Immunotherapy for Head and Neck Cancer Expands, but Concerns Persist About Patient Selection and Toxicities

Get Permission

SOME PATIENTS with advanced head and neck cancer may achieve durable responses with immunotherapy, and recent trial results suggest first-line immunotherapy may increase survival among patients with recurrent or metastatic disease. However, concerns remain about selecting patients most likely to benefit from this treatment and managing treatment-related toxicity. Advances and concerns about the use of immunotherapy for head and neck cancer were explored at the 2018 Multidisciplinary Head and Neck Symposium, sponsored by the Robert H. Lurie Comprehensive Cancer Center and Feinberg School of Medicine, Northwestern University, Chicago.1-3

Tanguy Seiwert, MD

Tanguy Seiwert, MD

“We now have some patients with metastatic head and neck cancer who are 4 years out from immunotherapy, which previously was unheard of,” stated Tanguy Seiwert, MD, Assistant Professor of Medicine, Section of Hematology and Oncology, University of Chicago Medicine. Eventually, immunotherapy “is going to change everything we do, probably for many cancer types and definitely for head and neck cancer.”

Duration of Responses

JEFFREY A. SOSMAN, MD, Professor of Medicine, Robert H. Lurie Comprehensive Cancer Center of Northwestern University, reviewed the milestones in the development of immune checkpoint inhibitors for several types of cancer. In testing anticytotoxic T-lymphocyte–associated protein 4 (anti–CTLA-4) agents, he noted, “The key finding was not how many patients responded, because in the initial study, it was about 10% to 15%.” It was the “potentially lifelong” benefit among those who responded.

Jeffrey A. Sosman, MD

Jeffrey A. Sosman, MD

Likewise, with the anti–programmed cell death protein 1 (anti– PD-1) agent nivolumab, “the real major finding is not the response rate,” Dr. Sosman said. “It is the duration of the responses.” These responses are ongoing, and they are maintained in many patients without continued therapy. “How long patients should be on therapy is still unknown,” he added.

Immunotherapy does not work for everybody, admitted Dr. Seiwert, but when it does work, “that patient will never forget. These treatments are non–cross-resistant with other modalities, including chemotherapy or radiation therapy, and they have durability, which is different from anything else we have.”

Move to First-Line Setting?

“UNTIL RECENTLY, the situation was pretty dismal for patients with metastatic head and neck cancer,” Dr. Seiwert remarked. “We cure a lot of patients upfront, but we don’t do very well once the disease is metastatic.” The response rate using cetuximab “is only about 10%,” he reported. “Progression-free survival was about 2 months with cetuximab. Overall survival was about 6 months. An effective treatment was a large unmet need.”

“We now have some patients with metastatic head and neck cancer who are 4 years out from immunotherapy, which previously was unheard of.”
— Tanguy Seiwert, MD

Tweet this quote

Two checkpoint inhibitors, nivolumab and pembrolizumab, have been approved for patients with recurrent, unresectable, or metastatic squamous cell carcinoma of the head and neck and disease progression on or after platinum-containing chemotherapy. Dr. Seiwert noted, however, that in the near future, pembrolizumab could be “moving forward to the first line.”

The results of the KEYNOTE-048 study, reported at the European Society for Medical Oncology (ESMO) 2018 Congress,4 showed that overall survival for patients with advanced head and neck cancer was significantly improved with first-line pembrolizumab alone compared with the standard EXTREME regimen of platinum, fluorouracil, and cetuximab. Among patients with high expression of the programmed cell death ligand 1 (PD-L1), overall survival was 14.9 months vs 10.7 months for the EXTREME regimen—a 39% improvement. For patients with low PD-1 expression, the overall survival was 12.3 months vs 10.3 months for the EXTREME regimen—a 22% improvement.

“Moving these treatments forward earlier makes sense,” Dr. Seiwert said. “If you move immunotherapy agents forward to highly immune-competent patients, they actually work better, although they will probably be associated with a bit more immune toxicity.”

Dr. Seiwert said he expected first-line pembrolizumab to be approved in the near future. “These data are practice-changing. The standard of care is literally changing under our feet, and I think it will continue to change over the next few years. My hope is that in the future, it will not be a one-size-fits-all treatment. We will actually personalize immunotherapy.”

