The changing reality of urothelial bladder cancer: should non-squamous variant histology be managed as a distinct clinical entity?

Abstract

Objectives: To assess the effect of non-squamous differentiation (non-SQD) variant histology on survival in muscle-invasive bladder urothelial cancer (UC).

Patients and methods: A cohort of 411 radical cystectomy (RC) cases performed with curative intent for muscle-invasive primary UC was identified between 2008 and June 2013. Survival analysis was evaluated using Kaplan-Meier methodology comparing non-variant (NV) + SQD histology to non-SQD variant histology (non-SQD variants). Multivariable cox proportional hazards regression assessed all-cause and disease-specific mortality.

Results: Of the 411 RC cases, 77 (19%) had non-SQD variant histology. The median overall survival (OS) for non-SQD variant histology was 28 months, whereas the NV+SQD group had not reached the median OS at 74 months (log-rank test P < 0.001). After adjusting for sex, age, pathological stage, and any systemic chemotherapy, patients with non-SQD variant histology at RC had a 1.57-times increased adjusted risk of all-cause mortality (P = 0.027) and 1.69-times increased risk of disease-specific mortality (P = 0.030) compared with NV+SQD patients.

Conclusions: While SQD behaves similarly to NV, non-SQD variant histology portends worse OS and disease-specific survival regardless of neoadjuvant or adjuvant chemotherapy and pathological stage. Non-SQD variants of UC could perhaps be considered a distinct clinical entity in UC with goals for developing new treatment algorithms through novel clinical trials.

Keywords: bladder urothelial cancer; radical cystectomy; survival; variant histology.