Immunotherapies Form New Frontier in Treating Head and Neck Cancers

Tanguy Seiwert, MD

In August 2016, the FDA approved pembrolizumab (Keytruda) for patients with platinum-refractory squamous cell carcinoma of the head and neck (SCCHN).¹ Not only was it the first immunotherapy approved for head and neck cancer (HNC), but it marked the first new drug approval for HNC in the United States in 20 years.

“Now we have an agent that really changes the paradigm—a new class of treatment—and we are seeing amazing benefit in some patients,” said Tanguy Seiwert, MD, during an OncLive Peer Exchange® panel held during the 2016 European Society for Medical Oncology (ESMO) Annual Meeting.

Less than a month later, the menu of immunotherapy options expanded as the FDA approved nivolumab (Opdivo) for the treatment of patients with recurrent or metastatic SCCHN with disease progression on or after a platinum-based therapy.

During the Peer Exchange, the panelists provided an overview of the immunotherapy terrain in HNC, a discussion that was filled with considerable hope and excitement. “When we try immunotherapies in the second-line setting, we see objective responses—sometimes deep, clinically meaningful, extremely durable responses—and we’re beginning to think that maybe, on some occasions, we may be able to cure patients with relapsed metastatic head and neck cancer,” said Kevin Harrington, MD, PhD. This is especially remarkable since such patients have generally had a survival of ≤1 year.

The panelists concurred that the care of patients with HNC will evolve significantly over the next 5 to 10 years, as the tip of the immunotherapy iceberg is just starting to be scratched. During the Peer Exchange, they provided a rationale for using immunotherapies in HNC, including human papillomavirus (HPV)-positive and HPV-negative disease; outlined key immunotherapy studies; and offered their thoughts on the future of immunotherapies in HNC, including use of biomarkers to guide therapy and the opportunity to improve response by using combination treatments.

Rationale for Immunotherapy in HNC

“Next-generation sequencing efforts are beginning to shed light on the hidden complexities of these tumors, leading to the identification of multiple molecular subtypes,” said Ezra Cohen, MD, who served as moderator for the session. “As key differences between tumors, with and without HPV infection, are beginning to emerge, the challenge is to find ways to use this information to personalize treatment for individual patients.”In patients with locally advanced HNC, HPV status has generally determined outcomes, with HPV-positive patients having a good prognosis and higher likelihood of cure, and HPV-negative patients having a poorer prognosis and a lower likelihood of cure.

However, outcomes with conventional therapy in recurrent metastatic disease have been poor across the board, especially in the setting of platinum- refractory disease, indicating a tremendous unmet need. Before pembrolizumab was approved in this setting, the recommendation was to use a taxane, such as methotrexate or cetuximab (Erbitux), as a single agent, but the outcomes have been unsatisfactory. In contrast, immunotherapy studies have shown promising results in these patients, with HPV-negative patients also benefiting.

“The rationale for [using immunotherapies] for HPV-positive tumors may be the virus, as well as mutations, and for HPV-negative tumors, it’s likely the mutation load,” said Seiwert. He explained that HPV-negative tumors are often smoking-associated tumors and, therefore, have high mutation loads, a factor that has been associated with good response to immunotherapy, whereas HPV-positive tumors resemble melanoma, with significant inflammation, another factor associated with good response.

Although efficacy was found to be the same for HPV-positive and HPV-negative tumors in KEYNOTE-012, which was the study that led to pembrolizumab’s approval for HNC, some CheckMate-141 subanalyses suggest there might be slightly more activity in HPV-positive patients, noted Seiwert.

Key Pembrolizumab Studies

KEYNOTE-012 Expansion Study

Despite such findings, he said, “HPV status should not actually dissuade us one way or the other from using immunotherapy—it’s clearly active in both HPV-negative and HPV-positive tumors.” And, as Harrington pointed out earlier in the discussion, since nothing else works well in the second-line setting, “why not try it?”The panelists proceeded to provide an overview of several instrumental pembrolizumab studies, including the KEYNOTE-012 expansion study and KEYNOTE-055 studies, and of the phase III CheckMate-141 study, which paved the way for nivolumab’s approval.^2-4 They also discussed a subanalysis of CheckMate-141 presented at ESMO that demonstrated good patient-reported outcomes following nivolumab therapy, lending further support to its use in SCCHN.⁵The phase Ib KEYNOTE-012 expansion study administered 200 mg of pembrolizumab intravenously once every 3 weeks to 132 patients with recurrent or metastatic SCCHN, irrespective of their programmed death-ligand 1 (PD-L1) or HPV status.² Primary endpoints included overall response rate (ORR), and safety and secondary endpoints included progression-free survival (PFS), overall survival (OS), and PD-L1 expression’s impact on response.

Pembrolizumab was well tolerated, and yielded a clinically meaningful ORR with evidence of durable responses; median duration of response was not reached. The ORR was 18% by central imaging vendor review and 20% by investigator analysis. A statistically significant increase in ORR was observed for PD-L1—positive versus PD- L1–negative patients (22% vs 4%, respectively; P = .021).