Room for Improvement

IN KEYNOTE-048, objective response rates were actually lower for patients receiving pembrolizumab alone. As Dr. Seiwert explained, “These treatments are very different from chemotherapy or targeted therapy. Patients live longer, even though they have only modest response rates. That is what we really want to do: give the patient more time and also good quality of life.”

According to Dr. Seiwert, there is still a lot of room for improvement. “We still have to help 60% to 80% of patients. However, at least for some patients, we have a proof of principle that we can change the course of metastatic disease.”

Combining Immunotherapies

DR. SOSMAN explained the rationale behind combining anti–PD-1 and anti–CTLA-4 antibodies. It is based on “the idea at both ends of the T-cell immunity cycle, you can block the antigen-presenting activation as well as the effector phase in the tumor.” Although combining nivolumab and the anti–CTLA-4 agent ipilimumab improved survival, “the downside of the combination is certainly increased toxicity,” he noted.

“We need to figure out the most efficient way to do trials with combination therapies as well as what trials are the most important to do,” Dr. Sosman said.

“I believe these agents may play a role also in locoregional advanced head and neck cancer,” Dr. Seiwert commented. There are a large number of trials, essentially using checkpoint inhibitors as add-on therapies to other modalities, such as chemoradiation therapy. “The question is whether this will change the cure rates” rather than just improve survival by a couple of months. 

“There is also an opportunity to give immunotherapy before surgery, as neoadjuvant treatment,” Dr. Seiwert stated. “We have a little bit of experience with that now in about 10 patients. We see dramatic shrinkage, but this is still experimental.”

Predicting Response

“THERE IS now a push toward biomarker testing,” Dr. Seiwert said. “PD-L1 is a modestly helpful marker,” he added, and “can be expressed on tumor cells or immune cells.”

In early studies of anti–PD-1 agents, “it looked like if you don’t express PD-L1, you don’t respond,” Dr. Sosman said. “However, this has not held up, virtually in any disease. There is a relationship, but it is not black and white. Patients who are PD-L1–negative, however you define it, may still respond to immunotherapy.”

“Patients who are PD-L1–negative, however you define it, may still respond to immunotherapy.”
— Jeffrey A. Sosman, MD

Tweet this quote

Mutation burden is now “most in vogue” as a biomarker, Dr. Sosman said. “These mutations are not necessarily related to driving the cancer,” he explained. “The tumors with the highest mutation rate are those with the highest responses. It is not across the board, and we know patients with a high mutation rate can fail to respond and vice versa, but it is clearly a factor.” Cancers with high mutation rates include head and neck cancer, he noted, and “certainly smoking is a big factor.”

Both human papillomavirus (HPV)-negative and HPV-positive head and neck tumors “are characterized by having a lot of genetic changes,” Dr. Seiwert noted. “The more changes you have, the more opportunities you have for the immune system to recognize the cancer. However, tumors are smart: they inherently figure out ways around the immune system.”

“It has become clear that some tumors are immunologically ‘cold’ and are not inflamed. Those tumors are not going to respond, and those tumors with active inflammation do respond. They can be responsive to anti–PD-1,” Dr. Sosman said. “For the cold tumors, you need to do even more. The idea of bringing in T cells can be approached in many ways,” he said, including with receptor agonists, activating antibodies, generating the needed T cells. Chimeric antigen receptor T-cell therapy could also be used to boost the ability of a patient’s T cells to recognize and attack cancer.

Immune Checkpoint Toxicity

ALTHOUGH IT is not known what regimens will be combined with anti–PD-1 agents in the future, “we know that almost everything you combine them with is still going to have the problem of immune checkpoint toxicity,” Dr. Sosman explained.

Autoimmunity can affect any organ system, but skin, gastrointestinal (GI), liver, and endocrine organs are most common. Multiple organ systems can be affected, either concurrently or serially, reviewed Dr. Sosman. “Toxicity mostly occurs during the first 6 months, but even after 1 or 2 years, you still see some toxicities. There are patients out 2 years who develop severe toxicity and die as a result. The onset is not necessarily predictable.”

“I always tell patients this is a natural treatment. It makes your own body’s immune system stronger. And they love this idea,” Dr. Seiwert said. “At the same time, it also explains the side effects. In principle, immunotherapy can generate autoimmune disorders. Most of them are mild, such as skin disorders. Sometimes, however, about 1% to 3% of the time, you can have a serious side effect, such as pneumonitis, which is potentially life-threatening. However, if it is recognized early, you can treat it.”

