Radical nephroureterectomy followed by adjuvant chemotherapy (RNU-AC) versus observation (RNU-O) in early-stage upper urinary tract cancers with variant histology (UUTC-VH).

Abstract

487 Background: Upper urinary tract cancers (UUTC) are less frequent and associated with poorer stage-for-stage prognosis compared to bladder cancer, with variant histology being an independent predictor of inferior outcomes. The POUT trial included only patients with predominantly urothelial tumors. We aimed to compare outcomes among patients with UUTC-VH who were treated with RNU-AC vs. RNU-O. Methods: We queried the National Cancer Database for adult patients with UUTC-VH diagnosed between 2004 and 2018. Only patients who underwent RNU with node-negative disease on pathological staging (pT2-4N0M0) were included and divided into two groups based on the postoperative treatment strategy - RNU-AC and RNU-O. Patients who received neoadjuvant chemotherapy were excluded from analyses. Fisher’s exact and Mann Whiney U tests were used to compare frequency distributions. Cox Proportional Hazards regression was employed for multivariate analysis of factors associated with overall survival. Models were adjusted for age, sex, race, income, educational level, clinical T stage, insurance status, and the Charlson Comorbidity Index. Results: A total of 522 patients were identified – 133 (25.5%) received RNU-AC while 389 (74.5%) underwent RNU-O. Patients in the RNU-AC group were younger (median 69 vs. 76 years, P <0.001). Patients with small cell (15.8% vs 4.9%), micropapillary (9.8% vs 5.9%) and adenocarcinoma (9% vs 6.7%) histologies were more likely while those with squamous histology was less likely to receive AC (38.3% vs 50.6%) (p < 0.001 for all comparisons). A significant majority of patients in each T stage were treated with AC – 87.1% of pT1, 73.2% of pT2, and 68.4% of pT3 (P = 0.009). Overall survival in the RNU-AC and RNU-O groups were comparable (median of 27 vs 24.1 months, log rank-P = 0.63). On multivariable analysis, neither AC nor histological subtype were not independently predictive of OS (HR for AC = 0.96, 95% CI 0.74-1.24, P = 0.75). Conclusions: This is the largest study to date evaluating outcomes with AC after RNU in UUTC-VH since these patients were largely excluded from AC clinical trials. We observed that AC was not associated with improved overall survival after RNU in this population.