External validation of E uropean O rganization for R esearch and T reatment of C ancer and S panish U rological C lub for O ncological T reatment scoring models to predict recurrence and progression in J apanese patients with non‐muscle invasive bladder cancer treated with bacillus C almette– G uérin

Objective To validate two prediction models ( E uropean O rganization for R esearch and T reatment of C ancer and S panish U rological C lub for O ncological T reatment) for recurrence and progression of non‐muscle invasive bladder cancer in J apanese patients who underwent bacillus C almette– G uérin instillation therapy. Methods From M arch 1985 to A pril 2007, data were analyzed from 366 patients who underwent transurethral resection of bladder tumor followed by bacillus C almette– G uérin instillation therapy. The ability of two scoring models to predict recurrence and progression was assessed by concordance index. Results For recurrence probability, the concordance index of the E uropean O rganization for R esearch and T reatment of C ancer and S panish U rological C lub for O ncological T reatment models was 0.514 and 0.576, respectively, which was lower than that (0.604) of a selected single prognostic factor (age) by our multivariate analysis. For progression probability, the concordance index of E uropean O rganization for R esearch and T reatment of C ancer and S panish U rological C lub for O ncological T reatment models was 0.693 and 0.764, respectively, which was higher than that (0.633) of a selected single factor ( T stage) by our multivariate analysis. The S panish U rological C lub for O ncological T reatment scoring system resulted in better stratification of tumor recurrence and progression when compared with the E uropean O rganization for R esearch and T reatment of C ancer model, probably because more patients underwent bacillus C almette– G uérin treatment in the S panish U rological C lub for O ncological T reatment cohort than in the E uropean O rganization for R esearch and T reatment of C ancer cohort. Conclusions The S panish U rological C lub for O ncological T reatment scoring system is a good predictor of tumor recurrence and progression in J apanese patients who underwent bacillus C almette– G uérin immunotherapy. A large prospective study is warranted to confirm the efficacy of this system.