Vokes on Increasing Impact of Immunotherapy in Head and Neck Cancer

Everett Vokes, MD

Everett Vokes, MD

As immunotherapy continues to make headway in several different cancer types, these agents are only just arriving to the treatment landscape of head and neck cancer.

Earlier this month, the FDA approved nivolumab (Opdivo) for patients with metastatic or recurrent squamous cell carcinoma of the head and neck following progression on platinum-based therapy. That approval was based on results from the CheckMate-141 study, which demonstrated a median overall survival (OS) benefit with nivolumab when compared with investigator’s choice.

This agent has now joined the ranks of pembrolizumab (Keytruda), which received its FDA approval in August 2016 as a treatment for patients with recurrent or metastatic head and neck squamous cell carcinoma following progression on a platinum-based chemotherapy, based on objective response rates in the phase Ib KEYNOTE-012 study.

OncLive: Please provide an overview of your talk at the meeting.

In an interview with OncLive at the 2016 Chemotherapy Foundation Symposium, Everett Vokes, MD, chair, Department of Medicine, University of Chicago Medicine, and 2013 Giants of Cancer Care winner for Head & Neck Cancer, offered his expert insight on the impact that these emerging immunotherapy agents will have on the treatment of patients with head and neck cancer.Vokes: My talk was an overview of where we are in head and neck cancer. Traditionally, we have 2 kinds of settings in head and neck cancer: those with recurrent disease and those that are previously untreated who are treated with curative intent for locoregionally advanced disease. In all of these settings, we have new drugs to consider and we have current treatments to consider.

Let’s start with recurrent disease, and by that, I mean recurrent where it is no longer treatable with salvage surgery or salvage radiation—there are some cases when the tumor comes back and can still be saved for cure. But when that is not an option, or the disease presents with distant metastases, then we look to systemic therapy.

Traditionally, that has been a triple-drug regimen of the so-called EXTREME regimen, with cetuximab (Erbitux) added to a doublet of platinum and 5-FU. Now many people substitute that with a carboplatin regimen or with a taxane, because the toxicity profiles differ, but with that kind of a 3-drug regimen, we get median survival ranges of about 10 months—maybe a little bit longer for those who have HPV-driven tumors, where survival may be longer. And maybe a group in between that is not known to be HPV-driven, but p16-positive; you get into various biomarker differentials there.

That is pretty much the state of the art—a triplet, and then we can follow that with a number of single agents, or sometimes cetuximab is given later, or a drug like gemcitabine is tried, or methotrexate. But it’s overall very unsatisfying.

Like in many other diseases, then, the checkpoint inhibitors come in, and here we now have 2 agents that have been FDA-approved after having been previously IRB-approved for clinical investigations. They are pembrolizumab and nivolumab. Those 2 agents are now available. They are active, and of course, we don’t have direct comparisons, so the trial setup is a little bit different for the 2 of them.

I will start with nivolumab because I think that that is a definitive phase III trial, where patients who had failed a platinum-based regimen within 6 months of initial treatment were then randomized to nivolumab or a standard of care that the investigator could choose from—second-line, in other words—after platinum, and that could be docetaxel, methotrexate, or cetuximab. And in that direct comparison, the checkpoint inhibitor was superior for OS, and it was also superior, but not immediately (there’s a fairly sharp dropoff in curves, but then a separation of curves), for progression-free survival.

It’s the picture we’ve seen in other diseases, that many patients don’t benefit. In fact, probably the majority don’t benefit, but those who do are benefitting enough to lift the entire survival curve to be superior. Furthermore, if a given patient responds, it can be for a long time, certainly much longer than what we’ve observed in chemotherapy-based regimens.

Pembrolizumab is the other drug, and that has not been tested in a phase III trial, but also, extensively, as a single agent in a large phase II trial. It, too, has been approved, and the activity looks very similar to what we would have expected, particularly now knowing how nivolumab faired in comparison. Response rates are around 20%, and sometimes we see very long responses, and what appears to be a group of patients who survive for very long periods of time.

These are the 2 agents, and of course, their toxicity profiles are known. They’ve been described by now, and they are not different in patients with head and neck cancer.

How would one go about choosing between these 2 agents?

Will these checkpoint inhibitors eventually move into the frontline space?

What is it like to be able to witness this explosion of immunotherapy in head and neck cancer, when we really hadn’t seen it previously in the field?

So where we go from here, is to try and move forward to the curative-intent treatment—the patients with locoregionally advanced disease, who nowadays, we treat with concurrent chemoradiation, sometimes with induction. Those are the ones we now want to see if we can either increase cure rates and/or decrease toxicity. So would these drugs—when added adjuvantly during chemoradiation or maybe even before—would they allow us to give less radiation or no chemotherapy and reduce toxicities both acutely and chronically? And it’s of course, in head and neck cancer, the chronic toxicities that we’re always very concerned about: salivary dysfunction, dental problems, neck fibrosis, swallowing problems, and aspiration that comes with that. So that’s really where we’re heading, but we don’t have data yet.I don’t think there’s a rational way to choose. There may be other considerations. It could be scheduling, it could be what a given institution has more readily available, it could be the investigator’s or the treating physician’s personal experience. Those who want to be truly evidence-based may be a little bit more comfortable with nivolumab, since it is a randomized trial. But really, the data are very similar.Yes, I think we need to do that carefully, because frontline is curative, so if we miss in a given patient, it could be the difference between life and death for that individual. This has to be done very considerably, very carefully. The safest way is always to give the standard chemotherapy or radiation therapy as we would, and then give it adjuvantly, as an add-on. The downside to that is we don’t really know who, individually, is benefitting. We need a control group, and we need long follow-up to really assess the impact. Then it could also be given prior to chemoradiation, prior to surgery. There are so-called window-of-opportunity trials. And all of those are currently in progress in one form or another.I think that that’s correct, but it’s also incorrect. We have made a lot of progress in the previously untreated population of patients. The addition of chemoradiation is not totally satisfactory because it is toxic. The discovery of HPV as a profile, and if you will, a new type of head and neck cancer, has allowed us to greatly distinguish patient groups, risk factors, prognosis, and increasingly distinguish treatment approaches.

What steps do you think we still need to take with biomarker research?

Having said that, the addition of a new systemic modality—and as medical oncologists, we are all about systemic therapy, treating the cancer for what it is as a systemic tumor, a systemic disease, because it rarely is just locoregional—it is very exciting.We need really selective biomarkers. PD-L1 is an imperfect selecting factor, although it is getting better in lung cancer, where in frontline, we can now probably pick patients who, based on their PD-L1 assay, should receive the therapy. In head and neck cancer, we similarly know that PD-L1 can enrich for patients who are likely to respond, but we don’t have a clear cutoff.