Cumulative dose, toxicity, and outcomes of spinal metastases re-irradiation : Systematic review on behalf of the Re-Irradiation Working Group of the Italian Association of Radiotherapy and Clinical Oncology (AIRO).

PURPOSE: The aim of this study was to identify patient-, tumor-, or treatment-related factors which may affect disease-related outcomes of re-irradiation (reRT) in patients with previously irradiated vertebral metastases. METHODS: A computerized search of the literature was performed by searching for terms related to reRT and spinal metastases in MEDLINE, EMBASE, OVID, and the Cochrane database from 1995 to 2019. Studies including at least 10 patients who had received reRT at the same site of initial radiotherapy for vertebral metastases with localized external beam radiotherapy were included. To determine the pooled ≥G3 acute and late toxicity rate, pain relief, local control, and overall survival, a meta-analysis technique of single-arm studies was performed. RESULTS: Nineteen studies including 1373 patients met the inclusion criteria for this systematic review. The pooled pain relief, neurological improvement, 1­year local control, and 1­year overall survival rates were 74.3%, 73.8%, 78.8%, and 54.6%, respectively, with moderate to high heterogeneity among studies. No difference in heterogeneity was evidenced for pain relief or local control after omitting studies not using stereotactic body radiotherapy (SBRT) or studies delivering biologically effective dose (BED) < 45 Gy10, whereas heterogeneity for 1­year OS was lower after omitting studies not using SBRT and delivering BED < 45 Gy10. The pooled results of grade ≥ 3 acute and late toxicity were 0.4% (95% confidence interval: 0.1-1.2%) and 2.2% (95% confidence interval: 1.2-37%), respectively, with low heterogeneity among studies. CONCLUSION: While this systematic review confirmed that reRT is both safe and effective for treating patients with recurrent spinal metastases, it could not identify factors which may affect outcomes of reRT in this patient population.