Abstract
BACKGROUND: High-dose proton beam therapy (PBT) uses excellent dose concentricity based on the unique characteristic termed the Bragg peak. PBT is a highly feasible treatment option that improves survival in select patients with newly diagnosed glioblastoma (GBM). However, selection bias remains an issue in prior studies that evaluated the efficacy of PBT. The aim of the present study was to compare the survival outcomes and toxicities of high-dose PBT and conventional radiation therapy (CRT) using propensity score-matched treatment cohorts. METHODS: The analysis included patients with newly diagnosed GBM treated with high-dose PBT of 96.6 Gy (RBE) or CRT of 60 Gy from 2010 to 2020. Propensity score generation and 1:1 matching of patients were performed based on the following covariates: age, sex, tumor location, extent of resection, chemotherapy, immunotherapy, and pre-radiation Karnofsky performance scale score. RESULTS: From a total of 235 patients, 26 were selected in each group by propensity score matching. The median overall survival (OS) of the PBT group was 28.3 months, while the median OS of the CRT group was 21.2 months. Although acute radiation-related toxicities were equivalent between the PBT and CRT groups, radiation necrosis as a late radiation-related toxicity was observed significantly more frequently in the PBT group. CONCLUSIONS: High-dose PBT provided significant survival benefits for patients with newly diagnosed GBM compared to CRT as shown by propensity score matching analysis. Radiation necrosis remains an issue in high-dose PBT; thus, the establishment of an effective treatment strategy centered on bevacizumab would be essential.