Abstract
Background Sinonasal undifferentiated carcinoma (SNUC) is a rare, aggressive malignancy with a poor prognosis, and multimodal therapy is the standard of care. We sought to characterize treatment delays in SNUC managed with surgery and adjuvant radiation and to determine the impact on survival using the National Cancer Database (NCDB). Methods This was a retrospective, population-based cohort study of patients with SNUC between 2004 and 2016 in the NCDB. The intervals of diagnosis to surgery (DTS), surgery to radiation (SRT), and radiation duration (RTD) were examined. Recursive partitioning analysis (RPA) was performed to identify the variables with the greatest impact on survival. The association between treatment delay and overall survival (OS) was then assessed using multivariate Cox proportional hazards regression. Results Of 173 patients who met inclusion criteria, 65.9% were male, average age at diagnosis was 56.6 years, and 5-year OS was 48.1%. Median durations of DTS, SRT, and RTD were 18, 43, and 46 days, respectively. Predictors of treatment delay included Black race, government insurance excluding Medicare/Medicaid, and positive margins. RPA-derived optimal thresholds were 29, 28, and 38 days for DTS, SRT and RTD, respectively. On multivariate analysis, positive margins (hazard ratio [HR]: 4.82; 95% confidence interval [CI]: 2.28-10.2) and DTS less than 29 days (HR: 2.41; 95% CI: 1.23-4.73) were associated with worse OS. Conclusion Our results likely reflect the aggressive nature of the disease with surgeons taking more invasive disease to the operating room more quickly. Median treatment intervals described may serve as relevant national benchmarks.