .06; HR, 0.87), Dr. Mark R. Gilbert reported at the meeting
RTOG 0525 was designed as a successor to the landmark EORTC (European Organisation for Research and Treatment of Cancer) trial that helped establish radiation plus concomitant and adjuvant temozolomide as the international standard of care for newly diagnosed glioblastoma (N. Engl. J. Med. 2005;352:987-96). It also helped to identify patients with MGMT (O6-methylguanine-DNA methyltransferase)–methylated tumors as a subpopulation that appeared to derive greater benefit from temozolomide (N. Engl. J. Med. 2005;352:997-1003).
It was hoped that dose-dense therapy would further enhance the efficacy of temozolomide in MGMT-methylated tumors and would deplete the endogenous reservoir of MGMT in nonmethylated MGMT tumors, as well as induce sensitivity to temozolomide, an oral alkylating agent. The MGMT gene encodes a DNA repair enzyme that can cancel the effects of alkylating chemotherapy.
Although the therapeutic end point was not reached, the study confirmed the prognostic value of MGMT promoter methylation status, said Dr. Gilbert, a professor of neuro-oncology at the University of Texas M.D. AndersonCancer Center in Houston.
MGMT-methylated tumors, compared with unmethylated tumors, were associated with significant improvements in median overall survival (21.2 months vs. 14 months; P less than .0001; HR, 1.74) and progression-free survival (8.7 months vs. 5.7 months; P less than .0001; HR, 1.63).
However, MGMT methylation status did not predict response to the dose-dense schedule, he said. No significant differences in overall or progression-free survival were observed between unmethylated and methylated tumors, regardless of treatment with standard or dose-dense temozolomide.
A Cox proportional hazard model showed that MGMT status and recursive partitioning analysis (RPA) class IV vs. III were significant predictors of overall and progression-free survival, whereas the treatment arm and type of radiation therapy (EORTC vs. RTOG) were not, Dr. Gilbert said.
He observed that RTOG 0525 had a slightly different population than did the EORTC trial in that it required a surgical procedure, enrolled patients older than 70 years, and required all patients to make it through chemoradiation to be randomized.
Invited discussant Dr. Jeffrey S. Abrams of the National Cancer Institute said "We’re all disappointed that it wasn’t positive, but there’s reason to take heart."
He observed that in 2.5 years of accrual, the group conducted what is likely the largest glioblastoma study to date, had a 98% tumor tissue collection rate, and ran a very well-conducted trial in an international collaboration. RTOG 0525 was conducted at 185 North American and 24 European centers by the RTOG, EORTC, and North Central Cancer Treatment Group.
"I think this bodes well for this sort of collaborative process that will speed the development of better treatments for glioblastoma," Dr. Abrams said.
He suggested that the trial may have failed because adjuvant temozolomide is not the active part of the regimen. He pointed out that a randomized, phase II Greek trial increased adjuvant temozolomide to 150 mg/m2 every 2 weeks, but saw no improvement in outcome (J. Clin. Oncol. 2005;23:2372-7) and that a small, single-institution study reported comparable survival outcomes with standard temozolomide when temozolomide was administered during the radiation phase at a lower dose of 50 mg/m2 without maintenance (Strahlenther Onkol. 2005;181:372-7).
In RTOG 0525, 833 patients received concurrent radiation at 200 cGy in 30 fractions plus temozolomide 75 mg/m2 daily, followed by 1:1 randomization to 6-12 cycles of standard temozolomide, starting at 150 mg/m2 and escalating to 200 mg/m2 if tolerated on days 1-5 of a 28-day cycle, or dose-dense temozolomide starting at 75 mg/m2 and escalating to 100 mg/m2 if tolerated on days 1-21 of a 28-day cycle.
Roughly one-fourth of patients were younger than 50 years old, more than half underwent total resection, and 45% had minor symptoms.
Grade 3-5 adverse events were significantly more common in the dose-dense arm, compared with the conventional arm (194 vs. 120; P less than .001), but they were mostly lymphopenia (107 vs. 51) and fatigue (33 vs. 12). No opportunistic infections occurred, Dr. Gilbert said.
The study was supported by Schering-Plough/Merck, the maker of temozolomide. Dr. Gilbert reported a consultant/advisory role with Genentech, and honoraria and research funding from Genentech and Merck. Several of his coauthors reported financial relationships with firms including Merck. Dr. Abrams reported that RTOG 0525 also receives funding from the National Cancer Institute, where he is employed.