Temporalis Muscle Thickness as a Prognostic Factor for 30-day, 90-day, and Overall Mortality in Newly-Diagnosed Glioblastoma

Abstract

Abstract Background The identification of novel prognostic biomarkers for glioblastoma (GBM) can guide clinicians and patients in treatment approaches. Frailty, as measured by sarcopenia, has been proven to predict overall survival in other oncologic processes. Objective We evaluated whether sarcopenia, as measured by temporalis muscle thickness, predicted survival in GBM, and we compared its accuracy to other survival markers. Methods A prospective GBM database identified 257 patients undergoing initial diagnostic surgery. Sarcopenia was quantified by temporalis muscle thickness and grouped into tertiles. Mortality hazard ratios were calculated using multivariate analysis. Results After multivariate analysis, sarcopenia at the time of initial surgery was the only factor associated with mortality at 30 days postoperatively (OR 0.10, P = 0.030). Sarcopenia at initial surgery predicted 90-day postoperative mortality; the most sarcopenic patients (1st tertile) had greater mortality than those in the 2nd (OR 0.28, P = 0.021) and 3rd tertiles (OR 0.04, P = 0.003). Sarcopenia predicted overall mortality, greater in the 1st tertile than the 2nd (OR 0.41, P < 0.001) and 3rd tertiles (OR 0.41, P < 0.001). Sarcopenia compared favorably to other predictors of mortality, including initiation of postoperative temozolomide and radiation treatment (OR 0.27, P < 0.001), gross total resection (OR 0.54, P = 0.007), and MGMT methylation status (OR 0.44, P < 0.001). Kaplan-Meier survival curves represent differences in survival (Log-Rank p < 0.001) Conclusions Sarcopenia predicts postoperative 30-day, 90-day, and overall survival from diagnosis in GBM. The frailty/sarcopenia paradigm is independent of patient demographic, oncologic, genetic, surgical, and therapeutic factors. Temporalis muscle thickness assessment provides a simple method to help guide treatment decisions in this population.