Noninvasive Papillary Urothelial Carcinoma of the Bladder: An Institutional Experience Focusing on Tumors With Borderline Features.

CONTEXT.—:Noninvasive papillary urothelial carcinomas (PUCs) comprise most urinary bladder tumors. Distinction between low-grade (LG-PUC) and high-grade (HG-PUC) PUCs is pivotal for determining prognosis and subsequent treatment. OBJECTIVE.—:To investigate the histologic characteristics of tumors with borderline features between LG-PUC and HG-PUC, focusing on the risk of recurrence and progression. DESIGN.—:We reviewed the clinicopathologic parameters of noninvasive PUC. Tumors with borderline features were subcategorized as follows: tumors that look like LG-PUC but have occasional pleomorphic nuclei (1-BORD-NUP) or elevated mitotic count (2-BORD-MIT), and tumors with side-by-side distinct LG-PUC and less than 50% HG-PUC (3-BORD-MIXED). Recurrence-free, total progression-free, and specific invasion-free survival curves were derived from the Kaplan-Meier method, and Cox regression analysis was performed. RESULTS.—:A total of 138 patients with noninvasive PUC were included, with the following distribution: LG-PUC (n = 52; 38%), HG-PUC (n = 34; 25%), BORD-NUP (n = 21; 15%), BORD-MIT (n = 14; 10%), and BORD-MIXED (n = 17; 12%). Median (interquartile range) follow-up was 44.2 months (29.9-73.1 months). Invasion-free survival was different between the 5 groups (P = .004), and pairwise comparison showed that HG-PUC had a worse prognosis compared with LG-PUC (P ≤ .001). On univariate Cox analysis, HG-PUC and BORD-NUP were 10.5 times (95% CI, 2.3-48.3; P = .003) and 5.9 times (95% CI, 1.1-31.9; P = .04) more likely to invade, respectively, when compared to LG-PUC. CONCLUSIONS.—:Our findings confirm a continuous spectrum of histologic changes in PUC. Approximately a third of noninvasive PUCs show borderline features between LG-PUC and HG-PUC. Compared with LG-PUC, BORD-NUP and HG-PUC were more likely to invade on follow-up. BORD-MIXED tumors did not statistically behave differently from LG-PUC.