At 6 months, PFS and OS rates were 23% and 59%, respectively. “And we have patients living far beyond what we usually expect for metastatic disease...we now have patients who have completed 2 years of treatment in a setting with a median life expectancy of about 6 months,” revealed Seiwert, who is involved with the study.

Pembrolizumab was also well tolerated. Grade ≥3 events occurred in 9% of patients. “[This is] within the range of toxicities that we have seen in other studies,” said Viktor Grünwald, MD. “Because we’re approaching a very sick and morbid patient population, we might expect different toxicity outcomes, so I think it’s very reassuring that we’re seeing the same amount of toxicity as in other studies,” he explained.

While checkpoint inhibitors are generally well tolerated and have favorable toxicity profiles, the panelists warned that severe side effects can occur and careful patient monitoring is required. A key concern they discussed is pneumonitis.

KEYNOTE-055 Preliminary Results

“Although it only occurs in about 1% to 2% of patients, you must screen for it because it’s life threatening,” said Seiwert. “If somebody says, ‘I am short of breath’ or ‘I have a little bit of a cough,’ I scan them right away to look for it,” he said, explaining that immediate treatment with high-dose steroids is warranted.KEYNOTE-055 enrolled 172 patients with recurrent or metastatic SCCHN to receive pembrolizumab 200 mg every 3 weeks after progression on platinum plus cetuximab. The preliminary analysis, which reported on 92 evaluable patients, was initially presented at the 2016 American Society of Clinical Oncology (ASCO) Annual Meeting.³ Primary endpoints include ORR and safety.

As with KEYNOTE-012, pembrolizumab was found to be well tolerated and to have significant antitumor activity, with 17% to 18% response rates. Evaluation of the full study cohort will include analyses of HPV status and response by anatomic site. “I think that’s the story we’re seeing for pembrolizumab in head and neck cancer—patients are already being treated in single-arm clinical studies, which is somewhat unusual, but reflects the speed of knowledge that we’re gaining that is leading to approvals,” said Grünwald.

Pivotal Nivolumab Study

In CheckMate-141, 361 patients with recurrent or metastatic SCCHN were randomized 2:1 to nivolumab (n = 240) or investigator’s choice (IC) of single-agent taxane (n = 121).4 The primary endpoint was OS and secondary endpoints included ORR and PFS. These endpoints were compared between treatment arms by PD-L1 expression (≥1%, 5%, and 10%) and HPV status.

Nivolumab demonstrated significant OS bene t compared with IC, with the greatest improvements in OS and ORR observed with PD-L1 expression ≥1%. “The data monitoring committee closed the study early because it had crossed the predefined statistically significant boundary with a median OS of 7.5 versus 5.1 months,” said Harrington, who is involved with the study. “And at the landmark analysis at 1 year, 36% of patients receiving nivolumab remained alive compared with about 17% in IC.”

Nivolumab also had a lower incidence of treatment-related adverse events (TRAEs) than IC. Any grade TRAEs occurred in 59.3% and 77.5% of patients on nivolumab or IC, respectively. Grade 3/4 TRAEs occurred in 13.6% and 35.1% of patients, respectively, indicating the treatment is well tolerated, which also translated to improvements in quality of life in the patient-reported outcomes study.⁵

“A very detailed analysis of patient-reported outcomes using 3 well-validated questionnaires showed nivolumab was able to maintain good patient-re- ported outcomes in terms of their quality of life, their functioning, and their symptom scores, whereas IC showed serious detriment in those scores,” said Harrington.

In the study,⁵ patients treated with nivolumab had delayed worsening of functioning and symptoms compared with IC at approximately 4 months of follow-up, with patients receiving nivolumab reporting longer maintenance of function and less pain, fatigue, and dyspnea on treatment, as compared with those receiving IC.

Future of Immunotherapy in HNC

“So not only do we have clear evidence that these drugs can work in terms of improving survival and delivering meaningful responses, but they do so with fewer episodes of treatment-related toxicity and disease-associated morbidity,” said Harrington.The panelists noted that use of biomarkers and combination therapies are key areas of future development for HNC. Both areas are already being examined in clinical trials and have relevance across the vast HNC spectrum, from those with minimal disease to those with previously treated advanced disease, potentially offering a curative pathway to more patients.

Biomarkers

“It’s fantastic to have drugs that work in second-line relapsed metastatic or first-line metastatic setting, but what I want and what patients want is to be cured at the time they first present with their disease so they never have treatment in relapsed metastatic setting,” said Harrington.Although biomarkers such as PD-L1 expression are already being used and can help identify patients who are more likely to respond, high PD-L1 expression does not guarantee response, nor does no or low PD-L1 expression ensure lack of response or lack of durable response.

Subsequently, use of pembrolizumab and nivolumab is without use of this biomarker for patient selection. “While [a PD-L1 assay] can help inform patients of their likelihood to respond, it is not an assay that can select patients,” said Seiwert, who is working to identify novel biomarkers.

“I’ve been involved in looking at a novel biomarker called interferon gamma signature, which can be assayed quite easily with a rapid turn-around, and seems to perform somewhat better than PD-L1 expression,” said Seiwert. “It seems to have a high negative predictive value, and it may eventually allow us to exclude some patients who have no chance of having benefit, but it needs further validation.” He said that other biomarkers are also under investigation, including mutational load, immunogenic mutations, and dynamic biomarkers, but all are still experimental.

