2544 - Treatment Patterns and Outcomes for Colorectal Cancer Spine Metastases Treated with SRS

D. Klebaner¹, J. A. Jaoude¹, A. Pratapneni¹, S. D. Chang², I. C. Gibbs¹, T. Kollipara¹, K. A. Kumar Jr³, E. Rahimy¹, D. Park², D. K. Fujimoto¹, M. Mendoza¹, A. Persad², S. G. Soltys¹, and E. Pollom¹; ¹Department of Radiation Oncology, Stanford University School of Medicine, Stanford, CA, ²Department of Neurosurgery, Stanford University School of Medicine, Stanford, CA, ³Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, TX

Purpose/Objective(s): There is evidence demonstrating that spine metastases from colorectal cancer (CRC) exhibit relative radioresistance, even in the setting of spinal stereotactic radiosurgery (SRS). We evaluated our institutional treatment patterns and outcomes for spine metastases from CRC treated with SRS. We hypothesized that right colon primary and SFED<18 Gy were associated with local failure (LF).
Materials/

Methods: This IRB-approved, retrospective analysis included patients with CRC non-sacral spine metastases treated with spine SRS between October 2002 and November 2021 with at least one instance of evaluable post-treatment spine imaging and no prior irradiation to the given segment. The exposures of interest were primary site (right colon, left colon, or rectum), epidural disease, fractionation, minimum dose to the GTV, and single fraction equivalent dose (SFED). The primary endpoint was cumulative incidence (CI) of LF at one year from SRS with death as a competing risk. We analyzed associations between exposures and LF using Gray’s test.


Results:

We included 36 patients with 52 spinal segments treated with SRS. Median follow-up after SRS was 8.3 months (interquartile range [IQR] 5.4 to 21 months). Of the 52 spinal segments, 12 arose from the rectum, 16 from the right colon, and 24 from the left colon. 23 (44 %) segments had some degree of epidural involvement, with 6 having a Bilsky score of 2 or 3. The median minimum SFED to the GTV was 15.1 Gy (IQR 12.0 to 18.3 Gy). Thirty-one segments (60%) received treatment with single fraction SRS.

The 1-year CI of LF was 16% (eight out of 52 levels). There was no significant difference in time to LF between cancers originating from the right colon, left colon, or rectum (P=0.70). There was increased 1-year CI LF for levels with Bilsky 1B or greater epidural disease (22% vs 11%), though this association was not significant. Fractionation was significantly associated with LF and SFED, with multi-fraction SRS (median SFED=16.4 Gy) associated with a 1-year local failure CI of 30%, compared to 6.7% for single fraction SRS (median SFED= 20 Gy, P=0.01). Minimum dose to the GTV was not associated with local failure, but there was a trend toward decreased 1-year CI LF for spinal levels receiving SFED of at least 18 Gy compared to less than 18 Gy (12% vs 24%, p=0.27). Median SFED used for treatment increased from 16.4 Gy prior to 2014 to 20 Gy 2014 and after.
Conclusion:

Though this analysis was limited by statistical power, we demonstrate that factors associated with local failure of CRC spine metastases include epidural disease and fractionation, likely mediated by SFED. Our institutional practices reflect using higher SFED over time for SRS treatments among this cohort. These findings support a need for further exploration of higher spinal cord tolerance limits as well as the safety and efficacy of dose-escalated fractionation for CRC spine metastases.