Maintenance Bacillus Calmette-Guérin Treatment of Non-muscle-invasive Bladder Cancer: A Critical Evaluation of the Evidence
Ehdaie B, Sylvester R, Herr HW. Eur Urol. 2013 May 18. pii: S0302-2838(13)00493-4. doi: 10.1016/j.eururo.2013.05.027. [Epub ahead of print]


Urology Service, Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York, NY, USA; Health Outcomes Research Group, Department of Epidemiology and Biostatistics, Memorial Sloan-Kettering Cancer Center, New York, NY, USA. Electronic address:



Despite the effectiveness of bacillus Calmette-Guérin (BCG) therapy in non-muscle-invasive bladder cancer (NIMBC) to delay recurrence and disease progression, the evidence supporting maintenance treatment and its optimal duration is unkown.


The purposes of this paper are to critically review the evidence supporting the use of maintenance BCG after an initial series of induction instillations and to illustrate the factors contributing to current dilemmas in establishing the optimal duration of BCG treatment.


The following terms were used in Medline database searches for original articles published before February 1, 2013: bladder cancer, urothelial cancer, bacillus Calmette-Guérin, maintenance, and induction. All randomized controlled trials and meta-analyses, including those based on indirect comparisons, were evaluated.


Seven randomized studies compared induction BCG plus maintenance to induction alone, with or without retreatment with BCG on recurrence. All but one of these studies were underpowered and the largest study used a broad, composite end point: worsening-free survival. Seven meta-analyses have been conducted, three of which included data from observational cohort studies. They demonstrated the benefit of maintenance BCG to reduce disease recurrence and delay progression compared to various control groups; however, the analyses were based on suboptimal data. Although there is new evidence that 1 yr of maintenance BCG is sufficient treatment in intermediate-risk patients, the optimal duration of BCG maintenance remains unknown. A new randomized trial is proposed, which includes induction BCG with retreatment on recurrence as a control arm, to study this question.


The optimal duration of BCG treatment in patients with NMIBC remains unknown and should be the subject of further studies. We recommend that in addition to 3 yr of maintenance BCG, guideline panels also include 1 yr of therapy and induction BCG with retreatment on recurrence as a possible treatment options for patients with NMIBC, albeit with a lower level of evidence and grade of recommendation.