Analysis of MRE11 and Mortality Among Adults With Muscle-Invasive Bladder Cancer Managed With Trimodality Therapy

IMPORTANCE Bladder-preserving trimodality therapy can be an effective alternative to radical cystectomy for treatment of muscle-invasive bladder cancer (MIBC), but biomarkers are needed to guide optimal patient selection. The DNA repair protein MRE11 is a candidate response biomarker that has not been validated in prospective cohorts using standardized measurement approaches. OBJECTIVE To evaluate MRE11 expression as a prognostic biomarker in MIBC patients receiving trimodality therapy using automated quantitative image analysis. DESIGN, SETTING, AND PARTICIPANTS This prognostic study analyzed patients with Ml BC pooled from 6 prospective phase I/II, II, or III trials of trimodality therapy (Radiation Therapy Oncology Group [RTOG] 8802, 8903, 9506, 9706, 9906, and 0233) across 37 participating institutions in North America from 1988 to 2007. Eligible patients had nonmetastatic MIBC and were enrolled in 1 of the 6 trimodality therapy clinical trials. Analyses were completed August 2020. EXPOSURES Trimodality therapy with transurethral bladder tumor resection and cisplatin-based chemoradiation therapy. MAIN OUTCOMES AND MEASURES MRE11 expression and association with disease-specific (bladder cancer) mortality (DSM), defined as death from bladder cancer. Pretreatment tumor tissues were processed for immunofluorescence with anti-MREll antibody and analyzed using automated quantitative image analysis to calculate a normalized score for MRE11 based on nuclear-tocytoplasmic (NC) signal ratio. RESULTS Of 465 patients from 6 trials, 168 patients had available tissue, of which 135 were analyzable for MREll expression (median age of 65 years [minimum-maximum, 34-90 years]; 111 [82.2%] men). Median (minimum-maximum) follow-up for alive patients was 5.0 (0.6-11.7) years. Median (Q1-Q3) MRE11 NC signal ratio was 2.41(1.49-3.34). Patients with an MRE11 NC ratio above 1.49 (ie, above first quartile) had a significantly lower DSM (HR, 0.50; 95% CI, 0.26-0.93: P = .03). The 4-year DSM was 41.0% (95% CI, 23.2%-58.0%) for patients with an MRE11 NC signal ratio of 1.49 or lower vs 21.0% (95% CI, 13.4%-29.8%) for a ratio above 1.49. MRE11 NC signal ratio was not significantly associated with overall survival (HR, 0.84; 95% CI, 0.49-1.44). CONCLUSIONS AND RELEVANCE Higher MRE11 NC signal ratios were associated with better DSM after trimodality therapy. Lower MRE11 NC signal ratios identified a poor prognosis subgroup that may benefit from intensification of therapy.