Insights From: Ezra Cohen, MD, UC San Diego Moores Cancer Center; Robert L. Ferris, MD, PhD, University of Pittsburgh Cancer Institute; Nabil F Saba, MD, FACP, Emory University
Ezra Cohen, MD: Without a doubt, immunotherapy has changed the landscape for patients with squamous cell carcinoma of the head and neck. We’ve seen approval of both nivolumab and pembrolizumab for patients with recurrent and metastatic disease. But the real excitement is what we think is going to be coming in the near future and what we’re already seeing in early phase clinical trials, especially with respect to combining different immunotherapy drugs.
For instance, we see response rates in the phase I and phase II studies that are higher than we would expect with single-agent PD-1 or anti-PD-L1. We’re beginning to see that many of those responses are what we call “deep responses”—that is, an almost complete remission or a 90% disappearance of those cancers. What we’ve also come to understand is that the patients that have those deep responses are the ones that usually have response durations that are measured in not many months, but, believe it or not, in years. And we’re beginning to see patients, in fact, with those deep responses. They are now 2 years or 3 years out from therapy without evidence of recurrence and ongoing response. And so, as we begin to combine these immunotherapy agents, you can imagine that there’s a lot of excitement about what they can do.
Robert L. Ferris, MD, PhD: Once a single immune checkpoint receptor is clinically active, such as nivolumab and pembrolizumab, it’s natural to try to combine them since there’s a large number of immune checkpoints expressed on cancer patient lymphocytes. Likewise, other diseases, such as melanoma and lung cancer, have combined immune checkpoint antibodies targeting not only PD-1 but also CTLA4. So clinical trials are now open in head and neck cancer, targeting both PD-1 and CTLA4. Positive data have been reported for lung cancer, targeting the 2 immune checkpoint receptors—PD-1 and CTLA4—and, clearly, data in melanoma would suggest that the trials in head and neck cancer, targeting both PD-1 and CTLA4, will be beneficial.
The toxicities appear to be higher when you target both immune checkpoint receptors because you stimulate a stronger anti-tumor or autoimmune toxicity profile. So, when you stimulate the immune system more strongly by targeting both CTLA4 and PD-1, the likelihood of an autoimmune adverse event increases. We have seen that with melanoma and lung cancer, but the clinical benefit outweighs the side effects. And so, the promising benefit of targeting immune checkpoint receptors at the same time leads us to run clinical trials, which are currently open and accruing. We don’t yet have clinical data for head and neck cancer with combination immune checkpoint receptors, besides the reported nivolumab/lirilumab data.
The use of single-agent immune checkpoint inhibitors has been promising and yet sobering at the same time because 80% or more of patients don’t derive measurable benefit. And so, with a response rate of 15% to 20% with nivolumab monotherapy, there was a great desire to combine anti–PD-1 nivolumab with other immune checkpoints. A promising one, called the killer inhibitory receptor, or the KIR, has an antibody that targets it, called lirilumab. And recent data have been promising and exciting, and were presented at a recent Society for Immunotherapy of Cancer meeting. Recent data combined nivolumab with lirilumab, demonstrating that outcome is improved in a small subset of patients, around 30 patients, over nivolumab monotherapy alone. This would suggest that targeting multiple checkpoint immune receptors could lead to enhanced benefit in head and neck cancer patients.
Nabil F. Saba, MD: So, immune therapy is clearly moving towards combination therapies, and this is not just in head and neck cancer, but in different tumor types. And there are several arms of the immune system. There’s the innate portion of the immune system that relies on the first defense mechanisms, including natural killer cells. Lirilumab is a drug that basically inhibits what we call the KIR, which are killer immunoglobulin-specific receptors. And by doing that, it inhibits the activity of natural killer cells. In a phase I and II trial, it was combined with the IgG4 antibody—which is against PD-1—nivolumab. And the results of this trial are interesting because even though the number of patients with head and neck cancer were small—there were about 41 patients with head and neck cancer enrolled—about 29 patients were evaluable for response. There was an about 24% response, which is superior to what we have seen with single-agent checkpoint inhibition. The more interesting aspect is that there were about 17% of patients who had significant responses and deep responses, with more than 80% reduction in tumor sites.
The other interesting aspect is if you looked at patients who had PD-L1 expression of more than 1%, the response rate was about 41%. So, even though the numbers are small, I think this is the first combination of its kind where an anti-KIR is combined with a PD-1 inhibitor. And the early results are encouraging and are certainly worth pursuing in future trials, given the responses we’re observing.
Ezra Cohen, MD: KIR is a receptor that is expressed on different cell types, but predominantly expressed on natural killer cells, or NK cells. It’s really somewhat a different part of the immune system. And so, one can begin to imagine that when you combine drugs that have different mechanisms that may be able to activate different parts of the immune system, that perhaps you begin to get additivity or even synergy. We’re beginning to get hints of that with that combination. For instance, the response rate was higher than we would expect with nivolumab alone, about 25%. And many of those responses were deep responses—patients who had complete responses or patients who had very good partial responses.
What was interesting about the data—and these are early data, small numbers—the responses appear to be restricted to tumors that expressed PD-L1 and almost paradoxically in patients who had HPV-negative disease. We’ll see how that plays out in larger studies. As more patients begin to get on the trials, I’m sure we’ll see responses in HPV-positive patients. But early on, we may be getting a signal in terms of which patients benefit from that combination.