The immunotherapy paradox in high-risk non-muscle-invasive bladder cancer: A tale of three trials

For five decades, intravesical Bacillus Calmette-Guérin (BCG) has remained the cornerstone of adjuvant therapy for patients with high-risk non-muscle-invasive bladder cancer (NMIBC). Nonetheless, up to 40% of patients experience recurrence or progression within 2 years, so there is a need for strategies that reinforce immune durability without compromising bladder preservation.[1] These limitations have driven efforts to develop systemic immunotherapeutic strategies that augment BCG’s local immune effect while preserving quality of life, based on the rationale that BCG-induced inflammation upregulates programmed death ligand 1 (PD-L1)–mediated immune exhaustion, which can be reversed by checkpoint inhibition to restore cytotoxic T-cell activity and prolong response. Building on the efficacy of pembrolizumab in BCG-unresponsive NMIBC, three phase III trials, POTOMAC, CREST, and ALBAN, evaluated systemic or subcutaneous immune checkpoint inhibitor (ICI) that blocks programmed death receptor (PD) and PD-L1 inhibition along with intravescial BCG in the BCG-naïve setting [Table 1].Table 1: Comparison of phase III trials of immune checkpoint inhibitor plus Bacillus Calmette-Guérin therapy in Bacillus Calmette-Guérin-naïve high-risk non-muscle-invasive bladder cancerThe phase III POTOMAC trial showed that adding 1 year of durvalumab to standard BCG induction and maintenance significantly improved event-free survival (EFS) in BCG-naïve high-risk NMIBC (Hazard Ratio [HR] 0.68), achieving a 32% reduction in recurrence risk with a manageable safety profile despite higher grade 3/4 adverse events (21% vs. 4%) that did not compromise BCG adherence.[2] The global phase III CREST trial demonstrated that adding subcutaneous PD-1 inhibitor sasanlimab to standard BCG induction and maintenance significantly improved EFS versus BCG alone (HR 0.68; P = 0.0095). In carcinoma in situ (CIS), responses were durable, with 91.7% maintaining a complete response at 36 months compared with 67.7% with BCG alone.[3] The phase III ALBAN trial found that adding intravenous atezolizumab to standard BCG in BCG-naïve high-risk NMIBC did not improve EFS (HR 0.98; P = 0.9106) and increased treatment-related adverse events.[4] These negative results contrast with CREST and POTOMAC, suggesting PD-L1 inhibition does not uniformly augment BCG efficacy. A meta-analysis by Scilipoti et al. (n = 2590; POTOMAC, CREST, ALBAN) showed a modest EFS benefit with ICI–BCG (HR 0.77), driven by prolonged-maintenance POTOMAC/CREST (HR 0.68) but attenuated by ALBAN and offset by higher grade ≥3 toxicity and no survival gain.[5] These studies raise important questions. The negative ALBAN results may relate to lower median BCG doses and a higher CIS burden. Moreover, the modest 5%–7% EFS gain, alongside significantly increased grade 3–4 treatment-related toxicity, may be statistically significant but of limited clinical relevance. For the Indian scenario, one has to analyse these results considering the theoretically higher efficacy rates of BCG due to endemic tuberculosis or universal childhood BCG vaccination (that may enhance intravesical BCG efficacy via trained immunity). Furthermore, Indian series have reported moderate-to-severe BCG toxicity in up to two-thirds of patients;[6] adding ICIs, with greater adverse events, requires cautious use in resource-limited settings. Finally, rates of CIS diagnosis may be limited in our country due to the predominant use of white light cystoscopy and the absence of trained pathologists. Therefore, unlike other drug combinations, combining ICI with intravesical BCG requires rigorous studies in our country before advocating their use for this indication. Financial support and sponsorship: Nil. Conflicts of interest: There are no conflicts of interest.