An international pooled analysis of SBRT outcomes to oligometastatic spine and non-spine bone metastases

Purpose: There is a paucity of data on SBRT to non-spine bone (NSB) lesions compared to spine metastases. We report local recurrence (LR), widespread progression (WSP), and overall survival (OS) for oligometastatic patients treated to bone lesions with SBRT and investigate the hypothesis that outcomes are different between patients with spine and non-spine bone oligometastatic disease. Methods: Patients with oligometastatic disease (≤5 cumulative extracranial metastases) treated with bone SBRT at 6 international institutions from 2007 to 2016 were reviewed. Fine and Gray competing risks and Cox regressions were used to analyze univariable and multivariable relationships between disease/treatment factors and outcomes. Results: In total, 288 spine and 233 NSB lesions are reported in 356 patients. Cumulative incidence of LR across all bone lesions was 6.3%, 12.6% and 19.3% at 6 mo, 1 yr and 2 yrs. While univariable analysis suggested inferior LC and OS in spine patients, this did not hold true in multivariable analysis. The final regression model for LR in NSB lesions included PTV ≥ median of 31.8 cc (HR 5.02, p = 0.014) and primary histology, with RCC and NSCLC conferring a 10.8- and 6.5-fold increased risk of LR compared to prostate histology, respectively. The spine LR model included radioresistant histology (HR 2.11, p = 0.0051), PTV Dmin (BED10) ≥ median of 19.1 Gy (HR 0.46, p = 0.0085), and epidural disease (HR 1.99, p = 0.016). Conclusion: This large multi-institutional series reports comparably excellent response to SBRT for a balanced distribution of oligometastatic NSB and spine lesions. Dose escalation for large and/or radioresistant NSB lesions should be explored, given the typical lack of an immediately adjacent dose-limiting critical structure.