Ezra Cohen, MD: Jared, let me talk to you. We’ve talked a lot about pembrolizumab and nivolumab, anti-PD-1 agents. We know that there’s another part of that receptor ligand system, and that’s PD-L1. What about the anti—PD-L1 agents in head and neck cancer?

Jared Weiss, MD: We can expect to see data on these fairly soon, probably first for durvalumab, atezolizumab, and avelumab. Although, there are now 23 PD-1 or PD-L1 drugs at some phase of development, so we will see more. Based on data from melanoma and from lung cancer, these agents seem relatively similar in safety, toxicity, and efficacy. I suspect that the story seen in these other cancers will transfer over to head and neck cancer, where the difference between PD-L1 and PD-1 will not be dramatic. I think if we’re going to make dramatic advances, it’s probably going to come in choosing other scenarios in which to integrate these agents, such as for a cure or as combination therapies, such as CTLA4 and IDO inhibitors.

Ezra Cohen, MD: And do you expect to see some data for anti—PD-L1 in the next few months?

Jared Weiss, MD: Yes. I think there are multiple trials ongoing looking, as a theme, at comparing single-agent I-O with chemotherapy or doublet-agent chemotherapy with I-O; in one case, all of them together. We expect readouts on those trials within the next 6 months.

Ezra Cohen, MD: We talked a lot about the current indication for nivolumab and pembrolizumab, but of course, as we’ve seen some promising data and an approval, we’d like to think about moving them forward into different settings. Barbara, what’s going on there?

Barbara A. Burtness, MD: The first thing to say is that we have seen the subset analysis of the CheckMate-141 trial, concentrating on those patients who were treated in the 6-month window after failing chemoradiation, and nivolumab did provide advantage in that setting. And then, as we’ve just been hearing, there are a number of trials ongoing in the first-line setting looking at either PD-1 inhibition or PD-L1 inhibition with or without CTLA4 inhibition—and not in a biomarker-selected fashion. Although I think we’ll see biomarker PD-L1 expression as a correlate. Thinking about what happened in lung cancer, which is not a notably more chemosensitive cancer than head and neck cancer, it did appear as if, in the unselected patients, first-line I-O was not able to beat chemotherapy. But if you had an environment for very high PD-L1 expression, you could expect greater benefit from nivolumab or pembrolizumab.

There’s a completed trial in head and neck cancer that attempts to integrate PD-1 inhibition with chemotherapy, so this was a trial that compared the EXTREME regimen—as we all don’t like to use it as written—with pembrolizumab alone or with the combination of a novel regimen of chemotherapy plus pembrolizumab. And that trial really included the pilot data of pembrolizumab with chemotherapy, and it was very closely monitored for evidence that ought to indicate that it should be closed early. They didn’t find that evidence, completed accrual, and we expect to see the data next year. So, just thinking about what happened in lung cancer, the questions are going to be, how does PD-1 inhibition combine with chemotherapy? That looked good in lung cancer. Is the answer to the question of first-line therapy going to be different for the PD-L1—negative, the PD-L1 modestly expressing, and the PD-L1–rich patient? I wouldn’t be surprised if we did find that there was a different first choice for different patients.

Ezra Cohen, MD: Yes, it’s incredibly exciting to think that immunotherapy will be part of first-line patients with recurrent or metastatic disease. Of course, we have to await the data, but there’s certainly a lot of excitement and a lot of anticipation for that.

Transcript Edited for Clarity