Background: Clinical application of a readily accessible biomarker for identification of bladder cancer patients most likely to respond to neoadjuvant chemotherapy (NAC) could help clinicians avoid unnecessary chemotherapy and prevent its subsequent complications in patients unlikely to drive clinical benefit. The primary objective of this study was to investigate the association of immunohistochemical markers of tumor subtype with response to NAC and survival of muscle-invasive bladder cancer (MIBC) patients. Methods: MIBC patients treated with NAC in two tertiary referral hospitals were retrospectively included in the study. The tissue microarrays (TMAs) were assembled from transurethral resection of bladder tumor (TURBT) specimens and immunohistochemistry (IHC) was performed on the TMA slides. The association of independent variables, including singular IHC markers, combined IHC markers, and clinical covariates with clinical complete response (CR) to NAC, and overall survival (OS) was assessed using logistic regression and Cox proportional hazard regression analysis, respectively. Kaplan-Meier curves were plotted for different IHC-based subtypes. Results: Data from 140 MIBC patients treated with NAC were retrospectively reviewed. A total of 63 patients with available TURBT specimen were eligible to be included in the analysis. Our results showed that the IHC signature of KRT5/6(+)/KRT20(-), as a combined marker of basal subtype, was the only covariate significantly associated with the CR to NAC (p = 0.037). After a median follow-up of 41 (range, 12-76) months, no statistically significant differences in OS were found between different IHC-based subtypes (Log rank P = 0.721). In Cox proportional hazard regression analysis, age >65yr was independently associated with poorer OS after NAC (hazard ratio [HR], 2.26; 95% confidence interval [CI], 1.02–5.05), but failed to remain significant after adjusting for creatinine clearance (HR, 1.54; 95% CI, 0.58–4.10).Conclusion: The IHC expression of KRT5/6 and KRT20, as a readily accessible combined marker may help us to identify patients most likely to benefit from chemotherapy. The clinical utility needs to be established in larger prospective studies.