Abstract
4575 Background: Urothelial carcinoma of the bladder (UCB) is a disease of alterations in several cellular pathways. Routine molecular profiling studies do not account for smoking, a well established risk factor for UCB, and its influence on outcome. This study assessed the prognostic potential of a multi-pathway protein panel across all UCB stages in a population-based cohort after accounting for clinicopathologic factors and smoking history. Methods: 212 UCB patients from the LA CSP, part of the NCI/SEER cancer registry, were included. "Smoking intensity" analyzed biologic and molecular impact of smoking by combining smoking status, duration of smoking and number of cigarettes smoked daily into a composite covariate. Tumors were profiled for Bax, caspase-3, Apaf-1, Bcl-2, p53, p21, COX2, VEGF and E-cadherin alterations by IHC. Univariate analyses and multivariable modeling examined associations with outcome. Results: Median follow up was 13.2 years. Age, pathologic stage, adjuvant therapy (all p< 0.001) and surgical modality (p = 0.05) were associated with survival. Increasing smoking intensity was associated with worse outcome (P < 0.001). Apaf-1 (p = 0.005), E-cadherin (p = 0.014) and p53 (p = 0.032) were univariately prognostic; E-cadherin remained prognostic after multivariate analysis (p = 0.04). Combined alterations in all 9 biomarkers were prognostic by univariate (p < 0.001) and multivariate (p = 0.006) analysis. A multivariate model that included all 9 biomarkers and smoking intensity was more accurate in predicting prognosis than models comprising of standard clinicopathologic covariates without (p < 0.001) or with (p = 0.018) smoking intensity. Conclusions: The study confirms detrimental effects of smoking on UCB prognosis. Apaf-1, E-cadherin and p53 individually predicted UCB survival. Increasing number of biomarker alterations was significantly associated with worsening survival, although markers contained in the panel were not necessarily prognostic individually. Predictive value of the nine-biomarker panel with smoking intensity was significantly higher than that of routine clinicopathologic parameters alone.