Prognostic factors in recurrent glioblastoma multiforme and anaplastic astrocytoma treated with selective intra-arterial chemotherapy.

background and purposeFactors predictive of primary brain tumor outcome have been studied extensively, although the prognostic value of radiologic data, such as MR imaging and angiographic characteristics, has not been studied in depth. The purpose of this study was to determine whether radiologic data were prognostic factors among patients with recurrent glioblastoma multiforme and anaplastic astrocytoma treated with selective intra-arterial chemotherapy.

methodsForty-six patients were enrolled in a Phase II study of intra-arterial chemotherapy with carboplatin and Cereport (Alkermes Inc.; Cambridge, MA), a bradykinin analog that selectively increases permeability of the blood-tumor barrier. MR imaging volumes of enhancing tumor, resection cavity, and T2 signal abnormality were measured with T1-weighted and T2-weighted sequences. Volumes were analyzed individually and in various combinations. Tumor vascularity was graded on angiograms. Outcome was measured by time to tumor progression and survival.

resultsOf 46 patients included in this study, 41 underwent evaluation. Thirty were male and 11 were female; mean age was 48.5 years. Karnofsky scores ranged from 70 to 100. Thirty-two patients had glioblastoma multiforme, whereas nine had anaplastic astrocytoma. Twenty-eight patients had tumor progression and 13 had stable disease. Twenty-three patients died after an average of 205 days; 18 were surviving at an average of 324 days from the start of intra-arterial chemotherapy. In multivariate analysis, time from diagnosis to intra-arterial chemotherapy was predictive both of time to tumor progression and survival. Net tumor volume and vascularity also were significant for survival. Age, Karnofsky performance status, histologic findings, gender, MR imaging area, resection cavity volume, T2 signal abnormality volume, and various combined volumes were not significant.

conclusionIf confirmed by further studies, radiologic factors such as tumor volume and angiographic vascularity should be considered in design and stratification of future chemotherapy trials.