The Current and Future State of Immune Checkpoint Inhibitors in NSCLC - Episode 4

Advanced NSCLC: IO Monotherapy Selection Based on PD-L1 Expression

,

Shared insight on the optimal selection and use of single-agent immunotherapy guided by the testing and expression of PD-L1.

Transcript:

Stephen V. Liu, MD: Now [concerning] these 3 options—pembrolizumab, atezolizumab, cemiplimab [-rwlc]—these are all very good options for PD-L1 high. That represents about a third of lung cancer. Now what if we lowered that threshold down to 1%, which would now allow us to treat two-thirds of patients? Well, that was what [the] KEYNOTE-042 [trial] looked at. This was another randomized trial, looking at patients with advanced non–small cell lung cancer, no prior treatment. But now we’re lowering that threshold to 1%. We’re, on one hand, diluting the population in terms of strong benefit, but we’re also expanding a potential indication by including more patients.

This was a fairly large study, with a randomization to pembrolizumab alone or standard chemotherapy. Looking at KEYNOTE-042, this was, overall, a positive study. And if we look here, in that middle row, looking at that 1% threshold, we have a hazard ratio of 0.80, which was statistically significant. And thus, this study did meet its primary end point of improving outcomes in patients with PD-L1–positive lung cancer. That led to the approval of pembrolizumab, using a PD-L1 cutoff of 1%. But if we look at this study a little more closely, we see that benefit was carried by that PD-L1–high group. If we look at PD-L1 high, where we already know pembrolizumab is better than chemotherapy for [the] KEYNOTE-024 [trial], we see a hazard ratio of 0.68. Clear benefit in that PD-L1 high. That’s not new. What we were interested in for this study was the PD-L1–low group. However, this was not the primary end point of the study. And hence, this is really considered an exploratory analysis. But if we look at that PD-L1–low group, not a whole lot of difference between pembrolizumab and chemotherapy. Not worse than chemotherapy, but not convincingly better. And I will say that pembrolizumab is an improved option in PD-L1–low non–small cell lung cancer. But in my practice, maybe not the preferred option. I don’t know if you feel different, Neal.

Neal Ready, MD, PhD: No, I agree completely. I think that [for] patients who are good candidates for chemotherapy, I would offer them combination therapy in that setting. But in patients who decline chemotherapy or are not appropriate for chemotherapy because of comorbid medical problems, it’s nice to have that option available.

Stephen V. Liu, MD: Absolutely. And [it is] so good to have the option. I’m glad that it’s available, [although] not something I reach for right away, in the right patients. Let’s go to our polling question. In what percentage of patients in your practice today are you using single-agent I/O [immuno-oncology]? A PD-1 inhibitor, PD-L1 inhibitor, whether it’s pembrolizumab, whether it’s atezolizumab, or cemiplimab. Would you say a minority, 5% to 10%, or higher numbers? Key in your selection there. We have many options here. And it really is an embarrassment of riches. I think the key part is we need to deliver immunotherapy in some form. We have a lot of different options to do that. But when are we thinking PD-L1 monotherapy? And it looks like the early results are pretty low. While people continue to key in their answer, at the risk of influencing things, Neal, what’s your answer for this?

Neal Ready, MD, PhD: I would say that I’m probably in the 20% to 25% range, something like that. I think that in patients with a really high PD-L1 score and a modest tumor burden, I am really enthusiastic about using monotherapy. When the PD-L1 score is closer to 50 than 100 and/or there’s a high tumor burden and a lot of symptoms, then I would tend to use combined treatment.

Stephen V. Liu, MD: That’s a great point that you make. We have these cutoffs of 1% and 50%. But PD-L1 expression is a continuous variable. And we know now, from a lot of work done by our colleagues, that 90% is measurably different from 50%. And hence, it is a continuum. The higher expression seems to do better. I’m completely with you that in that ultrahigh population, I’m very comfortable with monotherapy [and] less frequent infusions. You don’t have the toxicities of chemotherapy. And I’m right with you. I’m probably around that 25% range, for the most [part]. Each case is a little bit different. The breakdown generally of PD-L1, what we’re taught is about a third are high, a third are low, and a third are negative. But everyone’s practice is a little different. I think, at our centers, because we see a lot of referrals, especially in the salvage setting, we end up getting a lot more PD-L1 negatives and lows, just for trials and so forth. But I think I’m also around that 20% range. Our group here—it looks like [they are] not favoring a lot of monotherapy. Most answered a low number. But again, [it] really depends on your own specific patient population.

Transcript edited for clarity.