Anti-PD-1 Agent Improves Overall Survival in Bladder Cancer

Overall survival (OS) was improved on Keytruda (pembrolizumab) compared to chemotherapy for patients with advanced urothelial carcinoma whose disease progressed after prior treatment, according to results from KEYNOTE-045, a phase 3 trial presented by Joaquim Bellmunt, M.D., Ph.D., at the 2016 SITC Annual Meeting.

In the trial, patients treated with Keytruda achieved a median OS of 10.3 months compared with 7.4 months for those who received a chemotherapy regimen. The difference resulted in a hazard ratio of 0.73. The survival benefit was observed regardless of PD-L1 expression status and there were fewer adverse events (AEs) with the immunotherapy, Bellmunt said.

“Bladder cancer is a disease where nothing has changed in the last 20 years,” said Bellmunt, an associate professor of Medicine at Harvard Medical School and director of the Bladder Cancer Center at Dana-Farber Cancer Institute. “Now, for the first time, we have an agent that in fact improves survival in the second-line setting.”

The KEYNOTE-045 study was designed for patients with locally advanced or metastatic, unresectable urothelial carcinoma of the renal pelvis, ureter, bladder, or urethra who had progressed after one or two lines of platinum-based chemotherapy or who had experienced recurrence after 12 months of chemotherapy.

Overall, 542 patients were randomized to Keytruda (200 mg IV) every three weeks for two years versus chemotherapy consisting of either paclitaxel (175 mg/m2), docetaxel (75 mg/m2) or vinflunine (320 mg/m2) every three weeks for two years. The median age was 67 years in the Keytruda arm and 65 years in the chemotherapy cohort.

The primary endpoints were OS and PFS in the total population and among participants with a combined positive score (CPS) 10 percent or more for PD-L1 expression. The CPS consisted of the percentage of PD-L1—positive tumor cells (TCs) and infiltrating immune cells relative to the total number of TCs as measured using the PD-L1 IHC 22C3 pharmDx assay on samples collected by core needle or excisional biopsies or in resected tissue.

The treatment groups were well-balanced for four key prognostic factors: hemoglobin level; ECOG performance status; liver metastases (yes vs no); and time from last chemotherapy dose (less or more than three months). Bellmunt said these factors had proved significant differentiators of response in prior bladder cancer trials.

Keytruda therapy also resulted in a significantly higher objective response rate (ORR) of 21.1 percent compared with 11.4 percent with chemotherapy. Similarly, the complete response (CR) rate was much higher with Keytruda at 7.0 percent compared with a 3.3 percent CR with chemotherapy. The median duration of response in the Keytruda arm was not reached with an estimated 68 percent of responders considered likely to maintain a response for at least 12 months. By comparison, the median duration of response in the chemotherapy arm was 4.3 months with an estimated 35 percent likely to maintain a response for at least 12 months.

By contrast, progression-free survival (PFS) was not superior with Keytruda by the time of data cutoff on Sept. 7. The median PFS was 2.1 months with the immunotherapy versus 3.3 months with chemotherapy. Bellmunt noted that the PFS curves started to separate in favor of Keytruda at about 12 months, indicating that the immunotherapy benefit may be greater as time goes on. The results he presented at SITC 2016 were based on a median follow-up of 14.1 months.

In the OS analysis of patients with CPS 10 percent or greater, there was a 43 percent reduction in the risk of death with Keytruda compared with chemotherapy. The median OS was 8.0 months with Keytruda versus 5.2 months with chemotherapy.

Although the advantage was maintained in the PD-L1—high population, Bellmunt said it was surprising that the benefit was not more pronounced for these patients. This development might have occurred because primary resected tumors were used and patients subsequently underwent first-line therapy, which may have altered the PD-L1 expression, he said. Clinical trials are now using samples collected more recently to evaluate PD-L1 levels.

In terms of AEs, patients who received Keytruda had fewer toxicities than those treated with chemotherapy. The incidence of treatment-related AEs was lower with Keytruda compared with chemotherapy, respectively, for any grade (60.9 percent vs 90.2 percent) and for AEs of grade 3-5 severity (15.0 percent vs 49.4 percent).

Treatment-related AEs occurring in 10 percent or more of participants were generally lower with Keytruda as opposed to chemotherapy, respectively, including for fatigue (13.9 percent vs 27.8 percent), nausea (10.9 percent vs 24.3 percent), diarrhea (9.0 percent vs 12.9 percent), asthenia (5.6 percent vs 14.1 percent) and anemia (3.4 percent vs 24.7 percent with chemotherapy).

The incidence of pruritus was higher in the Keytruda arm at 19.5 percent versus the chemotherapy group at 2.7 percent. Immune-related AEs that were higher with Keytruda compared with chemotherapy, respectively, included thyroid abnormalities (9.4 percent vs 1.6 percent), pneumonitis (4.1 percent vs 0.4 percent) and colitis (2.3 percent vs 0.4 percent). Fifteen patients in the Keytruda arm and 28 patients in the chemotherapy group discontinued treatment due to a treatment-related AE. Each arm had four treatment-related deaths, according to Merck, which is developing the drug.