Stereotactic Body Radiotherapy (SBRT) for Sacral Metastases: Deviation from Recommended Target Volume Delineation Predicts Higher Risk of Local Failure

Abstract

A international consensus recommendation was published to guide target volume delineation specific to sacral stereotactic body radiotherapy (SBRT). We report outcomes after sacrum SBRT, focusing on the impact of contouring deviation on local failure (LF) risk, with an aim to validate this guideline. All patients who underwent SBRT to any level between S1 and S5 from 2010 to 2021 were identified from a prospectively maintained institutional database. The primary outcome was magnetic resonance-based LF. Secondary outcomes included vertebral compression fracture (VCF) and overall survival (OS). Cumulative LF and VCF rates were calculated per segment using the competing risk analysis method. Kaplan Meier analysis was used to estimate OS per patient. Cox proportional hazards model was used to assess predictive factors of LF, VCF, and OS. A total of 215 treated sacral segments in 112 patients were retrospectively reviewed. The median follow-up was 13 months (range, 0.4-116.9). The median age was 64 years (range, 18-86), and 56% were male. Most patients (52%) had treatment to a single segment. The median clinical target volume (CTV) was 129.2 cc (range, 5.8-753.5). Most segments were treated with 30 Gy/4 fractions (51%), 24 Gy/2 fractions (31%), or 30 Gy/5 fractions (10%). Thirty-one percent of segments were of radioresistant histology (gastrointestinal, kidney, melanoma, sarcoma, or thyroid primary), and 51% had extraosseous disease. Sixteen percent of segments were under-contoured per consensus guidelines, with incomplete coverage of the involved sector (71%), omission of the adjacent uninvolved sector (17%), or both (11%) as the causes for deviation. The cumulative incidence of LF was 18.4% (95% CI 13.5-24.0) at 12-months and 23.1% (95% CI 17.6-29.0) at 24-months. On multivariate analysis (MVA), under-contouring (HR 2.4, 95% CI 1.3-4.7, p = 0.008), radioresistant histology (HR 2.4, 95% CI 1,4-4.1, p = 0.001), and extraosseous extension (HR 2.5, 95% CI 1.3-4.7, p = 0.005) were predictors of increased risk of LF. The LF rates at 12/24-months were 15.1%/18.8% for segments contoured per guideline versus 31.4%/40.0% for those under-contoured. The cumulative incidence of VCF was 7.1% (95% CI 4.1-11.1) at 12-months and 12.3% (95% CI 8.2-17.2) at 24-months. On MVA, female gender was the only risk factor for VCF (HR 2.3, 95% CI 1.1-5.2, p = 0.04). The median OS was 29.5 months (95% CI 17.5-59.2). On MVA, primary kidney (HR 4.7, 95% CI 1.7-12.5, p = 0.002) or lung histology (HR 3.4, 95% CI 1.3-8.5, p = 0.010), the presence of liver (HR 2.8, 95% CI 1.2-6.4, p = 0.016) or lung (HR 2.5, 95% CI 1.3-5.1, p = 0.008) metastases, ECOG performance status 2 or 3 (HR 3.3, 95% CI 1.2-8.2, p = 0.013), and the presence of sensory or motor deficit (HR 2.6, 95% CI 1.2-5.4, p = 0.012) were prognostic for worse OS. Sacral SBRT is associated with high rates of efficacy and an acceptable VCF risk. Adherence to target volume delineation consensus guidelines reduces the risk of LF.