Outcomes following stereotactic body radiation therapy specific to spinal metastases with paraspinal disease extension: does volume matter?

OBJECTIVE: Paraspinal involvement has been consistently reported as a negative predictor of local control following stereotactic body radiation therapy (SBRT) for spinal metastases. The aim of this study was to investigate the characteristics of paraspinal disease and determine the impact on outcomes. METHODS: Patients who underwent SBRT for spinal metastases with paraspinal involvement, identified from a single-institutional prospective database, were retrospectively reviewed. Those with prior radiation therapy/surgery were excluded. The treated clinical target volume (CTV) was segmented into paraspinal (CTVPS), neuroforaminal (CTVNF), epidural (CTVEP), and osseous bone (CTVbone) components. The extent of extraosseous disease was classified according to the involvement of rib, neuroforamina, and muscle. Volume and dosimetric parameters were collected and dichotomized using recursive binary partitioning. The outcomes of interest were the cumulative incidence of local failure (LF), overall survival (OS), and reirradiation rates. RESULTS: One hundred fourteen patients with 125 treated spinal sites were identified. There were 38% (47/125), 66% (82/125), and 19% (24/125) treated spinal sites with involvement of rib, neuroforamina, and muscle, respectively. The median follow-up duration of the cohort was 17. 34 months (IQR 7. 79-40. 11 months). The 12-month and 24-month cumulative incidence rates of LF were 19. 5% (95% CI 12. 6%-27. 4%) and 29. 8% (95% CI 21. 4%-38. 7%), respectively. The 12-month cumulative incidence rates of LF were 12. 0% (95% CI 5. 9%-20. 5%) and 36. 3% (95% CI 20. 2%-52. 6%) for patients with CTVPS < 42. 9 mL and those with ≥ 42. 9 mL (p < 0. 001), respectively, and 55. 6% (95% CI 28. 7%-75. 8%) and 12. 2% (95% CI 6. 5%-19. 9%) for patients with and without muscle invasion (p = 0. 001), respectively. In the multivariable analysis, only CTVPS remained statistically associated with LF. CTVPS ≥ 42. 9 mL was associated with 2. 3 times (95% CI 1. 13-4. 83, p = 0. 02) increased risk of LF compared with CTVPS < 42. 9 mL. The 12-month and 24-month OS rates were 56% (95% CI 47%-65%) and 41% (95% CI 32%-50%), respectively. Patients with an Eastern Cooperative Oncology Group performance status score < 1 and oligometastatic disease (≤ 5 metastases) were associated with better OS in the multivariable analysis. The 12-month and 24-month reirradiation rates were 7. 3% (95% CI 3. 4%-13. 3%) and 16. 5% (95% CI 10. 2%-24. 1%), respectively. CONCLUSIONS: Spinal metastases with high-volume paraspinal involvement were associated with increased risk of LF following SBRT, and strategies to optimize local control are required.