Factors impacting contemporary management of high-grade extremity sarcoma: An analysis of 12,020 patients.

Abstract

11017 Background: Previous studies noted racial/ethnic disparities in high-grade extremity soft tissue sarcoma (ESS) treatment and overall survival (OS). Retrospective series have noted worse OS for Amputation (Amp) vs limb salvage surgery (LSS) and for LSS alone vs LSS plus radiation (RT). Given superior functional outcomes, LSS is now favored over Amp when possible. This study examines racial/ethnic disparities in receipt of Amp vs LSS and impact on OS using modern data from a national registry. Methods: The National Cancer Database was used to identify patients (pts) with stage II-III, high-grade ESS diagnosed between 2004-2014 and treated definitively with 1) Amp alone, 2) LSS alone, 3) Preoperative RT [pre RT] + LSS or 4) LSS + Post-operative RT [post RT]. Multivariate analyses (MVA) utilized logistic regressions for patterns of local treatment and Cox proportional hazards regression for OS. The Kaplan-Meier method was used to estimate 5-year OS. Results: Among 12,020 pts, receipt of LSS vs Amp did not differ significantly by race, ethnicity, age, insurance status, income, or educational attainment on MVA. The rate of Amp was higher in academic centers (OR 2.42; p = .006) or integrated network programs (OR 2.30; p = .014) vs community programs and rural vs. Metro settings (OR 1.88; p = .035). Actuarial 5-year OS by treatment was Amp 43%, LSS 59%, pre RT + LSS 61%, and LSS + post RT 66% (p < .001). On MVA, OS was worse with Amp vs. LSS alone (HR 1.37; p < .001) while pre RT + LSS (HR 0.70; p < .001) and LSS + post RT (HR 0.72; p < .001) had improved OS vs LSS alone. Treatment at a comprehensive community, academic, or integrated network programs vs community program; private insurance vs none; Hispanic vs. non-Hispanic (HR 0.85, p = .048); and higher educational attainment were associated with improved OS. More comorbidities, other primary cancers, older age, and no transitions in care were associated with worse OS. Conclusions: The only racial/ethnic disparity identified when evaluating rates of Amp and OS for ESS was a small OS benefit for Hispanic pts. OS was inferior with Amp and best with LSS with either pre RT or post RT. OS was improved at non-community programs, potentially indicating a need for referral to experienced centers.