Objectives: Previous publications have suggested that patients developing local and/or systemic side effects to Bacillus Calmette-Guerin (BCG) have a better clinical result, however it is necessary to determine if toxicity is responsible for improved efficacy.
Methods: After transurethral resection, intravesical instillations of BCG were given during a 6-week induction course followed by 3-weekly maintenance courses at 3, 6, 12, 18, 24, 30 and 36 months. The prognostic importance of delaying or stopping BCG due to local and/or systemic side effects was assessed in 487 patients.
Results: Patients with local BCG side effects had a significantly longer time to first recurrence, suggesting that side effects are related to efficacy. However patients with a better outcome remain on study for a longer period of time and receive more BCG, thus increasing their risk to develop side effects. To prove whether toxicity is responsible for improved efficacy, the prognostic importance of toxicity occurring prior to the 6 month instillations was assessed using the landmark method. Neither local nor systemic BCG toxicity prior to 6 months was related to subsequent recurrence.
Conclusions: While a correlation between BCG toxicity and efficacy exists, this study does not confirm that BCG toxicity is actually responsible for an improved outcome.