ASCO 2020 Data Update: CheckMate 227

Stephen Liu, MD: CheckMate 227 was a randomized trial that included patients with stage IV or recurrent non–small cell lung cancer and no prior systemic therapy for advanced disease. Patients with known EGFR mutations or ALK fusions were excluded, as were patients with untreated, unstable brain metastases. It did include patients with both squamous and nonsquamous histologies. About 30% of patients in this trial had squamous non–small cell lung cancer. It’s a bit higher than some other studies, like KEYNOTE-024, where it was about 20%. You want to factor that in when you look at the numbers, as some populations do have a slightly worse prognosis.

In the 3-year update, the median survival for the PD-L1–positive group was 17.1 months, compared with 14.9 months with chemotherapy. That’s a hazard ratio of 0.79 for death. The landmark survival rates favored duel-checkpoint blockade. The 2-year OS [overall survival] rate was 40% versus 33% with chemotherapy, and the 3-year survival rate was 33% versus 22%. What I thought was maybe more compelling was the PD-L1–negative cohort, where the hazard ratio was better at 0.64. The difference in hazard ratio for death, 0.64 in PD-L1 negative, 0.79 in PD-L1 positive, was primarily because of the performance of the chemotherapy arm. The nivo-ipi [nivolumab-ipilimumab] arm was remarkably consistent when we compare the PD-L1 positive and negative. In the positive group, 17.1 months was the median survival; in the negative, 17.2. We usually expect a bit of a drop-off in the PD-L1–negative cohort. We didn’t see that with nivolumab and ipilimumab. While chemotherapy was factoring in a bit worse in the PD-L1–negative group, nivolumab and ipilimumab performed consistently well.

Neal E. Ready, MD, PhD: CheckMate 227 was a large, randomized phase 3 trial, and a major component of that trial was looking at the immune combination of nivolumab plus ipilimumab compared with chemotherapy. And the overall survival end point was the primary end point for this trial in PD-L1–positive lung cancer. This trial met its primary end point. It was clearly an improvement in overall survival for the immune combination, compared with chemotherapy that was both clinically and statistically significant.

Stephen Liu, MD: In CheckMate 227, progression-free survival [PFS] was superior with dual-checkpoint blockade. Using nivolumab and ipilimumab compared with histology-specific chemotherapy, hazard ratio for PFS was 0.81 in the PD-L1–positive group, 0.75 in the PD-L1–negative group. The 3-year PFS rate was 18% among positive patients, 13% among negative. Response rate was also higher with just the nivolumab-ipilimumab combination, at 36% in the positive versus 27% in the negative group. What was most impressive, looking at the outcomes in CheckMate 227, was the duration of response. In the PD-L1–positive group, that duration of response was 23.2 months, almost 2 years, for a median duration of response. The PD-L1 negative cohort had an 18-month duration of response. These responses are remarkably durable when we consider the responses we’re achieving with dual-checkpoint blockade compared with chemotherapy-immunotherapy. When we look at chemotherapy-immunotherapy, some of the responses we see are those durable immunotherapy-based responses. But some are just chemotherapy responses, which are somewhat transient. When we’re using only dual checkpoint, the responses we see are immunotherapy-based responses, and those tend to be more durable. A duration of response of almost 2 years is quite compelling, and it’s difficult to ignore that.

Transcript Edited for Clarity