Adding an immune checkpoint inhibitor to standard Bacillus Calmette-Guérin (BCG) therapy for non–muscle invasive bladder cancer (NMIBC) might reduce recurrences in a small fraction of patients, but far more may have serious side effects, according to findings presented at the 2025 European Society for Medical Oncology (ESMO) annual meeting.

In two phase 3 trials, researchers tested the approach of adding an immune checkpoint inhibitor to standard care for patients with high-risk NMIBC. Only one trial found a modest reduction in recurrences, while both showed significantly increased toxicity.

And right now, there’s no reliable way to identify those patients who stand to benefit, according to Bradley McGregor, MD, a genitourinary medical oncologist at the Dana-Farber Cancer Institute in Boston. 

McGregor was the discussant on the two trials: POTOMAC, which paired BCG with durvalumab (Imfinzi), and ALBAN, which added atezolizumab (Tecentriq). He also highlighted findings from a third, recently published trial, CREST, which paired BCG with the investigational PD-1 blocker sasanlimab.

The combined take-home from the three trials, McGregor said, is “that adding immune checkpoint blockade to BCG is not a slam dunk.”

Efficacy and Toxicity

All three trials enrolled BCG-naive patients with high-risk NMIBC who had undergone standard treatment with transurethral resection of bladder tumor (TURBT) followed by BCG induction and maintenance. Each trial had a primary endpoint of disease-free or event-free survival.

The treatment arms included BCG induction and maintenance for up to approximately 2 years in POTOMAC and CREST, and 1 year in ALBAN. A second arm in the trials added the respective immune checkpoint inhibitors to TURBT and BCG induction and maintenance for up to approximately 1 year with durvalumab or atezolizumab, and 2 years with sasanlimab.

The two new trials, POTOMAC and ALBAN, reached different conclusions regarding the benefits of the add-on. In POTOMAC, patients in the durvalumab arm had a 2-year disease-free survival of 86.5% vs 81.6% in the stand-alone BCG arm — an absolute difference of 4.9%.

In contrast, ALBAN showed no benefit for event-free survival with atezolizumab, though McGregor noted that fewer patients in this trial had pure carcinoma in situ, the subgroup most prone to recurrence.

The previously published CREST was in line with POTOMAC. It showed a 3-year event-free survival of 82.1% with BCG plus sasanlimab vs 74.8% with BCG alone, for an absolute difference of 7.3%.

However, the modest benefits in the two trials came at the cost of greater toxicity, including potentially life-threatening immune-related events.

Overall, grade 3 or higher treatment-related adverse events occurred in 29.1% of sasanlimab patients in CREST vs 6.3% on BCG alone, and in 21% of patients in the durvalumab arm of POTOMAC vs 4% with BCG. The safety profile was similar in ALBAN.

Bladder Preservation Uncertain

A further concern, according to McGregor, is that none of the trials reported cystectomy-free survival as a mature or primary endpoint. So it remains uncertain whether adding immunotherapy helps patients keep their bladders — which is what they want most of all, he said. 

ALBAN lead investigator Morgan Roupret, MD, PhD, a urologic oncologist at Sorbonne University in Paris, made a similar point. Without information on cystectomy-free survival, he said, these trials are mostly about avoiding additional TURBTs for patients with recurrence, not necessarily avoiding bladder removal. 

“It’s very important to see what is at stake here,” Roupret said. “Not life. Cystectomy maybe, but just for the moment we are talking about more or less TURBTs. That's it.”

Updated Perspective on BCG

One message that came through in the trials is that the often-quoted figure that 40% of patients fail to respond to BCG within 2 years — based on the SWOG 8507 trial that established BCG 25 years ago — no longer holds true.

In the BCG induction and maintenance control arms of the new trials, 75% or more of patients were disease-free at 2 years, implying a much lower BCG failure rate. 

With modern TURBT and adequate BCG therapy, McGregor said, “our patients are doing much, much better now.”

He pointed another clear message: Immunotherapy is no substitute for BCG maintenance. All three trials also included an arm combining immunotherapy with BCG induction alone, and all three found that outcomes were worse when patients do not receive BCG maintenance, regardless of whether a checkpoint inhibitor is added. 

If add-on immunotherapy is to be useful in this setting, clinicians will need ways to refine patient selection, according to McGregor. “Ultimately,” he said, “we need to find biomarkers so we can find those 1 in 20 patients that would truly benefit.”

The trials were funded by the companies that make the immune checkpoint inhibitors: Pfizer for sasanlimab in CREST, Roche for atezolizumab in ALBAN, and AstraZeneca for durvalumab in POTOMAC. McGregor reported financial ties to Pfizer and AstraZeneca. Roupret reported financial ties to all three companies. 

M. Alexander Otto is a physician assistant with a master’s degree in medical science and a journalism degree from Newhouse. He is an award-winning medical journalist who worked for several major news outlets before joining Medscape. Alex is also an MIT Knight Science Journalism fellow. Email: aotto@mdedge.com