Safety of pemetrexed plus platinum in combination with pembrolizumab for metastatic nonsquamous non-small cell lung cancer: A post hoc analysis of KEYNOTE-189

Abstract

ObjectivesThis post hoc analysis assessed the safety of pemetrexed and platinum in combination with pembrolizumab, including time-to-onset and time-to-resolution of all-cause any-grade and grade ≥3 adverse events (AEs) and renal AEs. Materials and MethodsPatient-level data from KEYNOTE-189 were analyzed in the all-subjects-as-treated population (pembrolizumab arm, n = 405; placebo arm, n = 202), and among patients who received ≥5 cycles of pemetrexed (pemetrexed/pembrolizumab/platinum arm, n = 310; pemetrexed/placebo/platinum arm, n = 135). All-cause AEs were selected based on ≥2 % incidence from previously reported KEYNOTE-189 data and included neutropenia, febrile neutropenia, anemia, thrombocytopenia, asthenia, fatigue, dyspnea, diarrhea, nausea, vomiting, pneumonitis, and renal events. Descriptive statistics summarized all-cause AEs. Medians and interquartile ranges were used to examine time-to-onset and time-to-resolution. The data cutoff was November 8, 2017. ResultsIn both treatment arms, most non-hematologic (nausea, vomiting, diarrhea, and asthenia), and hematologic (febrile neutropenia, thrombocytopenia, and neutropenia) grade ≥3 AEs with ≥2 % incidence had a median time-to-onset within the first 4 cycles, and a median time-to-resolution of within 2 weeks from onset. A small number of AEs had longer median time-to-onset (pneumonitis and fatigue) and median time-to-resolution (pneumonitis, fatigue, acute kidney injury, and anemia). Among patients who received ≥5 cycles of pemetrexed, the incidence of any-grade renal toxicity in the pemetrexed/pembrolizumab/platinum arm was 2.3 % in Cycles 1−4, 4.8 % in Cycles 5−8, 2.6 % in Cycles 9−12, and 2.5 % in Cycles ≥13; and, in the pemetrexed/placebo/platinum arm, 0.7 % in Cycles 1−4, 1.5 % in Cycles 5−8, 1.3 % in Cycles 9−12, and 2.0 % in Cycles ≥13. ConclusionPemetrexed/pembrolizumab/platinum has manageable toxicity with longer duration of treatment. While the incidence of renal toxicity was slightly higher in the pembrolizumab combination as compared to pemetrexed, the incidence did not increase in later treatment cycles. These results support the safe use of the KEYNOTE-189 regimen in clinical practice. Clinical Trial Registration NumberNCT02578680 (clinicaltrials.gov).