Impact of Divergent Differentiation and/or Histological Subtype of Urothelial Carcinoma on Patient Outcomes in the GETUG-AFU V05 VESPER Trial.

ABSTRACT

PURPOSE: Variant histology or divergent differentiation (VH/DD) of urothelial carcinoma (UC) may impact outcomes after neoadjuvant chemotherapy (NAC) in muscle-invasive bladder cancer. Our aim was to assess the pathological response and progression-free survival (PFS) of patients with VH/DD in the prospective VESPER clinical trial. MATERIALS AND METHODS: This post hoc study included 300 NAC-treated patients with available transurethral diagnostic slides. Presence and percentage of VH/DDs were reviewed. For pathological response, logistic regression models were computed to measure association with VH/DD. For PFS, the associations were estimated in Cox proportional hazard regression model. All models were adjusted for randomization arm. RESULTS: VH/DD was identified in 177/300 patients (59%) and was predominant (≥50%) in 85/177. Compared to pure UC, VH/DD (≥10% or ≥50%) was not associated with a difference in proportion of complete pathological response (ypT0N0; OR adjusted 0.79, 95% CI 0.49-1.29), downstaging (≤ypT1N0; OR adjusted 0.62, 95% CI 0.37-1.02), or with an increased hazard of PFS (HR adjusted 1.24, 95% CI 0.83-1.85). However, comparing specific VH/DD to pure UC, nested subtype was associated with decreased odds of complete pathological response (OR adjusted 0.33, 95% CI 0.12-0.88) and downstaging (OR adjusted 0.30, 95% CI 0.13-0.74), and an increased hazard of PFS was observed for UC with ≥ 50% squamous differentiation (HR adjusted 2.11, 95% CI 1.01-4.38) or micropapillary subtype (HR adjusted 2.03, 95% CI 0.98-4.22). CONCLUSIONS: In the VESPER trial, we did not observe evidence for association of VH/DD with outcomes after NAC, but the specific presence of a predominant squamous differentiation or micropapillary subtype may be associated with shorter PFS.