Validation of Preoperative Neoadjuvant Bevacizumab Therapy for Newly Diagnosed Glioblastoma via Comparative Analyses with Propensity Score Matching

Résumé fourni par la source

Background/Objectives: This study assessed the clinical effects of preoperative neoadjuvant bevacizumab (neoBev) on newly diagnosed glioblastoma (GB) and investigated imaging-based predictors of outcome. Methods: We retrospectively reviewed 33 patients who received neoBev (10 mg/kg) between 2015 and 2024. Surgery was performed 21–30 days after treatment. Controls comprised 33 patients treated with standard chemoradiotherapy (radiation plus temozolomide) selected through propensity score matching. We assessed changes in performance status, extent of resection (EOR), and volumetric alterations on gadolinium-enhanced T1-weighted imaging (T1Gd) and fluid-attenuated inversion recovery (FLAIR). Median progression-free survival (mPFS) and median overall survival (mOS) were compared. Imaging responders were defined using receiver operating characteristic-derived thresholds. Results: All patients had GB (31 with isocitrate dehydrogenase (IDH)1 wild-type, 2 with IDH-mutant). Median Karnofsky Performance Status improved significantly after treatment in the NeoBev group (from 71 to 83), but remained unchanged in Controls. After matching, mPFS and mOS were 10.2 and 17.9 months with neoBev, and 13.3 and 18.2 months in Controls, respectively. Within the neoBev cohort, T1Gd responders (>37% reduction) achieved longer mPFS and mOS than non-responders. EOR-stratified analysis showed that outcomes favored standard therapy in cases with gross total resection, while subtotal resection cases demonstrated comparable or slightly better OS with neoBev. Conclusions: Preoperative neoBev improved functional status and identified an imaging-defined subgroup with favorable prognosis, although survival was not superior to that with standard therapy. NeoBev may benefit selected patients with limited resectability of newly diagnosed GB.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.