Combinations

“What we really need is a biomarker that would predict progressive disease,” said Grünwald. “To me, that would be much more usable than an assay that just allows us to say to patients ‘your chance of response is 30%.’ I see biomarkers as having the potential to guide development of treatment algorithms,” he said.Currently, PD-L1—targeted agents have seen the greatest development, and studies are starting to suggest that response with these agents can be enhanced when they are combined with other treatments, including chemotherapy, CTLA-4 blockade, and radiotherapy. “About 70% of HNCs have some level of PD-L1 expression—some level of inflammation—but we only see responses in 15% to 18% of patients, so the pool of patients who might benefit from combinations is huge,” said Seiwert.

He noted that the melanoma and lung cancer settings have already shown combining PD-1 inhibitors with chemotherapy or a second checkpoint inhibitor to be particularly promising in the front line, and he suspects one or both combinations will eventually receive approval in these settings and warrant serious investigation for patients with SCCHN.

“Some of our patients do not benefit from a checkpoint inhibitor, and we can’t identify these patients in advance, but giving them chemotherapy might buy us time,” he said. “It’s almost like a pharmacodynamic effect, where we have more time for the immunotherapy to work, and maybe, also make the immune system stronger and expose antigens.”

In the locally advanced setting, animal studies have shown promise combining chemoradiotherapy with immunotherapy, but results of a small study presented at ESMO revealed some dosing challenges in humans.⁶

In the study, 18 patients with various forms of intermediate- or high-risk SCCHN received ipilimumab, an anti—CTLA-4 antibody, in addition to standard intensity-modulated radiotherapy with cetuximab. Dermatologic immune-related adverse events limited dosing. “There are some safety hints that it may not be a piece of cake getting through radiotherapy, and maybe cetuximab might not be the optimal part now, but I think there is still a lot of promise combining radiochemotherapy with immunotherapy,” said Grünwald. “It could be a future way in how we successfully treat early forms of localized SCCHN.”

Combining checkpoint inhibition with radiation is another intriguing combination, and one that has the potential to act like an in situ vaccination that can lead to abscopal responses (ie, responses at distant sites), noted Harrington, something he has, thus far, only observed rarely with radiation.

“The idea behind combining checkpoint inhibition and radiation is that we could use the confluence of the two different mechanisms to make abscopal responses more predictable and more effective at a distant site, while engendering an immunologically relevant response that allows us to treat macroscopic metastatic disease while also getting rid of micrometastatic disease that could lead to metastatic failure,” he said.

Although immunotherapy combinations are showing promise in SCCHN and other cancers, the panelists warned that they should only be attempted as part of clinical trials. “There are still a lot of question marks about combinations, so they must be done as part of a clinical trial,” said Seiwert. Not only are toxicities and immunosuppressive effects best managed in clinical trials, but trials are essential in advancing these therapies and identifying the next breakthroughs in the field, he said.

References

1. US Food and Drug Administration. Pembrolizumab (KEYTRUDA). http://www.fda.gov/Drugs/InformationOnDrugs/ApprovedDrugs/ucm515627.htm. Updated August 9, 2016. Accessed December 4, 2016.
2. Chow LQM, Haddad R, Gupta S, et al. Antitumor activity of pembrolizumab in biomarker-unselected patients with recurrent and/or metastatic head and neck squamous cell carcinoma: results from the phase Ib KEYNOTE-012 expansion cohort. J Clin Oncol. 2016;34(32):3838-3845. doi:10.1200/JCO.2016.68.1478.
3. Bauml J, Seiwert TY, P ster DG, et al. Preliminary results from KEYNOTE-055: pembrolizumab after platinum and cetuximab failure in head and neck squamous cell carcinoma (HNSCC) [ASCO 6011]. J Clin Oncol. 2016;34(28). Abstract 6011.
4. Ferris RL, Blumenschein GR, Fayette J, et al. Further evaluations of nivolumab (nivo) versus investigator’s choice (IC) chemotherapy for recurrent or metastatic (R/M) squamous cell carcinoma of the head and neck (SCCHN): CheckMate 141 [ASCO 6009]. J Clin Oncol. 2016;34(38). Abstract 6009.
5. Harrington K, Ferris RL, Shaw J, et al. Patient-reported outcomes (PROs) in recurrent or metastatic (R/M) squamous cell carcinoma of the head and neck (SCCHN) treated with nivolumab (Nivo) or investigator’s choice (IC): CheckMate 141. Presented at: European Society for Medical Oncology 2016 Congress; October 7-11, 2016; Copenhagen, Denmark. Abstract LBA4.
6. Bauman J, Ferris RL, Clump DA, et al. Phase I trial of cetuximab, intensity modulated radiotherapy (IMRT), and ipilimumab in previously untreated, locally advanced head and neck squamous cell carcinoma. Presented at: European Society for Medical Oncology 2016 Congress; October 7-11, 2016; Copenhagen, Denmark. Abstract LBA34.