“The vast majority of toxicities are completely reversible,” Dr Sosman said. An exception is endocrine toxicity. “Patients can develop hypothyroidism and likely will be on levothyroxine for their entire lives. The bigger issue is adrenal insufficiency,” he noted. “Those patients, in my experience, are lifelong steroid-dependent.”

Toxicity/Efficacy Relationship

THERE IS likely a relationship between toxicity and efficacy, although it is not clear-cut, and probably there are a number of other factors as well, Dr. Sosman commented. “I tell patients all the time, if the toxicity is so severe that you stop therapy, it does not necessarily mean the therapy failed. In fact, it may be the opposite.”

Thus, side effects from immunotherapy are an essential indicator, Dr. Seiwert stated. “If you see no side effects, there is no main effect. The key is to manage them appropriately.” He tells his patients not to be afraid to mention they are experiencing a severe side effect. He tells them, “if you have a severe side effect, it is important to know early, because you can stay on treatment if you treat it early.”

Overlapping and Synergistic Toxic Effects

“THE TOXICITIES generally are completely overlapping,” Dr. Sosman said. “The one toxicity I don’t believe we have ever seen with anti–CTLA-4, but we do see with PD-1, is diabetic ketoacidosis.” This side effect has occurred among patients who did not have diabetes and came to the ER with a glucose of 700 mg/dL and a pH less than 7. Although it is treatable, it is not reversible, so it is something you need to be aware of,” Dr. Sosman remarked.

“The severity is dependent on the treatment, and certainly an anti–CTLA-4 agent has more toxicity than all the anti–PD-1 agents,” he noted. In addition, “with anti–CTLA-4 agents, the toxicity is dose-dependent. That is not at all true with the anti–PD-1 agents,” Dr. Sosman said.

Receiving two different agents sequentially might avoid toxicity. “However, rechallenging with the same agent may lead to toxicity more often, but not 100% of the time,” Dr. Sosman noted. “A high-grade adverse event with one class of agents does not preclude safe administration of the other class.”

Although the number of combination therapies is growing, toxicities are the major barrier to more effective combination immunotherapy, creating a sort of “synergistic toxicity,” Dr. Sosman said. About “60% of the combinations are associated with grade 3 and 4 toxicities.” Rare toxicities such as myositis or neurotoxicity, which can be life-threatening, are of serious concern, shared Dr. Sosman. However, such deaths are rare, he noted, occurring “in less than 1% of patients.”

Finally, Dr. Sosman noted the importance of multidisciplinary care. “You need to have a group of internists with whom you are familiar, such as in GI, pulmonary medicine, endocrinology, and neuro-oncology, to help you manage these toxicities.”

DISCLOSURE: Dr. Seiwert has received honoraria from AstraZeneca, Merck, and Bristol-Myers Squibb; is a consultant/advisor for AstraZeneca, Bayer, Roche Molecular Diagnostics, Nanobiotix, Celgene, Innate Pharma, Merck, Bristol-Myers Squibb, and Aduro Biotech; has received institutional research funding from Merck, Bristol-Myers Squibb, and Jounce Therapeutics; and has received travel/accommodations/expenses from Merck, Bristol-Myers Squibb, Innate Pharma, Nanobiotix, Bayer, Roche Molecular Diagnostics, AstraZeneca, and Celgene. Dr. Sosman has received honoraria from and is a consultant for Genentech and MSD.


1. Seiwert T: Immunotherapy for recurrent/metastatic head and neck cancer. 2018 Lurie Cancer Center Multidisciplinary Head & Neck Symposium. Presented November 11, 2018.

2. Sosman J: Advances in immunotherapy for cancer. 2018 Lurie Cancer Center Multidisciplinary Head and Neck Symposium. Presented November 11, 2018.

3. Sosman J: Management of adverse effects with immunotherapy. 2018 Lurie Cancer Center Multidisciplinary Head and Neck Symposium. Presented November 11, 2018.

4. Burtness B, Harrington KJ, Greil R, et al: KEYNOTE-048: Phase III study of first-line pembrolizumab for recurrent/metastatic head and neck squamous cell carcinoma. 2018 ESMO Congress. LBA8_PR. Presented October 22, 2018.