Selection bias, survival, and brachytherapy for glioma.

Interstitial irradiation is a promising treatment for malignant glioma. Longer than expected survival periods following treatment of recurrent tumor have led to the use of brachytherapy as an adjuvant treatment. The impact of patient selection on survival data was studied among candidates for this therapy. Consecutive, conventionally treated adults with newly diagnosed supratentorial tumors were identified retrospectively at a center where experience with glioma is population-based. Based on imaging and performance status, two surgeons and a radiation oncologist designated each patient as either eligible or ineligible for adjuvant brachytherapy. The survival and prognostic factors in the eligible and ineligible groups were analyzed. Overall, the patients eligible for brachytherapy (32% of the series) lived significantly longer than the ineligible patients (16.57 vs. 9.30 months), were younger, and had larger resections and better function. For glioblastoma, 40% of patients were eligible, and lived much longer than those who were ineligible (13.90 vs. 5.80 months). It is concluded that better outcome following adjuvant brachytherapy for glioma is at least in part the result of patient selection. Randomized trials of comparably selected patients will be necessary to demonstrate conclusively that longer survival is also a result of treatment.