All Bacillus Calmette-Guérin (BCG) Strains Are Equal, but Some BCG Strains Are More Equal Than Others

Abstract

Thirty-seven years have elapsed since the first report by Morales et al. [1] on intravesical bacillus Calmette-Guerin (BCG) treatment for urothelial bladder carcinoma. One could be forgiven for expecting that clinical differences between BCG strains would be well elucidated by now; unfortunately, that is not the case. The reasons for assuming a lack of clinical equivalence between available BCG strains are becoming more obvious. They include (1) observed differences among BCG response rates in randomised trials comparing one particular BCG strain with other chemotherapeutic intravesical agents (this reason was cited as a rationale for a trial started in the early 1980s [2]), (2) perceived differences in BCG strains in their role as a vaccine for tuberculosis, and (3) differences among commercially available BCG strains in terms of their number of colonyforming units per standard dose [3]. To this list, we can add new, emerging data of sophisticated genetic differences elucidated among the BCG strains [4]. What are the reasons for the lack of comparative trials among different BCG strains? Due to BCG’s unique discovery and history of global distribution (reviewed by Gan et al. [4]), many early researchers simply used the strain of BCG available to them in their countries and focused their attentions on proving BCG efficacy (against other intravesical chemotherapeutic agents) and on how best to administer it. Another speculative explanation could be a lack of commercial interest from BCG manufacturers in initiating such trials. A lack of comparative effectiveness may have contributed to wide variation in intravesical BCG strain use around the world, with more than seven different strains used in the contemporary era. Some of the variation is related to changes in governance, as illustrated in Switzerland in 1994, changes in availability due to global production shortages [5], and a lack of definitive level 1 evidence favouring any one particular strain. The Swiss situation arose when, in 1994, BCG Pasteur was withdrawn from the market, leaving Connaught and TICE as the only available options. Admirably, sensing an opportunity to evaluate the efficacy of the available BCG strains, a prospective randomised trial was established, the results of which are reported by Rentsch et al. [6] in this issue of European Urology. In this nonblinded prospective randomised controlled trial, 132 of the originally randomised 142 patients with high-risk, non–muscle-invasive bladder cancer (NMIBC) received either BCG Connaught (n = 71) or BCG TICE (n = 60) as six weekly instillations following transurethral resection of bladder tumour. The 5-year recurrence-free survival rate was 74% for BCG Connaught and 48% for BCG TICE, which was significantly different ( p = 0.01); however, no statistically significant difference was observed for progressionfree-survival and overall survival. This is the second time that BCG TICE has been compared with another BCG strain: Vegt et al. [7] in the Dutch Cooperative Trial also compared BCG TICE and BCG RIVM (Dutch) and mitomycin C (a summary of comparative trials of different BCG strains is shown in Table 1). The Dutch Cooperative Trial eventually concluded that there was no difference between the strains but hinted at the possible inferiority of BCG TICE [7]. Between 1988 and 1991, a multicentre UK study successfully recruited 99 patients for a randomised trial comparing the UK BCG strain (Glaxo) and the Pasteur strain [8]. In this trial, no statistical difference was observed at 3 mo between the two strains. It is unfortunate that the trial E U R O P E A N U R O L O G Y X X X ( 2 0 1 4 ) X X X – X X X