Abstract
Purpose
The optimal sequencing of radiotherapy (RT) with surgery in soft-tissue sarcomas (STS) remains undefined. We assessed the impact of RT sequencing on overall survival (OS), cause-specific survival (CSS), local failure, and distant failure.
Methods
A retrospective analysis was conducted using the National Oncology Database, a proprietary database of aggregated tumour registries owned by Impac(r) Medical Systems (Sunnyvale, CA). Eligible sites were soft tissues of the head/neck, thorax, abdomen, pelvis, extremities, trunk, and peritoneum. Only patients with known stage and grade were included. Prognostic factors were identified with multivariate analysis (MVA) using the Cox proportional hazards model. Survival was calculated using the Kaplan-Meier method, and compared for statistical significance (p< 0.05) using the log-rank test.
Results
A total of 821 patients met the inclusion criteria. The median follow-up time for living patients was 62 months. The 5-year CSS was 69%. MVA identified the following independent predictors for CSS (p< 0.01): age, stage, grade, histology, surgery, RT sequence, and tumour size. CSS was significantly improved with pre-op RT versus post-op RT [hazard ratio (HR) 0.7, 95% confidence interval (CI) 0.51-0.94, p< 0.05], with a 5-year CSS of 81% and 73%, respectively (log-rank, p< 0.01). Pre-op RT improved CSS in patients with lower extremity tumours, leiomyosarcoma, and synovial sarcoma (p< 0.05). OS was not significantly improved with pre-op RT. Pre-op RT also resulted in significantly reduced local and distant relapse rates than post-op RT. Adverse prognostic factors were balanced between both groups.
Conclusion
Pre-operative RT has a significant benefit in reducing cancer-specific mortality compared to post-operative RT in STS.
