Abstract
OBJECTIVE: To compare overall and cancer-specific outcomes between patients with upper tract urothelial carcinoma (UTUC) managed with either radical nephroureterectomy (RNU) or nephron-sparing measures (NSM) using a large population-based dataset. PATIENTS AND METHODS: Using Surveillance, Epidemiology, and End Results (SEER) data, patients diagnosed with low- or moderate-grade, localised non-invasive UTUC were stratified into two groups: those treated with RNU or NSM (observation, endoscopic ablation, or segmental ureterectomy). Cancer-specific mortality (CSM) and other-cause mortality (OCM) rates were determined using cumulative incidence estimators. Adjusting for clinical and pathological characteristics, the associations between surgical type, all-cause mortality and CSM were tested using Cox regressions and Fine and Gray regressions, respectively. RESULTS: Of 1227 patients [mean (sd) age 70.2 (11.00) years, 63.2% male] meeting inclusion criteria, 907 (73.9%) and 320 (26.1%) patients underwent RNU and NSM for low- or moderate-grade, low-stage UTUC from 1992 to 2008. Patients undergoing NSM were older (mean age 71.6 vs 69.7 years, P < 0.01) with a greater proportion of well-differentiated tumours (26.3% vs 18.0%, P = 0.001). While there were differences in OCM between the groups (P < 0.01), CSM trends were equivalent. After adjustment, RNU treatment was associated with improved non-cancer cause survival [hazard ratio (HR) 0.78, confidence interval [CI] 0.64-0.94) while no association with CSM was demonstrable (HR 0.89, CI 0.63-1.26). CONCLUSIONS: Patients with low- or moderate-grade, low-stage UTUC managed through NSM are older and are more likely to die of other causes, but they have similar CSM rates to those patients managed with RNU. These data may be useful when counselling patients with UTUC with significant competing comorbidities.