Systemic Atezolizumab and Intravesical BCG in High-Risk NMIBC: Lessons from the ALBAN (GETUG-AFU 37) Phase III Trial

Highlights

• The ALBAN trial (GETUG-AFU 37) is a pivotal phase III study investigating the synergy between systemic PD-L1 inhibition and local intravesical BCG in treatment-naive high-risk NMIBC.
• The addition of intravenous atezolizumab to a 1-year BCG regimen failed to improve investigator-assessed event-free survival (EFS) compared to BCG monotherapy (HR 0.98).
• Safety data revealed a higher incidence of treatment-related adverse events (TRAEs) in the combination arm, consistent with the known systemic profiles of checkpoint inhibitors.
• Discrepancies between ALBAN and other PD-(L)1/BCG combination trials suggest that therapeutic success in NMIBC may be sensitive to specific agent characteristics and trial design.

Background

Non-muscle-invasive bladder cancer (NMIBC) represents approximately 75% of all new bladder cancer diagnoses. Within this group, patients with high-risk features—including T1 stage, high-grade Ta tumors, or the presence of carcinoma in situ (CIS)—face a significant risk of recurrence and progression to muscle-invasive disease. For decades, the standard of care following transurethral resection of bladder tumor (TURBT) has been intravesical Bacillus Calmette-Guérin (BCG) therapy. BCG works by inducing a robust local inflammatory response and recruiting diverse immune effector cells to the bladder urothelium.

Despite the efficacy of BCG, approximately 30% to 40% of patients experience recurrence or progression within five years. Management of BCG-unresponsive NMIBC is challenging, often necessitating radical cystectomy, a procedure with high morbidity. The rationale for combining immune checkpoint inhibitors (ICIs) with BCG stems from the observation that BCG induction upregulates PD-L1 expression in the tumor microenvironment. It was hypothesized that blocking the PD-1/PD-L1 axis could overcome local immunosuppression and enhance the anti-tumor activity of BCG-primed T cells. Early-phase studies and the success of pembrolizumab in BCG-unresponsive NMIBC (KEYNOTE-057) set the stage for large-scale trials in the BCG-naive setting.

Key Content

The ALBAN Trial: Design and Patient Population

The ALBAN trial (GETUG-AFU 37) was an international, randomized, open-label phase III trial conducted across multiple centers. It enrolled 517 BCG-naive patients with high-risk NMIBC. High risk was defined according to European Association of Urology (EAU) guidelines, encompassing patients with high-grade Ta/T1 disease with or without concomitant CIS. Patients were randomized 1:1 to receive either intravesical BCG alone (Arm A) or a combination of intravenous atezolizumab and intravesical BCG (Arm B).

In Arm B, atezolizumab (1200 mg) was administered intravenously every three weeks for up to one year. BCG followed a standard 6-week induction course followed by maintenance instillations for one year. The primary objective was to compare event-free survival (EFS), where events included high-grade recurrence, progression to muscle-invasive or metastatic disease, or death from any cause. Secondary endpoints were meticulously designed to capture high-grade recurrence-free survival, complete response in the CIS cohort, and overall survival.

Efficacy Outcomes: A Neutral Result

The final analysis included 255 patients in the BCG monotherapy arm and 262 in the combination arm. With a median follow-up sufficient to capture initial recurrence patterns, the trial failed to meet its primary endpoint. The investigator-assessed EFS showed no statistically significant difference between the two cohorts. The hazard ratio (HR) was 0.98 (95% CI 0.71-1.36, P = 0.9106), indicating nearly identical survival curves.

Subgroup analyses, stratified by age, gender, T-stage, and the presence of CIS, confirmed the lack of benefit across all pre-specified categories. Specifically, for patients with CIS—a group traditionally thought to be more sensitive to immunotherapy—the addition of atezolizumab did not result in a superior complete response rate or increased duration of response. These results were unexpected given the preclinical synergy reported between intravesical immunotherapy and systemic checkpoint blockade.

Safety and Tolerability Profile

The safety data from ALBAN highlighted the inherent risks of adding systemic therapy to a localized treatment regimen. While intravesical BCG is associated with local irritative symptoms (cystitis, hematuria), the combination arm experienced systemic toxicities characteristic of atezolizumab. Treatment-related adverse events (TRAEs) of grade 3 or higher were significantly more frequent in the combination arm compared to BCG alone. Common TRAEs included fatigue, pruritus, and colitis. Importantly, the rate of discontinuation due to adverse events was higher in Arm B, suggesting that the added toxicity was not compensated for by clinical benefit in this early-stage patient population.

