EXPERT PERSPECTIVE VIRTUAL TUMOR BOARDPaul K. Paik, MD:In summary, this is a new diagnosis of stage IV lung adenocarcinoma in a patient who is PD-L1 [programmed-cell death ligand 1]negative. There are a couple of trials that have read out over the past few months that are relevant, in terms of how we should think about the management, because that has shifted. The first is KEYNOTE-189, and the second is CheckMate 227. So, we’ll talk a little about that data.
The first trial, KEYNOTE-189, was a follow up to KEYNOTE-021 cohort G, which is a subarm of a phase II trial taking a look at carboplatin/pemetrexed with or without pembrolizumab in the first-line setting. KEYNOTE-189 was a randomized phase III trial of dealer’s choice platinum with pemetrexed with or without pembrolizumab. Pemetrexed, in this setting, could be continued as a maintenance therapy with pembrolizumab as an option for the clinician and the patient.
At the end of the day, the trial was positive. It was positive in the intention-to-treat analysis, in terms of overall survival. The subgroup analysis showed that there was benefit in pretty much every single subgroup, in terms of PFS [progression-free survival], overall survival, and overall response rate. What was most powerful was that in each of the PD-L1 cut points that we care about, that are clinically relevantthe high expressers, with a PD-L1 greater than 50%; the intermediate category, with a PD-L1 between 1% to 49%; and the negative category, with a PD-L1 of 0%—we had a significant overall survival advantage. So, patients were living longer.
The hazard ratios were pretty much comparable. They were all around a hazard ratio of 0.5 for overall survival. This was also carried over to a PFS benefit, in each of the subgroups, as well as overall response benefit. So, all 3 clinical metrics were better.
The toxicity was not that bad. There was not much of a signal, in terms of any increased toxicity outside of what you would expect when you give patients pembrolizumab, which is a relatively low rate of autoimmune adverse events that we are sort of used to. But, this was not increased over what we’d expect from single-agent pembrolizumab.
The next trial that had read out was CheckMate 227. CheckMate 227 was a randomized phase III trial looking at the frontline setting of nonsmall cell lung cancer patients. The trial looked at a variety of different nivolumab-based regimens. This was ipilimumab/nivolumab versus chemotherapy/nivolumab, versus standard chemotherapy in non–small cell lung cancer patients. The trial was designed to take a look at stratification by PD-L1. Based on some interim analyses, where tumor mutation burden (TMB) seemed to be a predictive biomarker, the trial was amended to take a look at TMB as a potential biomarker stratified by a high and low status, with the cutoff being 10 mutations per megabase.
That is what was carried forward in the presentations. The trial was also positive. Of the entire population with TMB who were stratified by PD-L1 status, the benefit in terms of progression-free survival was found irrespective of PD-L1 expression. This was in the low as well as the high category. It’s important to note that in terms of the data that was presented, ipilimumab and nivolumab did seem to be the best regimen, particularly in the PD-L1negative population. This was followed by chemotherapy plus nivolumab, and then, finally, by chemotherapy, in terms of the PFS benefit.
Overall, we were left with a real predictive biomarker for a nivolumab-based regimen. It’s a bit difficult to figure out what the niche is, in part because overall survival data has not yet been presented for the CheckMate 227 regimen. I think that hampers our ability to compare this regimen to other first-line regimens that have come up. This would be, I think, a lot of the discussion that we will have when reviewing this specific case.
Transcript edited for clarity.