Introduction and First Clinical Application of a Simplified Immunohistochemical Validation System Confirms Prognostic Impact of KI-67 and CK20 for Stage T1 Urothelial Bladder Carcinoma: Single-Center Analysis of Eight Biomarkers in a Series of Three Hundred Six Patients

Stage pT1 is a challenging subentity of urothelial bladder carcinoma (UBC) because of its unpredictable clinical course. Prognostic markers are highly appreciated, but IHC markers are especially problematic because of high interobserver variability and varying cutoff values of this method. We proposed and used a simplified model of IHC analysis using a cutoff of < 10% vs. >= 10% expression for 8 biomarkers. Background: Biomarkers could help to estimate the prognosis of solid tumors. One of the reasons that many immunohistochemical (IHC) markers are not used routinely is the high interobserver variability and various cutoff values. In the present study, we used a simplified IHC method with a group of 8 biomarkers in stage pT1 urothelial bladder carcinoma (UBC). Patients and Methods: IHC expression of CK20, KI-67, STK15, MUC7, periostin, fibronectin, survivin, and CXCR4 was assessed independently by 2 reviewers in a series of 306 stage pT1 UBC specimens from a single center in 10% steps from < 10% up to > 90%. A general center < 10% vs. >= 10% was set for further analysis for all markers. All patients initially underwent a bladder-sparing approach. Kaplan-Meier analyses and multivariate Cox regression analyses of recurrence-free survival (RFS), progression-free survival (PFS), and cancer-specific survival (CSS) were performed. Results: A cutoff point >= 10% was shown to be valid and reliable for marker expression, with 96% interobserver agreement. Of the studied marker expressions, >= 10% for Ki-67 showed a statistically significant worse RFS (54% vs. 64%; P = .004), PFS (66% vs. 73%; P = .001), and CSS (71% vs. 77%; P = .015); >= 10% for CK20 showed a worse RFS (57% vs. 58%; P = .009). Multivariate Cox regression analysis revealed CK20 to be an independent prognostic factor for recurrence (hazard ratio [HR], 2.08; confidence interval [95% CI]; 1.21-3.57; P = .008) and Ki-67 for progression (HR, 2.11; CI, 1.02-4.37; P = .045). Conclusion: We proposed and applied a simplified IHC evaluation that increases interobserver agreement and confirms the prognostic role of Ki-67 and CK20 for stage T1 UBC.