Win for Perioperative Durvalumab in Muscle-Invasive Bladder Cancer

Adding perioperative durvalumab (Imfinzi) to neoadjuvant chemotherapy significantly improved survival outcomes in cisplatin-eligible patients with muscle-invasive bladder cancer (MIBC), a phase III trial showed.

Estimated 24-month event-free survival (EFS) rates reached 67.8% in the durvalumab group and 59.8% in the neoadjuvant chemotherapy-alone group (HR 0.68, 95% CI 0.56-0.82, P

The estimated overall survival (OS) at 24 months was 82.2% and 75.2% in the two groups, respectively (HR 0.75, 95% CI 0.59-0.93, P=0.01).

“NIAGARA is the first phase III perioperative immune therapy study in muscle-invasive bladder cancer,” said Bowles at the European Society for Medical Oncology congress in Barcelona. “It shows a significant event-free survival advantage, it shows a 25% reduction in the risk of death, it’s working across various subgroups of patients, and the pathological complete response results, the significant overall survival benefit, and the shape of the curves underline the importance of both periods of therapy [adjuvant and neoadjuvant].”

Thus, “NIAGARA supports perioperative durvalumab with neoadjuvant chemotherapy as a potential new standard treatment for patients with cisplatin-eligible muscle-invasive bladder cancer,” he concluded.

Results from the NIAGARA trial were published simultaneously in the New England Journal of Medicine.

“I think this is a practice-changing trial,” said invited discussant Petros Grivas, MD, PhD, of the Fred Hutch Cancer Center in Seattle.

Grivas pointed out that while immune checkpoint inhibitors demonstrated improved disease-free survival as adjuvant monotherapy in the CheckMate 274 and AMBASSADOR studies, they have yet to demonstrate a significant OS advantage.

“NIAGARA is the first trial that shows a clear EFS and OS benefit in the neoadjuvant and/or adjuvant setting, and in my opinion that is an important differentiator,” he said.

However, Grivas also cautioned that the trial design could not distinguish the impact of the neoadjuvant and adjuvant therapy phases on the outcome. “Do we need either, or do we need both?” he asked. “In the future, as we go forward, we can think about teasing out the individual contribution of the neoadjuvant and adjuvant phases.”

As background, Powles pointed out that neoadjuvant cisplatin-based chemotherapy with radical cystectomy improves OS compared with radical cystectomy alone, and has been the recommended treatment for MIBC for decades.

“Despite this, about half of patients will relapse and die from urothelial cancer, so it is an area of unmet need,” he said.

Powles said the perioperative approach has a strong biological rationale, and noted that perioperative durvalumab was shown to be safe and effective in a phase II study of MIBC.

In NIAGARA, 533 patients were assigned to receive neoadjuvant durvalumab plus gemcitabine-cisplatin every 3 weeks for four cycles, followed by radical cystectomy and adjuvant durvalumab every 4 weeks for eight cycles (durvalumab group); while 530 were randomized to receive neoadjuvant gemcitabine-cisplatin followed by radical cystectomy alone (comparison group).

The median age was 65 years in both groups, and 82% of patients were male.

A total of 530 and 526 patients assigned to the durvalumab and comparator arms, respectively, started neoadjuvant treatment, with 113 and 137 discontinuing treatment prior to surgery.

Cystectomy was completed in 470 patients in the durvalumab arm and 446 patients in the placebo arm, while 63 and 84 patients, respectively, elected not to undergo surgery. Among the 383 patients who started adjuvant treatment in the durvalumab arm, 95 discontinued treatment.

The study had dual primary endpoints — EFS and pathological complete response (pCR), which was analyzed about 6 months after the last patient had a cystectomy.

In a primary analysis, the pCR rate was 33.8% in the durvalumab group and 25.8% in the comparator group (risk ratio [RR] 1.30, 95% CI 1.09-1.56, P=0.004) — a difference that was not considered significant due to the statistical criteria for the trial.

A re-analysis of pCR included 59 evaluable samples from the intent-to-treat population that were incorrectly attributed as non-responders in the formal analysis, and this re-analysis showed a pCR rate of 37.3% in the durvalumab arm and 27.5% in the comparator arm, showing nominal statistical significance in favor of the durvalumab arm (RR 1.34, 95% CI 1.13-1.60, P=0.0005).

Powles observed that the EFS benefit was consistent across prespecified subgroups and that the percentage of patients who underwent surgery was comparable between the two trial groups, “which means that the addition of neoadjuvant durvalumab to chemotherapy did not lead to a reduction in the percentage of patients who underwent radical cystectomy.”

Treatment-related adverse events (TRAEs) of any grade occurred in 94.7% of the patients in the durvalumab group and in 92.6% of those in the comparison group, with grade 3/4 TRAEs occurring in 40.6% and 40.9%, respectively.

TRAEs leading to death occurred in three patients (0.6%) in each group.

Adverse events leading to the discontinuation of neoadjuvant treatment occurred in 14.9% of the patients in the durvalumab group and in 15.2% of those in the comparison group, with a similar percentage of patients with adverse events precluding or delaying surgery across groups.