Abstract
We have previously shown that diffusion MR characteristics are a predictive imaging biomarker for survival benefit in recurrent glioblastoma treated with anti-VEGF therapy; however, contemporary clinical use of bevacizumab is often limited to patients with very large tumors and/or after multiple recurrences. We hypothesize that diffusion MR characteristics can be used to predict long-term survival benefit in these patients, which may be beneficial for clinical decision-making. The current study identified 83 recurrent glioblastoma patients from our institution who were treated with bevacizumab over the past 5 years with high quality anatomic and diffusion MRI data. Of these 83 patients, 35 had large contrast enhancing tumors (>20cc or >3.4cm diameter, group average). Additionally, we identified 37 recurrent glioblastoma patients from the bevacizumab treated control arm of a recent multicenter phase III trial (NCT02511405) with high quality data and large enhancing tumors for validation. Pre-treatment tumor volume was quantified using T1 subtraction maps and apparent diffusion coefficient (ADC) histogram analysis was used to phenotype patients as having high (> 1.24 um2/ms) or low (< 1.24 um2/ms) ADCL, the mean value of the lower peak using a double Gaussian mixed model. Median overall survival in patients with large volume recurrent glioblastoma was ~5.7 months. High ADCL was associated with significantly longer overall survival (OS) compared with low ADCL in both single center (P=0.0271, HR=0.486, mOS=5.5 vs. 2.8mo) and multicenter phase III data (P=0.0457, HR=0.507, mOS=7.9 vs. 5.7mo). Accounting for absolute tumor volume and age, both cohorts showed that ADCL was an independent prognostic factor for OS (Cox, P< 0.01). In summary, pre-treatment diffusion MR imaging is an independent predictive biomarker for OS in recurrent glioblastoma with a large tumor burden.