Background: Although cisplatin-based neoadjuvant chemotherapy (NAC) has demonstrated an overall survival (OS) benefit in MIBC, only a subset of patients have pathologic complete response (pT0) at cystectomy. ATM, RB1 and FANCC mutations have shown correlation with pT0 to cisplatin-based NAC, as previously published. We now report updated OS and disease specific survival (DSS) from two phase II trials using these gene alterations as biomarkers. Methods: Patients with stage T2-T4 (N0 or N1) MIBC were enrolled in phase II trials of dose-dense NAC with MVAC (methotrexate, vinblastine, adriamycin, and cisplatin; NCT01031420) or GC (gemcitabine and cisplatin; NCT01611662). Patients were treated with NAC with plan for curative cystectomy. DNA from pretreatment tumor tissue was sequenced for coding exons of 287 cancer-related genes and analyzed for mutations. Survival in patients with one or more mutations in ATM, RB1, or FANCC genes was compared to those without mutations. Results: Of 58 pts treated, 38% (22/58 pts) had relevant mutations in the combined group of MVAC (13/34 pts) and GC (9/24 pts) trials. At a median follow-up of 56 months and minimum follow up of 16 months, patients with mutations had statistically significantly greater OS (p = 0.0043) and DSS (p = 0.0015). Median OS/DSS was not reached for patients with a mutation in any group. At 5 years post treatment, OS/DSS were greater in mutated vs non-mutated patients in all groups (see table). Conclusions: Long-term follow up reveals that previously reported improved responses to cisplatin-based NAC associated with mutations in ATM, RB1 and FANCC also confer a clinically meaningful and statistically significant survival benefit in these patients. These alterations may be useful as predictive biomarkers to allow clinicians to prioritize patients most likely to benefit from NAC prior to radical cystectomy.