Determining a cut-off residual tumor volume threshold for patients with newly diagnosed glioblastoma treated with temozolomide chemoradiotherapy: A multicenter cohort study.

Standard-of-care treatment of glioblastomas involves maximal safe resection and adjuvant temozolomide chemo-radiotherapy. Although extent of resection (EOR) is a well-known surgical predictor for overall survival most lesions cannot be completely resected. We hypothesize that in the event of incomplete resection, residual tumor volume (RTV) may be a more significant predictor than EOR. This was a multicenter retrospective review of 147 adult glioblastoma patients (mean age 53 years) that underwent standard treatment. Semiautomatic magnetic resonance imaging segmentation was performed for pre- and postoperative scans for volumetric analysis. Cox proportional hazards regression and Kaplan-Meier survival analyses were performed for prognostic factors including: age, Karnofsky performance score (KPS), O(6)-methylguanine methyltransferase (MGMT) promoter methylation status, EOR and RTV. EOR and RTV cut-off values for improved OS were determined and internally validated by receiver operator characteristic (ROC) analysis for 12-month overall survival. Half of the tumors had MGMT promoter methylation (77, 52%). The median tumor volume, EOR and RTV were 43.20 cc, 93.5%, and 3.80 cc respectively. Gross total resection was achieved in 52 patients (35%). Cox proportional hazards regression, ROC and maximum Youden index analyses for RTV and EOR showed that a cut-off value of <3.50 cc (HR 0.69; 95% CI 0.48-0.98) and ≥84% (HR 0.64; 95% CI 0.43-0.96) respectively conferred an overall survival advantage. Independent overall survival predictors were MGMT promoter methylation (adjusted HR 0.35; 95% CI 0.23-0.55) and a RTV of <3.50 cc (adjusted HR 0.53; 95% CI 0.29-0.95), but not EOR for incompletely resected glioblastomas.