Objectives: To investigate whether the occurrence of side effects or discontinuance of bacille Calmette-Guérin (BCG) therapy because of side effects is associated with its therapeutic efficacy.
Methods: We analyzed the data from 145 patients who had had nonmuscle-invasive bladder cancer (Stage pTa, pT1, or pTis) and had undergone an initial course of adjuvant BCG therapy after transurethral resection of bladder tumor from 1996 and 2006 at Keio University Hospital. Side effects were classified as minor and major, and the association between the occurrence of side effects or discontinuance of BCG therapy because of side effects and tumor recurrence or progression was analyzed.
Results: Side effects occurred in 106 patients (73.1%) during BCG instillation. Of these 106 patients, 38 had major side effects and 95 had minor; 27 patients had both. BCG therapy was discontinued in 19 patients (13.1%) because of severe hematuria in 2, high fever in 8, and severe lower urinary tract symptoms in 9. Multivariate analyses demonstrated that discontinuance of BCG therapy (P = .018) was an independent predictor for tumor recurrence, in addition to multiplicity (P = .043). However, the occurrence of side effects was not an independent predictor for tumor recurrence (P = .935). Multivariate analyses also demonstrated that neither discontinuance of BCG therapy (P = .308) nor the occurrence of side effects (P = .333) was an independent predictor for tumor progression.
Conclusions: Discontinuation of BCG therapy has a significantly deleterious effect on reducing the incidence of tumor recurrence. When major side effects occur, it might be preferable to attempt to mitigate the major side effects to maintain the BCG therapy on schedule.