Bacillus Calmette-Guerin strain may not effect recurrence-free survival when used intravesically with interferon-alpha2b for non–muscle-invasive bladder cancer

Abstract

ObjectivesAdjuvant intravesical Bacillus Calmette-Guerin (BCG) remains the standard-of-care for high-grade non–muscle-invasive bladder cancer (NMIBC). Conflicting reports exist regarding disparate outcomes among BCG strains. We sought to determine whether a difference in recurrence-free survival (RFS) existed between TICE BCG and Connaught BCG strains used with interferon (IFN) for the treatment of NMBIC. Materials and methodsA post hoc analysis of the phase 2 BCG/IFN study, conducted from May 1999 to February 2001. A total of 901 patients had sufficient records for analysis. Enrollment criteria were liberal and included primary and recurrent NMIBC, patients with and without carcinoma in situ, and patients with prior BCG failure. At the beginning, 3 to 8 weeks after transurethral resection or biopsy, patients received induction with 6 weekly intravesical treatments of BCG (TICE or Connaught) with 50 million units of IFN. Surveillance for recurrence began 4 to 6 weeks after induction and quarterly thereafter for 2 years. If no recurrence was identified, patients received maintenance therapy. Separate models were created for BCG naïve and failure patients. Multivariable analysis was performed using Cox proportional hazards regression. ResultsOverall, 609 patients received TICE BCG and 292 received Connaught BCG with similar baseline characteristics. BCG strain was not associated with worse RFS in both the multivariable BCG naïve model (P = 0.28) and BCG failure model (P = 0.53). Duration of disease, tumor focality, tumor size, and BCG failure interval (in the BCG failure model) were associated with worse RFS. ConclusionNo significant difference in RFS was observed among patients treated with TICE BCG or Connaught BCG in combination with IFN.