Immunotherapy with anti–PD-1 (programmed cell death protein 1) and anti–PD-L1 (programmed cell death ligand 1) agents continues to advance in metastatic urothelial cancer, with positive showings in two clinical trials presented at the 2016 European Society for Medical Oncology (ESMO) Congress. The first study, KEYNOTE-052, found that pembrolizumab (Keytruda) achieved a 24% response rate as first-line therapy for cisplatin-ineligible patients with advanced urothelial cancer.1 The second study, CheckMate 275, showed encouraging response rates with nivolumab (Opdivo) in patients with previously treated advanced urothelial cancer that failed to respond to platinum-based therapy.2
These two phase II studies suggest that pembrolizumab and nivolumab may join atezolizumab (Tecentriq) as U.S. Food and Drug Administration–approved agents for locally advanced or metastatic urothelial cancer that has progressed on prior therapy. [In fact, based on the results of CheckMate 275, the FDA has granted Breakthrough Therapy designation to nivolumab, permitting expedited review for a potential indication in second-line metastatic urothelial cancer.] Nivolumab and pembrolizumab are anti–PD-1 agents, whereas atezolizumab is an anti–PD-L1 agent.
Patients with advanced urothelial cancer have few good treatment options. Cisplatin-based therapy is the standard of care for first-line treatment of advanced urothelial cancer, but many patients are not cisplatin-eligible due to impaired renal function, poor performance status, or other comorbidities, and other patients treated with cisplatin-based therapy have a suboptimal response or continue to have disease progression on therapy. Other approaches are needed.
Many of the patients enrolled in this trial would otherwise not be expected to enroll in trials involving chemotherapy. Therapy was well tolerated, and responses occurred quickly, but with a median follow-up of 8 months, we need more time to assess the durability of response. These data are promising but will need to be replicated in randomized trials.— Arjun Balar, MD
Arjun Balar, MD, of NYU Langone Medical Center, New York, presented an interim analysis of response rates in the first 100 patients with cisplatin-ineligible advanced urothelial cancer enrolled in the phase II, open-label KEYNOTE-052 study. Patients received first-line pembrolizumab given intravenously in a flat dose of 200 mg every 3 weeks until disease progression or unacceptable toxicity.
These patients were typical of a cisplatin-ineligible population. Median age was 75 years, with 34 patients older than age 80. Eighty-seven percent had visceral metastasis, 43% enrolled due to a poor Eastern Cooperative Oncology Group [ECOG] performance status, and 45% had impaired kidney function (11% had both ECOG 2 performance status and renal insufficiency).
In these very sick patients, overall response rate was 24%, and complete response rate was 6%; an additional 15% had stable disease on pembrolizumab. In patients with lymph node–only metastasis, the overall response rate was 40% vs 21% in patients with visceral metastasis.
At a median follow-up of 8 months, median duration of response was not yet reached; the range of response duration was from 1.4 to 9.8 months. The duration of response was 6 months or longer in 83% of patients.
“Many of the patients enrolled in this trial would otherwise not be expected to enroll in trials involving chemotherapy. Therapy was well tolerated, and responses occurred quickly, but with a median follow-up of 8 months, we need more time to assess the durability of response. These data are promising but will need to be replicated in randomized trials,” Dr. Balar told listeners.
“We need to define a biomarker for response,” he continued. He suggested that the combined positive score of PD-L1 expression in both tumor and infiltrating immune cells could be a useful biomarker to predict who is most likely to respond. However, the greater challenge that remains unanswered is identifying who should not receive treatment,” he said. The study used a tentative cutoff of 10% or higher for PD-L1 expression in tumor and immune cells. “The PD-L1 cutpoint needs to be validated in the remaining 274 patients,” he added.
Responses were seen across all PD-L1 subgroups, but the best responses were observed in the combined positive score ≥ 10% group (high tumor and immune cell PD-L1 expression), who had an overall response rate of 37% and a complete response rate of 13%.
Regarding safety, 16% of patients had grade 3 or 4 adverse events. Rates of immune-mediated events were low, and only 5% of all patients discontinued treatment due to toxicity.
The phase II CheckMate 275 study enrolled 270 patients with platinum-resistant advanced urothelial cancer treated with nivolumab at 3 mg/kg intravenously every 2 weeks until disease progression or unacceptable toxicity. Outcomes were analyzed in all treated patients and by tumor PD-L1 expression at cutoffs of ≥ 1% and ≥ 5% (using the Dako PD-L1 PharmDx test).
In this large study of advanced/unresectable urothelial cancer, nivolumab had clinically meaningful efficacy and a manageable safety profile. The efficacy benefit of nivolumab was seen across all PD-L1 subgroups.— Matthew D. Galsky, MD
“In this large study of advanced/unresectable urothelial cancer, nivolumab had clinically meaningful efficacy and a manageable safety profile. The efficacy benefit of nivolumab was seen across all PD-L1 subgroups,” said Matthew D. Galsky, MD, of the Tisch Cancer Institute, Icahn School of Medicine at Mount Sinai, New York.
Median age was 66 years, about 30% had been treated with 2 or more prior regimens for advanced disease, and about 30% received nivolumab as first treatment in the metastatic setting. Forty-six percent of patients had tumors with ≥ 1% PD-L1 expression, and 31% had ≥ 5% PD-L1 expression.
The objective response rate was 19.6%; 2.3% had a complete response, and an additional 23% had stable disease. Higher levels of PD-L1 expression were associated with higher responses, but patients with very low or no PD-L1 expression also responded: The objective response rate was 16.1% for patients with PD-L1 expression < 1% and 28.4% for those with the highest level of PD-L1 expression, ≥ 5%.
Duration of response had not yet been reached in 52 responders followed for at least 6 months; 40 of 52 patients had ongoing responses.
Median progression-free survival was 2.0 months. For those with PD-L1 < 1%, median progression-free survival was 1.87 months, whereas for those with PD-L1 expression ≥ 1%, it was 3.55 months.
Median overall survival for the entire study population was 8.74 months; for those with PD-L1 expression < 1%, 5.95 months; and for those with PD-L1 expression ≥ 1%, 11.3 months.
The most common adverse events of any grade were fatigue (16.7%), pruritus (9.3%), diarrhea (9.3%), decreased appetite (8.1%), hypothyroidism (7.8%), and nausea (7%). The most common grade 3 or 4 adverse events were fatigue and diarrhea (1.9% for each).
Dr. Galsky noted that quality of life remained stable over time. ■
Disclosure: Dr. Balar has been a consultant for Genentech, Merck, AstraZeneca/MedImmune, and Cerulean Pharma and has received research funding from Genentech, Merck, and AstraZeneca/MedImmune. Dr. Galsky owns stock in Dual Therapeutics; has been a consultant for Genentech, Merck, Novartis, and Astellas; has received research funding from Bristol-Myers Squibb, Dendreon, Janssen, Novartis, and Merck; and has intellectual properties in association with Mount Sinai School of Medicine.
1. Balar A, Bellmunt J, O’Donnell PH, et al: Pembrolizumab as first-line therapy for advanced/unresectable or metastatic urothelial cancer: Preliminary results from the phase 2 KEYNOTE-052 study. 2016 ESMO Congress. Abstract LBA32_PR. Presented October 8, 2016.
2. Galsky MD, Tetz M, Siefker-Radtke AO, et al: Efficacy and safety of nivolumab monotherapy in patients with metastatic urothelial cancer who have received prior treatments: Results from the phase II CheckMate 275 study. 2016 ESMO Congress. Abstract LBA31_PR. Presented October 8, 2016.