Predictive model for systemic recurrence following cisplatin-based neoadjuvant chemotherapy and radical nephroureterectomy for high risk upper tract urothelial carcinoma.

INTRODUCTION:Neoadjuvant chemotherapy (NAC) is increasingly used prior to radical nephroureterectomy (RNU) for upper tract urothelial carcinoma (UTUC). Systemic recurrence (SR) carries a dismal prognosis. We sought to determine risk factors associated with SR in this setting. METHODS:We evaluated a multi-center database of patients with UTUC who received cisplatin-based NAC before RNU. Final pathology at RNU was dichotomized into ypT<2 vs ypT≥2. Univariable and multivariable analyses were performed to identify risk factors associated with SR. Three groups were defined based on the number of significant risk factors (groups 1, 2, 3 for 0-1, 2, 3 risk factors, respectively) and evaluated for recurrence-free survival (RFS) using the Kaplan-Meier method. RESULTS:106 patients were identified between 2004 and 2018. Median age was 67.0 years [IQR = 61-73.3]; 57 (54%) and 49 (46 %) patients received MVAC and GC, respectively. Final pathological stage was ypT<2 in 57 (54%); 23% (24/106) had SR. On univariable analysis, pathological variables on final specimen including ypT≥2, lymphovascular invasion (ypLVI), and nodal involvement were associated with SR. On multivariable analysis, ypLVI OR = 4.1 (95% CI 1.2-13.6; P = 0.024) and pathological nodal involvement OR = 4.5 (95% CI 1.3-15.7; P = 0.017) were predictive of recurrence. Stratifying by the number of risk factors, the 2-year RFS was 95%, 55%, and 18% for groups 1, 2, and 3 respectively (log-rank <0.001). CONCLUSION:This model evaluates the risk of SR following NAC and RNU to guide counseling and decision-making after surgery. Adverse pathological variable including ypLVI and nodal involvement, in combination with ypT-stage, are strongly associated with SR.