KEYNOTE-012 Investigator Explains Promise of Pembrolizumab in HNSCC

Ranee Mehra, MD

Given the lack of treatment options for patients with recurrent/metastatic head and neck squamous cell carcinoma (HNSCC), the need for novel therapies is dire, says Ranee Mehra, MD, chief of Head and Neck Oncology at Fox Chase Cancer Center.

“There is activity seen with both pembrolizumab [Keytruda] and nivolumab [Opdivo] in this disease and this is significant—especially in a disease that lacks a lot of standard therapies,” says Mehra. “It is very exciting.”

Mehra was the lead investigator on the KEYNOTE-012 trial, which explored pembrolizumab in patients with recurrent/metastatic HNSCC.

The results were promising, says Mehra. Patients who received treatment with pembrolizumab had an overall response rate (ORR) of 18% and a stable disease rate of 17%.

In patients with HPV-positive disease, the ORR was 24%, and in those with HPV-negative disease, the response rate was 16%. There were 4 complete responses (CR) in each of the HPV groups. In patients treated with a prior platinum-based agent, the ORR was 17% (95% CI, 12-23), with a CR rate of 5%. Patients who were treated with prior cetuximab (Erbitux) and platinum therapy demonstrated an ORR of 15% (95% CI, 9-23) and a CR rate of 5%.

OncLive: What were the most significant findings from this study?

In an interview with OncLive, Mehra discusses what the KEYNOTE-012 findings could mean for patients, the growing role for the immunotherapy agent in head and neck cancer, and what else is on the horizon for the treatment landscape.Mehra: KEYNOTE-012 was an expansion of a phase Ib trial of pembrolizumab in squamous cell carcinoma of the head and neck. Patients were enrolled in 2 cohorts. We presented the data at the 2016 ASCO Annual Meeting of the combined analysis of 192 patients, including HPV-positive and HPV-negative patients. Our results show responses in both of those subtypes of head and neck cancer, as well as progression-free survival (PFS) and overall survival (OS) data from the single-arm study.

What should oncologists take away from these findings?

Are there any safety challenges that oncologists should be aware of with pembrolizumab?

What immunotherapy combinations do you see potential for?

The PFS for the combined analysis was 2 months; the OS was 8 months. What is notable is that there were some patients who had CRs and some patients did complete the 24-month treatment period; therefore, the duration of response in a subset of patients was actually quite impressive.These data give us a longer experience with this agent. As we learn more about the role of immunotherapy in head and neck cancer, it is useful to see not just the response rate, but also the duration of activity that is potentially possible in a subset of patients.The toxicity profile is certainly very different from cytotoxic therapy. While most patients tolerate it very well, oncologists need to be aware of the potential for immune-related toxicities. While it is rare, serious toxicities could include colitis or pneumonitis; we are also seeing endocrinopathies with these patients with thyroid dysfunction. It will require oncologists to be aware of a different type of toxicity profile.There have been combinations of PD-1/PD-L1 inhibitors with CTLA-4 inhibitors, and those trials are ongoing.

Where are we right now, in terms of biomarker development?

Another class of agents that is being looked at in combination with PD-1 inhibitors is IDO inhibitors. Certainly, there are a lot of other emerging combinations that are in phase I studies right now and are still being explored.In KEYNOTE-012 and other trials that have been done, we have been looking at PD-1 and PD-L1 expression. From KEYNOTE-012, there are also data regarding the NanoString interferon signature.

What challenges remain with immunotherapy in head and neck cancer?

What are the biggest overall challenges that remain in head and neck cancer?

What do you see on the horizon for head and neck cancer?

Biomarkers should continue to be studied. It is likely premature to say that this is standard of care without more data; however, these types of biomarkers, as well as others, should be incorporated into future trial designs.Clearly, many patients do not respond. Therefore, our challenge for the future is to develop better biomarkers to select which patients could respond and to develop new immune combination therapies. Certainly, what we have seen so far with PD-1 inhibitors in head and neck cancer is encouraging.Our patient population is unique in that they really can be very symptomatic from advanced locoregional disease; therefore, it is very important that we are able to manage the symptoms of the treatment in all of these trials. A small growth in a tumor can result in significant changes in speech or swallowing function. As we develop new combinations of therapies, we have to keep in mind that there is a need for palliative care for these patients through treatment.In the next 5 to 10 years, I hope that there will be more agents that have activity in this disease. Hopefully, we can incorporate these novel agents earlier in the course of the disease, especially for patients with high-risk disease.

Ideally, you want to prevent the recurrence, as well as prevent a situation where patients have more loss of function. We hope that the signals of activity we are seeing in the recurrent metastatic setting can be moved forward into locally advanced disease to cure more patients.

1. Mehra R, Seiwert TY, Mahipal A, et al. Efficacy and safety of pembrolizumab in recurrent/metastatic head and neck squamous cell carcinoma (R/M HNSCC): Pooled analyses after long-term follow-up in KEYNOTE-012. J Clin Oncol. 2016;34 (suppl; abstr 6012).