Outcomes by line of therapy and programmed death ligand 1 expression in patients with advanced melanoma treated with pembrolizumab or ipilimumab in KEYNOTE-006: A randomised clinical trial

Background: Predictive biomarkers of patients likely to benefit from anti–programmed death 1 inhibitor therapy have clinical relevance. We examined whether line of therapy or tumour programmed death ligand 1 (PD-L1) expression affects the efficacy and safety of pembrolizumab, compared with ipilimumab, in advanced melanoma. Methods: Of 834 patients enrolled in the randomised, open-label phase III KEYNOTE-006 study, 833 were included in this analysis. Patients were randomly assigned 1:1:1 to receive pembrolizumab 10 mg/kg every 2 or 3 weeks (for 24 months) or ipilimumab 3 mg/kg every 3 weeks (for four doses) until disease progression/intolerable toxicity. This analysis evaluated progression-free survival (PFS), overall survival (OS) and objective response rate (ORR). Data cut-off: 03 November 2016. Results: Of the patients, 60.3% were male, 65.9% were treatment naive and 80.6% had PD-L1–positive tumours (median follow-up was 33.9 months). Twenty-four–month survival rates were higher with pembrolizumab than with ipilimumab in treatment-naive (PFS 31.0% versus 14.6%; OS 58.0% versus 44.7%) and previously treated patients (PFS 25.7% versus 11.3%; OS 49.2% versus 37.9%). Twenty-four–month survival rates were higher with pembrolizumab than with ipilimumab in patients with PD-L1–positive tumours (PFS 33.2% versus 13.1%; OS 58.4% versus 45.0%) and similar in PD-L1–negative tumours (PFS 14.9% versus NR [no data at 24 months for a PFS estimate]; OS 43.6% versus 31.8%). Safety of pembrolizumab by subgroup was consistent with previous reports. Conclusions: Findings support pembrolizumab monotherapy as standard of care in patients with advanced melanoma, regardless of first- or second-line therapy or PD-L1 status. ClinicalTrials.gov identifier: NCT01866319.