Benefit of Pembrolizumab in NSCLC: The KEYNOTE Trials

Alexander Drilon, MD:The KEYNOTE-024 and -042 trials randomized patients in the first-line setting to either immune therapy with pembrolizumab or the prior standard of care, which is platinum-doublet therapy. The difference between the 2 trials is that KEYNOTE-024 looks specifically at patients with a tumor proportion score of 50% or greater by IHC. And in the KEYNOTE-042 study that was more recently presented, the cutoff was 1% or higher. Now, both trials were positive for a survival advantage—ie, progression-free survival and overall survival for KEYNOTE-024 and overall survival for KEYNOTE-042. However, if you look at an exploratory analysis of KEYNOTE-042, it seems like the benefit is largely limited to patients, again with a proportion score of 50% or greater. So, the take-home message for me is that we now have independent trials showing substantial benefit for patients with PD-L1—expressing cancers that are 50% or greater with single-agent pembrolizumab in the first-line setting. However, for patients who fall below that cutoff, despite the fact that there was a signal for survival in the KEYNOTE-042 trial, I might consider other strategies like giving chemotherapy plus pembrolizumab, like carboplatin/pemetrexed/pembrolizumab for those patients for example.

Justin Gainor, MD:For patients with a TPS of 50% or greater, we have clear level 1 evidence showing that pembrolizumab results in improvements in overall survival in those patients, based upon both the KEYNOTE-024 study as well as the KEYNOTE-042 study. In my practice, I tend to use pembrolizumab monotherapy in those patients unless patients have a very large disease burden in which I really want to get a response to therapy, in which case I may use the triplet combination of platinum/pemetrexed plus pembrolizumab. Right now, we don’t have enough data to really compare the role of pembrolizumab monotherapy versus the triplet combination for patients who are high PD-L1 expressers, and by that, I mean 50% or greater. So, we really need additional follow-up from KEYNOTE-024 as well as KEYNOTE-189 to help guide clinicians for those particular patients.

With respect to patients who are TPS 1% or greater, KEYNOTE-042 did show that there was an improvement in survival compared with platinum-doublet chemotherapy for those patients. However, it really looks like all of that benefit was driven by the high—PD-L1 expressers, the greater than 50% patients. Because if you look at the post hoc analysis looking at the 1%-to-49% patients, there wasn’t an improvement in those patients. And it really seemed that it was the greater-than-50% patients who were deriving the overall survival benefit in that study. And so, in my mind, as long as someone has reasonable performance status to tolerate a triplet combination for patients who are between 1% and 49% PD-L1 expression, I would actually give those patients a combination of chemotherapy plus pembrolizumab.

For patients who may have an inadequate performance status or who are borderline performance status and yet have a PD-L1 expression score of 1% or greater where you don’t think that they’d be able to tolerate a triplet or a doublet chemotherapy may be challenging, those patients you may treat with pembrolizumab monotherapy, provided the understanding that in KEYNOTE-042, patients still had to meet a performance status criterion for study entry and those patients may not be adequately captured in that study. But I do think it would be a reasonable approach.

Transcript edited for clarity.