Background: The histological variants of urothelial carcinoma (UC) may impact the outcomes after neoadjuvant chemotherapy in muscle invasive bladder cancer (MIBC). Previous studies provided contradictory results because of bias related to their retrospective design. Our aim was to assess pathological response and 3-year (yr) progression free survival (PFS) according to UC variants for patients included within the French prospective Vesper clinical trial (NCT01812369). Methods: 493 patients received dose-dense Methotrexate, Vinblastine, Doxorubicin and Cisplatin (dd-MVAC) or Gemcitabine and Cisplatin (GC) after randomization in the VESPER trial. This ancillary study was restricted to 437 patients treated in neoadjuvant setting. A central pathological review of initial transurethral diagnostic specimens allowed to assess UC variants (if ≥10% tumor area) and the presence of tumor emboli, perineural invasion, and CIS. Results: Slides were available for central review in 303 cases. UC variants were present in 180 patients (59%). Tumor emboli, perineural invasion, and CIS were noticed in 119 (39%), 24 (8%) and 143 (47%) cases. In comparison with pure UC, the presence of UC variants was not associated with different complete pathological response (ypT0N0) (OR 0.78, 95% CI 0.48-1.27) or 3 yr PFS (HR 1.25, 95%CI 0.83-1.86). Tumor emboli were negatively associated with ypT0N0 (OR 0.57, 95%CI 0.35-0.95, p = 0.029), organ confined disease (OR 0.44, 95%CI 0.26-0.75, p = 0.002) and 3 yr PFS (HR 1.81, 95%CI 1.23-2.67, p = 0.003). Perineural invasion were negatively associated with organ confined disease (OR 0.29, 95%CI 0.12-0.68, p = 0.005) and 3 yr PFS (HR 2.54, 95%CI 1.47-4.40, p < 0.001). There was no association for CIS. Conclusions: In the VESPER trial, UC variants were not associated with distinct pathological response or 3 yr PFS after neoadjuvant chemotherapy. Future studies will determine whether molecular subtypes are associated or not with distinct outcome.