SRS227 - Prognostic accuracy of the revised tokuhashi scoring system in breast cancer patients undergoing surgery for spinal metastases

Abstract Purpose Breast cancer is a leading cause of symptomatic spinal metastases and cancer mortality in women worldwide. The Revised Tokuhashi Scoring System (RTS) is one of the most widely utilised scoring systems for life expectancy in metastatic spinal disease, however, the predictive value of the RTS specific to breast cancer spinal metastases (BCSM) remains unexamined. This study aimed to evaluate the prognostic accuracy of the RTS for postoperative survival in patients undergoing surgery for BCSM. Methods A review was conducted of 90 patients treated surgically for BCSM at a National Spinal Injuries Unit between 2005 and 2020. RTS parameters were recorded at initial surgery and correlated with actual survival up to two years postoperatively. Survival analyses employed Kaplan-Meier and Cox regression models to assess RTS predictive accuracy and the prognostic value of individual parameters. Results There was no statistically significant correlation between RTS-predicted and actual postoperative survival (P = 0.06). The RTS predicted survival accurately in 45.6% of cases, overestimated in 37.8%, and underestimated in 11.1%. Among the six RTS parameters, only the presence and resectability of metastases to major internal organs significantly predicted survival (P = 0.03). We suggest an alternative scoring model combining general condition and organ metastases which showed improved prognostic value (P = 0.01). Conclusions The RTS lacks prognostic accuracy for patients with BCSM undergoing spinal surgery, primarily due to limited predictive value of parameters other than primary tumor type. These findings highlight the need for pathology-specific prognostic tools to better guide surgical decision-making in this population.