Stereotactic Body Radiation Therapy for Large Volume Solid Tumor Spinal Metastases Involving Three or More Contiguous Vertebral Levels

Abstract

Data for treating 3 or more contiguous vertebral bodies with SBRT is limited. The purpose of this study was to explore oncologic outcomes and toxicity of spine SBRT delivered to large volume solid tumor metastases involving 3 or more contiguous vertebral levels. Patients treated with spine SBRT for vertebral metastases involving 3 or more contiguous levels between 2009 and 2021 were retrospectively reviewed. Data on demographics, dosimetry, toxicity, and outcomes were collected. Radiographic local failure was defined as progressive disease on CT and/or MRI in the treatment volume compared to the pre-SBRT baseline. Local control (LC) and overall survival (OS) were calculated from end of SBRT using the Kaplan-Meier and log-rank test. A total of 141 patients were included with a median follow-up of 9.7 months. The majority of patients had either 3 (74%) or 4 (16%) involved contiguous vertebral levels. Twenty-two percent, 57%, 14%, and 7% of treated lesion began in the cervical, thoracic, lumbar, and sacral vertebral levels, respectively. The most common primary sites were NSCLC (15.6%), kidney (14.9%), and breast (13.5%). Thirty-two percent of patients had prior external beam radiation to the same vertebral level, 63% had prior surgery, and 43% had prior vertebral fracture. Paraspinal extension was identified in 26% at time of SBRT. Median pre-SBRT Bilsky grade was 2 (range 0-3). The median total prescription dose was 27 Gy (range: 12-43 Gy) with the most common dose (Gy)/fractionation schedules of 27/5 (32%), 30/5 (25%), and 25/5 (18%). The median Dmax was 4727 cGy (range 2070-7857) with a median prescription isodose of 58% (range 48-97). The median OS was 12.4 months (95% CI 9.6-15.3). 1- and 2- year OS were 51% and 35%, respectively. Local progression occurred across all histologies, most commonly in colon and NSCLC (16.7%). 1- and 2-year LC were 70% and 57%, respectively. In the radiation-naïve cohort, 1- and 2-year LC were 75% and 63%, respectively and 52% and 34% in the prior radiation group with no statistically significant difference in time to local progression (p = 0.075). There was no statistically significant difference in LC when comparing 3 versus 4 or more contiguous lesions (p = 0.66). Forty percent of patients that progressed underwent salvage treatment (58% surgery, 33% SBRT, 8% systemic therapy). Post-SBRT vertebral compression fracture rate was 11.6%. We present the largest series to date of patients treated with SBRT for large volume spinal metastases involving 3 or more contiguous levels. These data suggest reasonable local control and low toxicity with SBRT; therefore, SBRT should be considered a practical modality to offer this cohort.