Spine stereotactic radiosurgery for metastases from hepatobiliary malignancies: patient selection using PRISM scoring

Objective Dose escalation via stereotactic radiation therapy techniques has been necessary for hepatobiliary malignancies in the primary and oligometastatic setting, but such dose escalation is challenging for spine metastases due to spinal cord proximity. Here, we investigate the role of spine stereotactic radiosurgery (SSRS) in the management of such metastases. Methods We retrospectively reviewed patients treated with SSRS to spinal metastases from hepatobiliary malignancies between 2004 and 2017 at our Institution. We used the Kaplan-Meier method to calculate overall survival (OS) and local control (LC) and Cox regression analysis to identify factors associated with disease-related outcomes. Results We identified 28 patients treated to 43 spinal metastases with SSRS for either HCC or cholangiocarcinoma. The 1-year LC and OS were 85% and 23%, respectively. The median time to death was 6.2 months, while median time to local failure was not reached. Tumor volume > 60 cc (SHR 6.65, p = 0.03) and Bilsky >= 1c (SHR 4.73, p = 0.05) predicted for poorer LC, while BED10 > 81 Gy trended towards better local control (SHR 4.35, p = 0.08). Child-Pugh Class (HR 3.02, p = 0.003), higher PRISM Group (HR 3.49, p = 0.001), and systemic disease progression (HR 3.65, p = 0.001) were associated with worse mortality based on univariate modeling in patients treated with SSRS; on multivariate analysis, PRISM Group (HR 2.28, p = 0.03) and systemic disease progression (HR 2.67, p = 0.03) remained significant. Four patients (10%) developed compression deformity and one patient (2%) developed radiation neuritis. Conclusion SSRS provides durable local control in patients with metastatic hepatobiliary malignancies, with higher BED necessary to ensure excellent LC. PRISM scoring is a promising prognostic tool to aid SSRS patient selection.