Pembrolizumab Is Safe and Effective in Cisplatin-Ineligible Advanced Urothelial Carcinoma

October 5, 2017

As first-line therapy in cisplatin-ineligible advanced urothelial cancer, pembrolizumab (Keytruda) was safe and provided tumor reduction as well as durable responses.

Thomas Powles, MD

As first-line therapy in cisplatin-ineligible advanced urothelial cancer (UC), Pembrolizumab (Keytruda) was safe and provided tumor reduction as well as durable responses, according to an update on the phase II KEYNOTE-052 trial presented at the 2017 Global Congress on Bladder Cancer (GCBC), held in Edinburgh, United Kingdom.

Patients treated with pembrolizumab demonstrated an objective response rate (ORR) of 29% (95% CI, 25%-34%), including 27 (7%) patients with complete response (CR) and 81 (22%) patients with partial responses (PR). Stable disease was achieved by 67 (18%) patients as best response, providing a disease control rate of 47%.

“Compared with the earlier report of initial results (September, 2016) from KEYNOTE-052, the ORR increased by 5%, with 10 additional patients achieving CR and 9 additional patients demonstrating PR,” said Thomas Powles MD, Centre for Experimental Cancer Medicine, Barts Cancer Institute, Queen Mary University of London, England.

The median time to response was 2 months (range, 1—9 months). The median duration of response was not reached (95% CI, 12 months to not reached) at a median follow-up of 10 months (range, 0.1–23 months). At this point, 67% of responses were ongoing and 82% of responses had lasted 6 months or longer.

“Advanced urothelial cancer most often afflicts patients who have comorbidities and poor performance status. First-line cisplatin-based chemotherapy improves survival, but age-related complications such as renal dysfunction or poor ECOG performance status (PS) preclude approximately 50% of patients from receiving standard first-line treatment,” said Powles. “Alternative first-line therapies have an inferior outcome and substantial toxicity, leaving best supportive care the most reasonable option for these patients.”

Powles noted that the IMvigor 210 study recently demonstrated an ORR of 23% with another immunotherapy, the anti-PD-L1 antibody atezolizumab, in patients with previously untreated metastatic urothelial cancer.

The phase II KEYNOTE-052 (NCT02335424) trial is evaluating the efficacy and safety of PD-1 blockade with pembrolizumab as first-line therapy in cisplatin-ineligible patients with metastatic or locally advanced bladder cancer. The trial included patients with ECOG PS 2, creatinine clearance ≥30 to <60 mL/min, grade ≥2 neuropathy/hearing loss, and New York Heart Association class III heart failure who had not received prior systemic chemotherapy.

All patients were treated with pembrolizumab at 200 mg IV every 3 weeks. Imaging was performed at week 9, and then every 6 weeks for the first year, and every 12 weeks thereafter.

Efficacy and safety were assessed in the 370 patients receiving 1 or more doses of pembrolizumab. The patients’ median age was 74 (range, 34—94 years), and 77% were males. Nearly a third (29%) of patients were aged 80 years or older and 50% of patients had renal dysfunction. Other reasons for cisplatin ineligibility included ECOG PS 2 for 32% of patients, and both renal dysfunction and ECOG PS 2 in 9% of patients, with other reasons given for another 9% of patients. Twenty-one percent of patients had metastases to the liver.

The primary endpoint was confirmed ORR according to RECIST v1.1 by independent review.

“Analysis of the ORR by patient subgroup revealed similar ORRs for patients aged less than 85 and those 85 years and older, and also between patients with ECOG PS 0/1 and 2,” Powles commented. “In contrast, the presence of liver metastasis was associated with a decrease in ORR.”

The ORR was 18% versus 33% in patients with and without liver metastasis, respectively.

More than half (58%) of patients treated with pembrolizumab demonstrated tumor reduction of 20% or greater.

The safety analysis revealed that pembrolizumab was well tolerated across all subgroups of cisplatin-ineligible patients, including those who were elderly or had poor performance status. Drug-related adverse events (AEs) of any grade occurred in 243 (66%) patients, and 70 (19%) patients experienced grade ≥3 AEs. The most commonly reported treatment-related AEs were fatigue (18%), pruritis (17%), and rash (12%).

The incidence of immune-mediated AEs was low and these occurred in 84 (23%) patients. Mild grade 1/2 hypothyroidism, reported in 11% of patients, was the most common immune-mediated AE.

Discontinuation due to an AE occurred in 7% of patients.

One treatment-related death occurred in a woman aged 83 years.

Investigators also evaluated the association between the ORR and the results of each patient’s 18-gene T-cell-inflamed gene expression profile (GEP) and with a PD-L1 combined positive score (CPS) that was determined by immunohistochemistry (IHC). The PD-L1 cutpoint was 10% or greater total PD-L1 expression in immune cells or tumor cells, and was used to identify patients most likely to respond to pembrolizumab. Of the 30 patients with positive PD-L1 expression, 11 (37%) responded to treatment.

The IHC training set comprised 96 patients; of those with a CPS <10, the ORR was 17% (95% CI, 9% 28%), whereas the ORR was 37% (95% CI, 20%-56%) in patients having a CPS ≥10. In comparison with the IHC validation set, which included 265 patients, the ORR was 23% (95% CI, 17% 29%) for patients with a CPS <10 and the ORR was 51% (95% CI, 40% - 63%) in patients with CPS ≥10.

“An appreciable number of additional responders was captured using the T cell-inflamed GEP as compared with the PD-L1 immunohistochemistry biomarker,” Powles observed.

Evidence supporting a positive association of PD-L1 expression with response was also demonstrated in the 275 patients. “After longer follow-up, results confirm that first-line pembrolizumab elicits clinically meaningful, durable anti-tumor activity in cisplatin-ineligible advanced UC,” Powles summarized.