Contextualizing ALBAN within the NMIBC Landscape

The results of ALBAN must be viewed alongside other major trials in the NMIBC space. For instance, the KEYNOTE-676 trial, which evaluated pembrolizumab plus BCG in a similar population, reported more favorable EFS data in certain contexts. The divergence between ALBAN (atezolizumab, an anti-PD-L1) and trials using PD-1 inhibitors (like pembrolizumab) raises critical questions. It is possible that the mechanism of action—specifically the difference between targeting the ligand (PD-L1) versus the receptor (PD-1)—plays a role in the unique environment of the bladder. Furthermore, the duration and timing of the PD-L1 blockade relative to the BCG-induced inflammatory peak may influence the success of the synergy.

Expert Commentary

The failure of the ALBAN trial to improve EFS is a sobering reminder of the complexities of the tumor microenvironment in NMIBC. Several factors may explain why systemic atezolizumab did not enhance the efficacy of BCG. First, the timing of administration is crucial. BCG triggers a massive infiltration of immune cells into the bladder wall; however, the maximal expression of inhibitory checkpoints may not align perfectly with a fixed 3-week atezolizumab schedule. Second, the “BCG-naive” population already has a relatively high response rate to BCG alone, making it statistically difficult to demonstrate a significant incremental benefit without a very large sample size or a highly selected high-risk cohort.

From a biological standpoint, we must consider whether PD-L1 is the primary driver of immune evasion in all high-risk NMIBC patients. The results suggest that NMIBC is a heterogeneous disease, and a “one-size-fits-all” approach to adding immunotherapy is likely inefficient. Experts argue that future trials should utilize biomarker-driven inclusion criteria. For example, quantifying baseline PD-L1 expression, tumor mutational burden (TMB), or the presence of specific interferon-gamma signatures might identify the subset of patients who would truly benefit from the combination. Without such selection, the systemic toxicity of agents like atezolizumab may outweigh their modest oncologic contributions in the NMIBC setting.

Another area of controversy is the route of delivery. While ALBAN utilized systemic IV administration, ongoing research is investigating intravesical delivery of ICIs. Local delivery might achieve higher mucosal concentrations and lower systemic toxicity, potentially providing a more favorable therapeutic index. The ALBAN trial reinforces current EAU and AUA guidelines, which maintain BCG monotherapy as the standard of care for high-risk NMIBC, while signaling that systemic atezolizumab should not be routinely added in the BCG-naive setting outside of further clinical investigation.

Conclusion

The ALBAN (GETUG-AFU 37) trial provides high-level evidence that adding systemic atezolizumab to a standard 1-year BCG regimen does not improve EFS in BCG-naive, high-risk NMIBC patients. Despite a strong mechanistic rationale, the trial yielded a neutral result with an unfavorable increase in systemic toxicity. These findings highlight that the success of immune checkpoint combinations in NMIBC is not a class effect and may depend heavily on the specific agent, the timing of delivery, and patient selection. Future efforts must focus on identifying predictive biomarkers to refine patient selection and exploring alternative delivery methods or combination strategies to improve outcomes for this high-risk population without compromising quality of life.

References

• Roupret M, Bertaut A, Pignot G, et al. ALBAN (GETUG-AFU 37): a phase III, randomized, open-label international trial of intravenous atezolizumab and intravesical Bacillus Calmette-Guérin (BCG) versus BCG alone in BCG-naive high-risk, non-muscle-invasive bladder cancer (NMIBC). Ann Oncol. 2026;37(1):44-52. PMID: 41110692.
• Balar AV, Kamat AM, Kulkarni GS, et al. Pembrolizumab monotherapy for the treatment of high-risk non-muscle-invasive bladder cancer unresponsive to BCG (KEYNOTE-057): an open-label, single-arm, multicentre, phase 2 study. Lancet Oncol. 2021;22(7):919-930. PMID: 34051159.
• Steinberg GD, Roupret M, Iori F, et al. Intravesical Bacillus Calmette-Guérin (BCG) in the treatment of non-muscle invasive bladder cancer. Lancet Oncol. 2020;21(11):e523-e532.