RESUMO
OBJECTIVE: To analyse the impact of histological discordance of subtypes (subtypes or divergent differentiation [DD]) in specimens from transurethral resection (TUR) and radical cystectomy (RC) on the outcome of the patients with bladder cancer receiving RC. PATIENTS AND METHODS: We analysed data for 2570 patients from a Japanese nationwide cohort with bladder cancer treated with RC between January 2013 and December 2019 at 36 institutions. The non-urinary tract recurrence-free survival (NUTR-FS) and overall survival (OS) stratified by TUR or RC specimen histology were determined. We also elucidated the predictive factors for OS in patients with subtype/DD bladder cancer. RESULTS: At median follow-up of 36.9 months, 835 (32.4%) patients had NUTR, and 691 (26.9%) died. No statistically significant disparities in OS or NUTR-FS were observed when TUR specimens were classified as pure-urothelial carcinoma (UC), subtypes, DD, or non-UC. Among 2449 patients diagnosed with pure-UC or subtype/DD in their TUR specimens, there was discordance between the pathological diagnosis in TUR and RC specimens. Histological subtypes in RC specimens had a significant prognostic impact. When we focused on 345 patients with subtype/DD in TUR specimens, a multivariate Cox regression analysis identified pre-RC neutrophil-lymphocyte ratio and pathological stage as independent prognostic factors for OS (P = 0.016 and P = 0.001, respectively). The presence of sarcomatoid subtype in TUR specimens and lymphovascular invasion in RC specimens had a marginal effect (P = 0.069 and P = 0.056, respectively). CONCLUSION: This study demonstrated that the presence of subtype/DD in RC specimens but not in TUR specimens indicated a poor prognosis. In patients with subtype/DD in TUR specimens, pre-RC neutrophil-lymphocyte ratio and pathological stage were independent prognostic factors for OS.