LAUSR

BACKGROUND Despite recent improvements in treatment of glioblastoma (GBM), some patients still show a short survival. OBJECTIVE In this study, we sought to develop a new risk score for preoperative assessment of short-term survival (STS, < 6 months) in GBM patients. METHODS All adult patients that underwent surgical resection of GBM between 2004 and 2014 were included (n=379). Various demographic and clinical parameters, which are available at admission, were assessed. Variables were evaluated in univariate and multivariate analyses. The score was validated in a separate GBM cohort that underwent surgical resection between 2015 and 2018. RESULTS The following independent predictors of STS were integrated into a new score: body height (Small, <169 cm, 1point), arterial Hypertension (1point), patients' age (Older: ≤54 years - 0points, 55-74 years - 1point, ≥75 years - 2points), and poor clinical status (Reduced Karnofsky performance status scale: ≤60% - 2points, 70-80% - 1point, ≥90% - 0points). Therefore, the new risk score (SHORT [Term]) ranged from 0 to 6 points and showed a good accuracy of risk estimation for STS in GBM (AUC: 0.715). STS rates were 9.7%, 23.1% and 70% in GBM patients scoring <2 points, 2-4 points and >4 points respectively (P<0.0001). The score was successfully validated (AUC: 0.770). CONCLUSION This study suggests a risk score for preoperative assessment of STS risk in GBM patients. Still, this risk score needs external validation in larger patients' cohorts from other institutions. Our score might be a tool to facilitate treatment decisions in GBM patients prior to surgery.