Chemotherapy response criteria in malignant glioma.

Abstract

No one has ever proven a relationship between the extent of response to chemotherapy in malignant glioma and time to progression or survival. We studied the predictive value of "imaging response" following two courses of nitrosourea-based chemotherapy in 136 patients with recurrent astrocytoma/malignant glioma. We performed image analysis by blinded side-to-side comparison of sequential studies, and categorized response into: partial response (PR) (>50% reduction), minor response (MR) (25-50% reduction), stable disease (SD) (<25% change), progressive disease (PD) (>25% increase). Patients with PR, MR, and SD did not differ with respect to time to progression (TTP) (p > 0.2) or survival (p > 0.2). Median TTP was 27 weeks for SD, 43 weeks for MR, and 30 weeks for PR. Patients with PD had a significantly reduced survival (p < 0.001). Median survival was 21 weeks for PD, 53 weeks for SD, 63 weeks for MR, and 48 weeks for PR. The lack of relationship between response and TTP may be due to early relapses in patients with response, a cytostatic benefit of chemotherapy in some patients who do not have an objective response, or a relatively favorable natural history in some tumors that do not respond to chemotherapy. Our data do not support the validity of current response grading, assessed after two courses of chemotherapy. Further research and validation of response criteria is necessary.