Purpose: This phase III Eastern Cooperative Oncology Group-Southwest Oncology Group intergroup study was conducted to determine whether three 72-hour infusions of carmustine (BiCNU) and cisplatin administered monthly before external-beam radiotherapy would improve the survival of patients with newly diagnosed glioblastoma multiforme. The control arm consisted of radiation with standard adjuvant BiCNU.
Patients and methods: A total of 223 patients were accrued from 1996 to 1999. Of these, 219 patients were eligible; 109 were randomly assigned to the experimental arm, and 110 were randomly assigned to the control arm. Randomization was stratified by age, performance status, and extent of resection.
Results: The median age of the patients was 55 years; 55% were male, 93% were white, 26% had a biopsy only, and 84% were ambulatory. Treatment arms were well balanced with respect to baseline characteristics. Median follow-up time of the 15 patients still alive at the time of analysis was 3.3 years (range, 2 to 5 years). Median survival times for the standard and experimental arms were 11.2 and 11.0 months (P =.33, two-sided log-rank test), and survival at 1 year was 45% versus 44%, respectively. Fifty-six percent of patients received all three cycles of BiCNU/cisplatin, 12% received two cycles, and 31% received only one cycle. Toxicity was primarily hematologic and was more common in the experimental arm (P <.01).
Conclusion: This study demonstrates that 72-hour infusions of BiCNU and cisplatin followed by radiation do not improve median survival, survival at 1 year, or time to progression. Furthermore, this treatment requires more time in the hospital and is associated with more serious toxicities than standard therapy.