Purpose: Older patients with newly diagnosed glioblastoma have poor outcomes, and optimal treatment is controversial. Hypofractionated radiation therapy (HRT) is frequently used but has not been compared to patients receiving standard fractionated radiation therapy (SRT) and temozolomide (TMZ).
Methods and materials: We conducted a retrospective analysis of patients ≥65 years of age who received radiation for the treatment of newly diagnosed glioblastoma from 1994 to 2013. The distribution of clinical covariates across various radiation regimens was analyzed for possible selection bias. Survival was calculated using the Kaplan-Meier method. Comparison of hypofractionated radiation (typically, 40 Gy/15 fractions) versus standard fractionation (typically, 60 Gy/30 fractions) in the setting of temozolomide was conducted using Cox regression and propensity score analysis.
Results: Patients received SRT + TMZ (n=57), SRT (n=35), HRT + TMZ (n=34), or HRT (n=9). Patients receiving HRT were significantly older (median: 79 vs 69 years of age; P<.001) and had worse baseline performance status (P<.001) than those receiving SRT. On multivariate analysis, older age (adjusted hazard ratio [AHR]: 1.06; 95% confidence interval [CI]: 1.01-1.10, P=.01), lower Karnofsky performance status (AHR: 1.02; 95% CI: 1.01-1.03; P=.01), multifocal disease (AHR: 2.11; 95% CI: 1.23-3.61, P=.007), and radiation alone (vs SRT + TMZ; SRT: AHR: 1.72; 95% CI: 1.06-2.79; P=.03; HRT: AHR: 3.92; 95% CI: 1.44-10.60, P=.007) were associated with decreased overall survival. After propensity score adjustment, patients receiving HRT with TMZ had similar overall survival compared with those receiving SRT with TMZ (AHR: 1.10, 95% CI: 0.50-2.4, P=.82).
Conclusions: With no randomized data demonstrating equivalence between HRT and SRT in the setting of TMZ for glioblastoma, significant selection bias exists in the implementation of HRT. Controlling for this bias, we observed similar overall survival for HRT and SRT with concurrent TMZ among elderly patients, suggesting the need for a randomized trial to compare these regimens directly.
Copyright © 2015 Elsevier Inc. All rights reserved.