Development and validation of a novel prognostic model for overall survival in patients with micropapillary urothelial carcinoma.

e16618 Background: Micropapillary urothelial carcinoma (MUC) is a rare variant of bladder cancer with limited prognostic data in the literature. We hereby present a novel prognostic model for overall (OS), for patients with MUC, derived from the Surveillance, Epidemiology and End Results (SEER) database. Methods: The SEER database (18 registries/November 2020) was queried with the case listing method by using the ICD-10 topography codes C67.0-C67.9 (bladder cancer), and the morphologic code 8131 (MUC). Patients were randomly divided into a training (TC) and a validation cohort (VC) (7:3 ratio). Variables significantly associated with OS were identified with multivariate Cox regression and backwards stepwise conditional approach; each variable received “points” according to its beta coefficient, based on the formula: Probability of an event at time t=S 0 (t) exp(β 1 x 1 , β 2 x 2 ...) , (β = regression coefficients, x = observed covariate values, S0(t) = survival function). Cumulative risk points were assigned to each patient and three distinct risk categories for OS were constructed. Harrel’s optimism-corrected C-statistic with bootstrap resampling was used for internal (TC) and external (VC) validation. Discriminatory performance of the model was compared to the AJCC 8 th edition with the likelihood ratio test (LRT). Results: 548 patients were included in the study (374 TC, 174 VC). No statistically significant differences in baseline characteristics and survival outcomes were identified between TC and VC. Age, marital status, tumor size, AJCC stage, radical cystectomy, radiation and chemotherapy were significant variables inserted in the multivariate model. Patients were stratified to three risk groups (low, intermediate, high) based on the total risk points collected (150, Table). Median OS was NR (95%CI; NR, NR), 41 (95% CI; 25-57) and 14 months (95% CI; 11-17) for the low, intermediate, and high-risk categories respectively (log-rank p<.001). Optimism-corrected C-statistic for the prognostic score was 0.74 (95%CI;0.71-0.78) in the TC and 0.70 (95%CI; 0.65-0.78) in the VC. C-statistic for AJCC 8 th edition was 0.65 (95% CI; 0.61-0.69) and 0.62 (95% CI; 0.56-0.68) in the TC and VC respectively. The LRT demonstrated superior discriminatory performance of our prognostic model compared to AJCC (p<.001). Conclusions: We present a novel prognostic model for OS of patients with MUC composed of clinicopathologic and treatment variables. Our model had superior performance to the AJCC system and may serve as a useful tool for individualized risk assessment in patients with MUC. Table: see text