Enter PD-1 Inhibitors; A New Paradigm for HNSCC

Transcript:Ezra Cohen, MD: They’re certainly exciting data, and we’re going to talk a lot about that in just a second. But, let’s frame it. Kevin, let me ask you: what’s the rationale for immunotherapy in this disease?

Kevin Harrington, MD, PhD: That’s a very good question, and there’s a short answer and a long answer. The short answer is that in the second-line setting, we had nothing else that worked, so why not try it.

Ezra Cohen, MD: Why not try it.

Kevin Harrington, MD, PhD: But actually, there’s a longer answer that maybe will sound a little more scientifically cogent. We know that a number of patients with head and neck cancers have HPV-positive virally-driven cancers, and we know from other settings that virally-driven cancers may be good targets for immunotherapy approaches. The problem is up until now, we haven’t had very good immunotherapy approaches to break that tolerance of the tumor and allow the immune system to engage with it. So, that’s one aspect of it.

I think the other aspect is that we now have those data that have emerged from clinical trials where we have tried it. We’ve known that some tumors have had an appearance of having had T-cell infiltrates. We know that they look like they could be candidates for immune responses. And, indeed, when we begin to break those data apart—and Tanguy has been instrumental in helping us understand this—we know that when we try patients with immunotherapies in the second-line setting that we do see objective responses. We sometimes see deep, really clinically meaningful, and beneficial responses, and they can be extremely durable. And we’re beginning to think maybe, on occasion, we may cure some patients with relapsed metastatic head and neck cancer.

So, I think the data are very clear, both with pembrolizumab and with the CheckMate-141 data, in the relapsed metastatic setting for the use of immunotherapy in the platinum refractory setting. That then opens the door, I think, for a rationale for testing these agents earlier in the patient’s course. And, of course, we’re now in the middle of a real rush to complete phase III studies in the first-line setting, and those will read out in the coming year. I think we may completely change the paradigm of what we need to do. Because, of course, the setting in the first-line treatment approach is actually to challenge extreme regimen triplet chemotherapy and try to replace that with something that is potentially better tolerated and works better.

Ezra Cohen, MD: And I’m going to touch on that in just a couple of minutes, Kevin, but I want to just follow up on something that you said about the rationale in HPV-positive disease. We see these agents work in HPV-negative disease as well. And Tanguy, maybe you can provide some insight on why that might be.

Tanguy Y. Seiwert, MD: Initially, when we looked—and others have looked at tumor inflammation and associated with that also—at PD-L1 expression, there was a sense of maybe HPV-positive tumors are extremely sensitive. In fact, they look like melanoma; they look very inflamed. And people were thinking maybe it’s as effective as in melanoma. However, when we actually saw the data, the efficacy was actually the same in HPV-positive and HPV-negative tumors. And I would like to add that the HPV-negative tumors, they’re actually very similar to lung cancer and we see a similar level of activity in HPV-negative tumors. And when you look at it, there was a smoking-associated tumor, and smoking causes, you know it’s like a cluster bomb that causes mutations. And when we have high mutation loads, those diseases are oftentimes very amenable to immunotherapy as well. So, I think the rationale for HPV-positive tumors may be the virus, but also mutations. But for HPV-negative tumors, it’s likely also the mutation load contributes to that. So, the efficacy we saw is actually good in both settings.

There is a little bit of debate on whether you see somewhat more activity in the HPV-positive patients. They’re different studies. So, in KEYNOTE-012—the study that tested pembrolizumab in head and neck cancer and ultimately actually led to the accelerate approval of pembrolizumab—we saw the same amount of activity in both diseases: HPV-positive and HPV-negative. However, in CheckMate-141, in some subanalyses, maybe there was more of an effect on survival in the HPV-positive. But in my mind, the HPV status should not actually dissuade us one way or the other from using immunotherapy. This is clearly a drug that’s active in both HPV-negative and HPV-positive tumors.

Ezra Cohen, MD: So, let’s drill down on some of those data. And Viktor, we’ve seen the data from KEYNOTE-012, we’ve seen the data from KEYNOTE-055. Please tell us, first of all, what those studies were and what are the practical implications for the treatment of our patients?

Viktor Grünwald, MD: The KEYNOTE-012 study was a phase I study extolling different pembrolizumab dosing, one of which would be per kilogram or flat dosing. So, there were different cohorts. It’s mixed in terms of the patient population, so it’s not that precisely clear cut in terms of that specific operation. And is it a problem? I don’t think so, no. It’s an early clinical study with a decent size. It’s 170 patients or more for answering the question. First of all, it’s the question: is it safe? Is it safe to administer? And yes, it is. It’s within the range of toxicities, what we have seen in other studies, because we’re approaching a very sick and morbid tumor in the patient, and, therefore, it really might be different in terms of outcome of toxicity. I think it’s very reassuring that we see the same amount of toxicity as we do see in others. And the grade 3-4 toxicities are really in the range of between 10% and 15% in all of these studies actually.

I think it’s a very good message that we have. And, as Tanguy referred to, is there a difference between HPV-positive and HPV-negative disease? The answer is no, at least in the subgroups that we have in all of these studies. But you should not overinterpret this data. It’s a subgroup analysis. We don’t have specific clinical trials prospectively testing for a specific subgroup of patients the best treatment approach with PD-1 or PD-L1 inhibitors. So, yes, I think it’s very reassuring. And there’s activity of a 15% response rate. I think this is a very good signal.

And then the question will be, how does it really translate into a patient population that is pretreated with platinum and pretreated with EGFR inhibitors in both of them? That is what KEYNOTE-055 data looked into, and it kind of mirrored what we’ve seen in KEYNOTE-012 basically. And I think that’s the story that we see for pembrolizumab in head and neck, having quite a bit of patients actually already being treated in single-arm clinical studies, which is somewhat unusual but reflects the speed of knowledge that we have that actually leads to approval.

Ezra Cohen, MD: Yes, we’re almost doing away with the traditional phases of studies and morphing into a different clinical trial paradigm. And you’re right, now between those 2 trials, KEYNOTE-12 and KEYNOTE-55, we’re talking about over 300 patients that have been treated with pembrolizumab.

Transcript Edited for Clarity