Diffusion restriction as a predictor of response in recurrent glioblastoma patients receiving bevacizumab.

Abstract

2024 Background: Bevacizumab (BEV) improves outcome in a subset of patients with recurrent glioblastoma (GBM). Identification of this subset would improve patient selection for BEV therapy. The observation that BEV treatment induces diffusion weighted imaging (DWI) changes in some patients led to this study evaluating the relationship of DWI and apparent diffusion coefficient (ADC) to clinical outcome in recurrent GBM patients receiving BEV. Methods: We reviewed clinical data and brain MRIs of 262 adults with recurrent GBM treated with BEV. Patients with relevant imaging studies (MRI with DWI and ADC) at pretreatment baseline and at least 1 follow-up were categorized based on presence of DWI hyperintensity (DWI-HI) at baseline. Each group was subdivided based on the presence or absence of low ADC signal. The overall survival (OS) and progression free survival (PFS) rates were analyzed using the Kaplan Meier method and the log rank test. Results: Of 191 evaluable patients, 107 had new or increased DWI-HI after BEV. Patients with low ADC (n = 46) had a statistically significant improvement over those without low ADC (n = 61) in PFS (median 249 vs. 130 days; p = 0.0024; HR = 0.5) and OS (median 425 vs. 232 days; p < 0.0001; HR = 0.34). Of 60 patients with baseline DWI-HI which increased after BEV, patients with low ADC (n = 25) had better PFS and OS compared with those who did not (n = 35) (median PFS 218 vs. 97days; p = 0.0035, HR = 0.4; median OS 382 vs. 205 days; p = 0.0002; HR = 0.27). 47 patients developed DWI-HI after BEV; those with low ADC showed statistically significant improvement in OS but not PFS. Conclusions: Patients with a diffusion restricted (“ischemic”) MRI phenotype which appears or increases upon BEV treatment have improved PFS and OS compared with those without such changes, indicating a potential for using DWI to predict clinical outcome with BEV treatment in patients with recurrent GBM. A prospective study is being developed to better characterize the implications of these results. Author Disclosure Employment or Leadership Position Consultant or Advisory Role Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration Bayer, Genentech, Merck, Novartis, Pfizer, Schering-Plough IMEDEX, Merck, Novartis, Schering-Plough Celgene, Genentech, Lilly, MGI Pharma, Novartis, Pfizer, Schering-Plough