RE-IRRADIATION FOR RECURRENT HIGH-GRADE GLIOMA: AN ANALYSIS OF PROGNOSTIC FACTORS FOR SURVIVAL AND PREDICTORS OF RADIATION NECROSIS

Recurrent high-grade glioma (rHGG) is a heterogeneous population, and the ideal patient selection for reirradiation (re-RT) has yet to be established. This study aims to identify prognostic factors for rHGG patients treated with re-RT. METHODS: We retrospectively reviewed consecutive adults with rHGG who underwent re-RT from 2009-2020 from our institutional database. The primary objective was overall survival (OS). The secondary outcomes included prognostic factors for early death (<6 months after re-RT) and predictors of radiation necrosis (RN). RESULTS: For the 79 patients identified, the median OS after re-RT was 9.9 months (95% CI 8.3-11.6). On multivariate analyses (MVA), re-resection at progression (HR=0.56, p=0.027), interval from primary treatment to first progression ≥16.3 months (HR=0.61, p=0.034), interval from primary treatment to re-RT ≥23.9 months (HR=0.35, p<0.001), and re-RT PTV volume <112 cc (HR=0.27, p<0.001) were prognostic for improved OS. Patients who had unmethylated-MGMT tumours (OR=12.4, p=0.034), ≥3 prior systemic treatment lines (OR=29.1, p=0.22), interval to re-RT <23.9 months (OR=9.0, p=0.039), and re-RT PTV volume ≥112 cc (OR=17.8, p=0.003) were more likely to survive <6 months. The cumulative incidence of RN was 11.4% (95% CI 4.3-18.5) at 12 months. Concurrent bevacizumab use (HR<0.001, p<0.001) and cumulative equivalent dose in 2 Gy fractions (cEQD2, a/b=2) <99 Gy2 (HR<0.001, p<0.001) were independent protective factors against RN. CONCLUSIONS: We observe favorable OS rates following re-RT and identified prognostic factors, including methylation status, that can assist in patient selection and clinical trial design. Concurrent use of bevacizumab and cEQD2 <99 Gy2 mitigates the risk of RN.