Ipilimumab/Chemo Combo Misses OS Endpoint in Phase II Urothelial Cancer Trial

The addition of the CTLA-4 inhibitor ipilimumab to cisplatin and gemcitabine failed to significantly improve the primary endpoint of overall survival (OS) in a phase II study for patients with metastatic urothelial cancer.

Matthew Galsky, MD

The addition of the CTLA-4 inhibitor ipilimumab (Yervoy) to cisplatin and gemcitabine failed to significantly improve the primary endpoint of overall survival (OS) in a phase II study for patients with metastatic urothelial cancer, according to data presented at the 2016 GU Cancers Symposium.

In the single-arm study, the 1-year OS rate was 59%, which did not meet the primary endpoint of 80%. The median OS was 14.6 months (95% CI, 10.5-18.6) and the objective response rate (ORR) was 64%. These findings were consistent with historical data for cisplatin and gemcitabine alone.

The study remains an important milestone, as the first trial initiated to explore an immune checkpoint inhibitor in metastatic urothelial cancer. Additionally, the findings represent the first reported data for the combination of a checkpoint inhibitor and chemotherapy for patients with bladder cancer.

“We set a high bar for the improvement in 1-year overall survival of 20% compared with historical controls. This was based on the toxicity profile of gemcitabine, cisplatin, and ipilimumab independently,” said lead investigator Matthew Galsky, MD, director of Genitourinary Medical Oncology at the Tisch Cancer Institute at Mount Sinai. “We felt that we would really need to see something robust to warrant moving this regimen forward.”

In the phase II study, 36 patients with metastatic urothelial carcinoma received induction therapy for 2 cycles with gemcitabine and cisplatin followed by the doublet plus ipilimumab for 4 cycles. Across both phases of the study, gemcitabine was administered at 1000 mg/m2on days 1 and 8 and cisplatin was given at 70 mg/m2on day 1 of each 21-day cycle. Ipilimumab was administered at 10 mg/kg on day 1 of each cycle. Following treatment with the triplet, patients with stable disease or better went on to receive maintenance ipilimumab at 10 mg/kg every 3 months.

The primary endpoint of the study was the 1-year OS rate. To be significant, the lower bound of the 90% confidence interval needed to exceed 60%, which is equivalent to a 1-year OS rate of 80%. The secondary endpoints of the study focused on ORR, progression-free survival, and safety.

The median age of patients was 60 years and the majority were males (81%). The primary tumor site was the bladder (78%), with 20% having renal pelvic disease. Nearly a third of the patients (30%) had a Karnofsky performance status of 80% and 58% had visceral metastasis. Prior treatments included systemic chemotherapy (15%) and surgical removal of the primary tumor (53%).

Across the full study, the complete response (CR) rate was 14%. Overall, 17% of patients experienced an improvement in response with the addition of ipilimumab following induction therapy. There was 1 CR following induction therapy and 4 CRs with the addition of ipilimumab. “It’s unclear, of course, if this is related to ipilimumab or to additional chemotherapy,” noted Galsky.

The partial response rate was 50%, across both treatment phases. These responses were primarily achieved during induction therapy (53%), some of which converted to CRs. The stable disease rate was 31%. After the predefined 6 cycles of treatment, 15 patients had stable disease or better and were eligible for maintenance therapy.

Patients received a median of 5 cycles of gemcitabine plus cisplatin and 3 cycles of ipilimumab. Disease progression was the most common cause of treatment discontinuation.

At least 1 grade 3/4 adverse event (AE) was experienced by 72% of patients. Potentially immune-related grade 3/4 AEs included diarrhea (8%), colitis (6%), rash (6%), adrenal insufficiency (3%), hypopituitarism (3%), and hyperthyroidism (3%). The most commonly occurring grade 3/4 AEs in the trial were neutropenia (36%), anemia (28%), thrombocytopenia (19%), and diarrhea (9%).

“Based on the character, severity, and frequency of the adverse events, the side effect profile was consistent with additive toxicity rather than synergistic toxicity,” said Galsky.

In an exploratory analysis of immune cells, a significant depletion was not seen in circulating regulatory T cells or myeloid-derived suppressor cells following induction chemotherapy. Additionally, induction therapy did not appear to impact CD4 and CD8 levels, based on peripheral blood measurements. After the addition of ipilimumab, there was an expected expansion in CD4 and CD8 cells, noted Galsky.

“We did see pharmacodynamic effects consistent with the mechanism of ipilimumab, despite administration of concurrent cytotoxic chemotherapy,” he said. “This does potentially bode well for upcoming trials combining cytotoxic chemotherapy with PD-1 and PD-L1 blockade.”

There are currently several trials assessing immune checkpoint inhibitors in combination or as monotherapy for patients with urothelial cancer. The PD-L1 inhibitors atezolizumab and avelumab are currently being assessed primarily as monotherapy for patients with urothelial cancer. The leading agent being assessed in various combinations is pembrolizumab.


  1. Galsky MD, Hahn NM, Albany C, et al. Phase II trial of gemcitabine + cisplatin + ipilimumab in patients with metastatic urothelial cancer. J Clin Oncol. 2016; 34 (suppl 2S; abstr 357).