Value of [(18)F]-FDG positron emission tomography in patients with recurrent glioblastoma receiving bevacizumab

BACKGROUND: Treatment of recurrent glioblastoma (GBM) with bevacizumab can induce MRI changes that confound the determination of progression. We sought to determine the value of [(18)F]-fluorodeoxyglucose (FDG) positron emission tomography (PET) in GBM patients receiving bevacizumab at the time of suspected progression and, thereby, its utility as a potential prognostic adjunct in progressive disease. METHODS: This retrospective study included patients who underwent brain FDG PET within 4 weeks of receiving bevacizumab for recurrent GBM with suspected progression. Volumes-of-interest were placed over the reference lesion with measurement of maximum standardized uptake value (SUV(max)), peak standardized uptake value (SUV(peak)), metabolic tumor volume, total lesion glycolysis (TLG), and tumor-to-normal contralateral white matter ratios (TNR-WM). Tumors were additionally categorized as non-avid or avid based on qualitative FDG uptake. Associations between baseline variables and overall survival (OS) were examined using univariable and multivariable Cox proportional hazards regression, with P < .05 considered significant. RESULTS: Thirty-one patients were analyzed. Qualitative FDG uptake was significantly associated with OS (P = .03), with a median OS of 9.0 months in non-avid patients versus 4.5 months in avid patients. SUV(max), SUV(peak), TNR-WM, and TLG were significantly associated with OS (P < .001, TLG: P = .009). FDG avidity and SUV(max) remained significantly associated with OS (P = .046 and .048, respectively) in the multivariable analysis including age, KPS, and MGMT status. Dichotomizing patients using an SUV(max) cutoff of 15.3 was associated with OS (adjusted P = .048). CONCLUSION: FDG PET is a promising imaging tool to further stratify prognosis in recurrent GBM patients on antiangiogenic therapy.