P08.04.B The Clinical Frailty Scale is superior to the Karnofsky Performance Status as predictor of overall survival in patients with surgical treatment of brain metastases

Abstract

Abstract Background The Clinical Frailty Scale (CFS) evaluates patients’ level of frailty on a scale from 1 (very fit) to 9 (terminally ill) and is commonly used in geriatric medicine, intensive care and orthopaedics but not in patients harboring neuro-oncological diseases. Our study was conducted to reveal if the usage of CFS generates more reliable prediction of overall survival in patients after brain metastases resection rather than Karnofsky Performance Status (KPS). Material and Methods All patients which were operated for brain metastatic disease at our department from 2005-2019 were included. CFS and KPS were retrospectively assessed for the timepoints pre- and postoperatively as well during follow-up 3-6 months after resection. Results 205 patients with mean follow-up of 22.8 months (CI95% 18.4-27.1) were evaluated. Mean estimated OS was 32.1 months (CI95% 25.0-39.1). CFS showed a median of 3 points (IqR 2-4) at all 3 assessment-points which means patients were “managing well”. Median KPS was 80 preoperative (IqR 80-90) and 90 (IqR 80-100) postoperative as well as on follow-up after 3-6 months. CFS strongly correlated with KPS: preoperatively (r=-0.92; p< 0.001), postoperatively (r=-0.85; p<0.001) and at follow-up (r=-0.93; p<0.001). In the same time, CFS pre- and postoperatively showed only weak correlation with CFS at follow-up after 3-6 months (r=0.30, p<0.001; and r=0.37, p<0.001, correspondingly). In multivariate integrated Cox regression model, the CFS predicted the expected reduction of OS superior to KPS at all 3 assessment-points. One point increase of preoperative CFS represented 30% additional hazard to decease (HR=1.30, CI95% 1.15-1.46; p<0.001), correspondingly postoperative CFS provides 39% (HR=1.39, CI95% 1.25-1.54; p<0.001) and at follow-up 42% of hazard (HR= 1.42, CI95% 1.27-1.59; p<0.001). In case of KPS, decrease of 10 points resulted in additional hazards to decease of 26% (HR=0.974/point, CI95% 0.962-0.987; p<0.001) postoperatively 14% (HR=0.986/point, CI95% 0.978-0.993; p<0.001) and 31% (HR=0.969/point, CI95% 0.959-0.978; p<0.001) at follow-up. Conclusion CFS is a feasible and reliable performance scoring in patients undergoing brain metastasis resection, that provides better OS prediction as compared to KPS. Whereas, CFS at follow-up after 3-6 months specifies the expected OS most accurately. Due to weak correlation between CFS preoperatively vs. postoperatively and in 3-6 months, initial frailty does not predict the patients’ postoperative frailty score.