Treatment with the PD-1 inhibitor pembrolizumab (Keytruda) significantly prolonged overall survival (OS) over investigators’ choice of chemotherapy in patients with recurrent advanced urothelial carcinoma, according to results reported at the 19th European Cancer Congress (ECCO 2017). Findings from the phase III Keynote-045 study support pembrolizumab as a new standard of care for patients with advanced urothelial carcinoma that progressed during or after platinum-base chemotherapy.
Urothelial carcinoma is currently treated in the first-line setting with platinum-based chemotherapy, and there is no standard therapy for recurrent or progressive disease.
At a median follow-up of 14.1 months (range: 9.9-22.1), the median OS with pembrolizumab was 10.3 months (95%CI, 8.0- 11.8) compared with 7.4 months (95%CI, 6.1- 8.3) with standard chemotherapy (HR, 0.723; 95% CI, 0.59-0.91; P
<.0022). One-year OS rates were 43.9% with pembrolizumab versus 30.7% with chemotherapy.
The confirmed objective response rate was 21.1% with pembrolizumab compared to 11.4% with chemotherapy (P = .0011). Complete response (CR) was reported for 7% of patients, and partial response (PR) for 14.1% of patients treated with pembrolizumab, compared to a CR in 3.3% and a PR in 8.1% of patients receiving chemotherapy.
“Pembrolizumab is the first agent to demonstrate OS improvement compared with chemotherapy in patients with advanced urothelial carcinoma after failure of platinum-based chemotherapy,” said Andrea Necchi, Fondazione IRCCS Istituto Nazionale dei Tumori, in Milan, Italy. “This is the first immunotherapy to demonstrate OS benefit over an active comparator in urothelial carcinoma.”
In the KEYNOTE-045 trial, 270 patients with advanced urothelial carcinoma were randomized after the failure of platinum-based doublet chemotherapy to pembrolizumab at 200 mg every 3 weeks for 24 months, and 272 patients to receive the investigator's choice of paclitaxel at 175 mg/m2
, docetaxel at 75 mg/m2
, or vinflunine at 320 mg/m2
, each administered every 3 weeks. Randomization was stratified by ECOG performance status (0/1 versus 2), liver metastases (yes versus no), hemoglobin level (<10 versus ≥10 g/dL), and the time from last chemotherapy (<3 versus ≥3 months). The median patient age was 66 years, and nearly all patients had an ECOG performance status of 0 or 1.
There was no significant difference in progression-free survival (PFS); median PFS was 2.1 months with pembrolizumab compared with 3.3 months with chemotherapy (P
“The objective response rate was significantly higher, and the response was more durable with pembrolizumab,” said Necchi.
For those receiving pembrolizumab, the median time to response was 2.1 months (range: 1.4-6.3) compared with 2.1 months (range: 1.7-4.9) with chemotherapy. The median duration of response (DoR) was not reached (range: 1.6-15.4+) with pembrolizumab, and 4.3 months (range: 1.4-15.4+) with chemotherapy.
A combined positive score was determined in patients with quantifiable PD-L1 expression on tumor and infiltrating immune cells; patients with high PD-L1 expression were described CPS ≥10.
In the CPS ≥10 cohort 74 patients received pembrolizumab and 90 patients received chemotherapy. Median OS was 8.0 months versus 5.2 months, and 1-year OS rates were 39.8% with pembrolizumab versus 26.97% with chemotherapy. However, the ORR with pembrolizumab was 21.6% compared with 6.7% with chemotherapy.
“Pembrolizumab benefit was observed regardless of PD-L1 expression, “Necchi commented.
The KEYNOTE-045 investigators also performed a meta-analysis of OS in studies of second-line immunotherapy for advanced urothelial carcinoma that showed the median OS of 10.30 months demonstrated in this study was greater that the 9.7 months seen with nivolumab in a phase I/II trial1
, the 8.74 months with nivolumab in a phase II trial2
and the 7.9-month OS shown with atezolizumab in a phase II trial.3
The median exposure with pembrolizumab was 3.5 months (range, 0.03-20.0) with pembrolizumab compared to 1.5 months (range, 0.03-14.2) with chemotherapy.
Treatment-related adverse events (TRAEs) were less frequent with pembrolizumab compared with chemotherapy. The rates of all-grade TRAEs were 60.9% and 90.2%, and Grade 3-5 TRAEs occurred in 15% of patients treated with pembrolizumab and 49.9% in patients treated with chemotherapy. TRAEs with pembrolizumab occurring at an incidence of ≥10% were pruritus (20%), fatigue (10%), nausea (10%), and diarrhea (10%).
Treatment discontinuation due to a TRAE was reported for 15 (5.6%) pembrolizumab patients compared to 28 (11%) of chemotherapy patients. Four deaths due to a TRAE occurred in each arm.