The Approval of Pembrolizumab for Recurrent HNSCC

Transcript:Ezra Cohen, MD: The introduction of immunotherapy in squamous cell carcinoma of the head and neck has really completely changed the landscape. It now provides an option for patients that can extend their survival, that can improve their symptoms, and can do so without the cost of high toxicity. Again, we have to learn to manage those immune-related toxicities. But the percentage of patients who have a serious adverse event from immunotherapy is going to be much, much lower than with chemotherapy. So, it really has completely changed the world.

For a drug like pembrolizumab, we know—and we have data to demonstrate that even in patients who have had platinum-refractory disease and whose cancers have progressed, patients who have had cetuximab-refractory disease and whose cancers have progressed—this drug works. And that is from the KEYNOTE-055 study that enrolled patients with both platinum- and cetuximab-refractory cancers and demonstrated about an 18% response rate, again, at a very favorable toxicity profile.

Conversely, for a drug like nivolumab, we have randomized data that demonstrate that nivolumab extends survival in patients with platinum-refractory disease when compared to a choice of either methotrexate, docetaxel, or cetuximab. And so, now we have 2 agents that are approved in the United States to treat patients with recurrent or metastatic disease, either pembrolizumab or nivolumab, that have demonstrated efficacy with a very favorable toxicity profile.

When we think about pembrolizumab and how we use it in patients with head and neck cancer, we have to realize that it’s a flat dose. And that is a little bit different than how the drug was developed in, let’s say, melanoma or non—small cell lung cancer. Let’s spend a moment thinking about why that is and how this came to be.

When the drug was initially developed, it was developed as a weight-based dosing in the phase l study. It was eventually settled as a 2-mg/kg dose, mostly because the efficacy appeared to be similar than higher doses and the pharmacology appeared to be very similar to 10 mg/kg. When the study then began to enroll patients with head and neck cancer, actually the phase I trial entered those patients at a dose of 10 mg/kg because that was the dose that was being used in other studies.

About the first 60 patients were entered at that dose. We have efficacy data and we have toxicity data at 10 mg/kg. The pharmacology began to get better understood, and it was quickly realized that a flat dose of 200 mg delivered the same efficacy, a very similar toxicity profile, and very similar pharmacokinetics—at least when you think about exposure—to 10 mg/kg of pembrolizumab. And so, for the latter part of the KEYNOTE-012 study, the phase lb trial that enrolled head and neck cancer patients, patients were enrolled at a dose of 200 mg. In fact, the approved dose in head and neck cancer is 200 mg. That offers some advantages over weight-based dosing, but it also has introduced, unfortunately, some confusion—when you think about it—in the context of other diseases like melanoma or non—small cell lung cancer. What is going to make things easier in the long run is that in all likelihood, pembrolizumab is going to go to a flat dose across all of its diseases once we have the data in all of those other cancers.

Nabil F. Saba, MD: So, KEYNOTE-012 is an open-label phase 1b clinical trial that enrolled patients with recurrent or metastatic head and neck cancer that had failed a platinum-based therapy. The trial required that patients have a PD-L1 expression of more than 1%. And the trial basically led to about a 17% rate of grade 3 and 4 toxicities incurred from the single-agent pembrolizumab. In addition to this, the response rate that was observed was about 18% in a pretreated population.

KEYNOTE-055 followed up on the results of KEYNOTE-012 and treated about 172 patients with recurrent or metastatic head and neck cancer. These patients also had to fail platinum-based therapy and had to fail cetuximab as well. And the preliminary information from this trial, that was reported at ASCO of 2016, basically confirms the results of KEYNOTE-012, which is that pembrolizumab in a heavily pretreated patient population results in a response rate of about 18%. So, this basically confirms the data from KEYNOTE-012.

Transcript Edited for Clarity