First-Line Nivolumab Plus Ipilimumab Does Not Boost OS in Cisplatin-Ineligible Unresectable Urothelial Carcinoma

Urothelial Carcinoma |
Image Credit: © Matthieu
- stock.adobe.com

Frontline treatment with the combination of nivlumab (Opdivo) and ipilimumab (Yervoy) did not lead to a statistically significant improvement in overall survival (OS) compared with gemcitabine plus carboplatin in patients with unresectable or metastatic urothelial carcinoma who were ineligible for cisplatin, according to data from the phase 3 CheckMate 901 trial (NCT03036098).

However, findings presented at the 2025 ASCO Annual Meeting showed that the treatment combination did produce a notable duration of response and “favorable landmark OS after 12 months,” according to Michiel Van der Heijden, MD, PhD, a medical oncologist at the Netherlands Cancer Institute in Amsterdam, who presented the results.

“Unfortunately, the prespecified thresholds for statistical significance were not met for the dual primary end points of OS in cisplatin-ineligible and PD-L1–positive patients. For cisplatin-ineligible patients, the hazard ratio was 0.79 with a P-value of .0245, where the P-value boundary was .0173. For PD-L1–positive patients, the hazard ratio was 0.87,” Van der Heijden said. “Despite these negative topline results, some interesting observations could be made. Durable response with [nivolumab plus ipilimumab] and favorable landmark OS after 12 months show meaningful activity from this chemotherapy-free regimen.”

CheckMate 901 was a phase 3, open-label, randomized trial in patients with untreated, unresectable or metastatic urothelial carcinoma. Patients who were cisplatin ineligible were randomly assigned 1:1 to nivolumab plus ipilimumab or gemcitabine-carboplatin.

Patients were stratified by tumor PD-L1 expression (≥1% or <1%) and liver metastasis. Patients had to be at least 18 years of age and have an ECOG performance status of 0-1. Median OS follow-up was 69.2 months (range, 58.8-83.2 months) for the nivolumab plus ipilimumab arm and 69.2 months (range, 58.3-86.3 months) for the gemcitabine-carboplatin arm.

In the patients who were cisplatin ineligible, the primary end point was OS.

“OS in all randomized patients was a key secondary end point, but the analysis was contingent on the cisplatin-ineligible population being positive, which was not the case,” Van der Heijden noted. Another key secondary end point was progression-free survival per blinded independent central review (BICR), and key exploratory end points included ORR per BICR, duration of response per BICR, and safety.

Regarding baseline characteristics, “The baseline characteristics for cisplatin-ineligible patients were well balanced between the 2 arms and are similar to other study populations in this disease setting,” Van der Heijden said.

Van der Heijden reported that median time to treatment discontinuation was 2.2 months (95% CI, 2.1-3.5) in the nivolumab plus ipilimumab arm and 3.8 months (95% CI, 3.5-3.9) in the gemcitabine arm.

“Approximately 45% of patients completed the first part of their study treatments,” Van der Heijden noted.

Regarding OS, “CheckMate 901 did not meet its primary end point of OS in the cisplatin-ineligible population, with a hazard ratio of 0.79 and a P-value of .0245. The median OS for [nivolumab plus ipilimumab] was 19.1 months, and for gemcitabine-carboplatin, it was 13.2 months,” Van der Heijden said. The 36-month OS rate was 29.6% for the nivolumab plus ipilimumab arm and 19.3% for the gemcitabine-carboplatin arm, and the 60-month OS rate was 23.0% for the nivolumab plus ipilimumab arm and 14.4% for the gemcitabine-carboplatin arm.

Median PFS “was similar for both arms; however, the curve separated after 8 to 9 months and then remained separated, showing durable progression-free survival for [nivolumab plus ipilimumab],” Van der Heijden said.

Median duration of response was 25.0 months (95% CI, 14.8-61.8) in the nivolumab plus ipilimumab arm vs 7.4 months (95% CI, 5.8-8.5) in the gemcitabine-carboplatin arm. Thirty-four percent of patients in the nivolumab plus ipilimumab arm received subsequent systemic therapy compared with 46% in the gemcitabine-carboplatin arm.

In terms of safety, “As expected, immune-related adverse [effects] were more commonly observed in [nivolumab plus ipilimumab] arm, whereas hematological toxicity was more common in the [gemcitabine-carboplatin] arm. Out of 8 treatment-related deaths, 7 occurred in the [nivolumab plus ipilimumab] arm and 1 in the [gemcitabine-carboplatin] arm,” Van der Heijden said.

“In summary, the prespecified thresholds for statistical significance were not met for the dual primary end points of overall survival in cisplatin-ineligible and PD-L1 positive patients…Durable response with [nivolumab plus ipilimumab] and favorable landmark OS after 12 months show meaningful activity from this chemotherapy-free regimen. The safety profile of [nivolumab plus ipilimumab] is consistent with that observed in other studies," Van der Heijden said.

Reference

Van der Heijden M, Galsky M, Powles T, et al. Nivolumab plus ipilimumab (NIVO+IPI) vs gemcitabine-carboplatin (gem-carbo) chemotherapy for previously untreated unresectable or metastatic urothelial carcinoma (mUC): Final results for cisplatin-ineligible patients from the CheckMate 901 trial. J Clin Oncol. 2025;43(suppl 17):4500. doi:10.1200/JCO.2025.43.16_suppl.4500