Immunotherapy Outcomes in HPV-Associated Head and Neck Cancer Fuel Investigative Efforts

Partner | Cancer Centers | <b>UPMC Hillman Cancer Center</b>

Robert L. Ferris, MD, PhD, discusses the current standard of care for patients with head and neck cancer, the role of HPV in treatment decisions, and the need to personalize therapy for patients.

Promising data from studies evaluating immunotherapy and cellular therapy for patients with head and neck cancers associated with human papilloma virus (HPV) have helped to widen the treatment landscape. However, many clinical questions remain unanswered as to the appropriate sequencing for the trials. Additionally, investigators have looked to the success in this patient population to develop a model for other HPV-associated cancers, according to Robert L. Ferris, MD, PhD.

Key clinical insights derived from pivotal trials have provided a foundation of information regarding the biology of HPV-positive cancers. For example, Ferris referenced data from the from the phase 3 CheckMate 141 (NCT02105636).

CheckMate 141 was designed to evaluate the efficacy of nivolumab (Opdivo) in patients with head and neck cancer. Investigators observed an overall survival (OS) which was approximately the same for patients who with HPV-positive disease and HPV-negative disease. However, the responses were deeper and more quickly achieved in patients who were HPV-positive.

Investigators enrolled patients with squamous cell carcinoma of the head and neck were randomized to nivolumab (n = 240) or investigators choice of therapy (n = 121). Among those who received nivolumab, 64 patients had HPV-positive disease and 56 had HPV-negative disease. In the investigator’s choice arm 29 and 37 patients had HPV-positive and HPV-negative disease, respectively.

At 2-year follow-up, the median OS was 9.1 months (95% CI, 6.5-11.8) with nivolumab vs 4.4 months (95% CI, 3.3-9.8) with investigator’s choice in the HPV-positive population (HR, 0.60; 95% CI, 0.37-0.97). In the HPV-negative population the median OS was 7.7 months (95% CI, 4.8-13.0) with nivolumab vs 6.5 months (95% CI, 3.9-8.7) with investigator’s choice (HR, 0.59; 95% CI, 0.38-0.92).1

Although a benefit was observed with the PD-L1 inhibitor, Ferris noted that efforts either through combination therapy or proper sequencing among this population are needed to maximize success.

“If the PD-1 pathway is not the primary suppressive mechanism in [a particular patient], we need [identify other ways] to personalize treatment,” Ferris said in an interview with OncLive® ahead of the 6th Annual International Congress on Immunotherapies in Cancer™: Focus on Practice-Changing Application. “There are individuals who are cured, remarkably, by PD-1 monotherapy; [however,] tends to be a minority [of approximately] 10% to 15% of patients. For the [remaining patients in] the vast majority, what combination do they need? How can we rationally design those combinations?”

Ferris, director of University of Pittsburgh Medical Center (UPMC) Hillman Cancer Center, professor of oncology, otolaryngology, immunology, and radiation oncology, discussed the current standard of care for patients with head and neck cancer, the role of HPV in treatment decisions, and the need to personalize therapy for patients.

OncLive®: Please expand on the incidence of HPV-related malignancies.

Ferris: We know HPV can cause cancers in the anal/genital area, the cervix, the anus, the vulva, as well as in the head and neck, particularly in the tonsil or the base of the tongue.

HPV-related [head and neck] cancer has been rising dramatically. Most people think of HPV-related cancers as cervical cancer, but a few years ago, HPV-associated head and neck cancer in the oropharynx overtook cervical cancer in [the highest] number of cases occurring per year in North America. [Therefore], the head neck subsite of HPV-related cancers is a [significant] clinical problem that happens to be my area of specialty as a head neck surgeon.

[As a] cancer immunologist and an HPV biologist, [my focus is on] immune response.[Specifically,] I focused heavily on head and neck cancer given that incidence has been rising at a rate of 5% per year for a few decades. It has now gone up 300%, so I thought that would be interesting as a model for HPV-related cancers.

[There is a] role for immunotherapy, not only in head and neck cancer specific to anti-HPV immunity, but also in cervical cancer, [such as] anti­­–PD-1 immunotherapy, anti-CTLA4, and combinations which are used as immuno-oncology agents. [These therapies] for HPV-related head and neck [and] cervical cancer, are particularly useful in the preoperative neoadjuvant window—4 to 6 weeks before surgery where 1 can get a biopsy because of the accessibility through natural orifices.

We can get to the cervix, [or] the anal cancer, or tonsil and pharynx cancer through the mouth [to] get baseline tissue before immunotherapy, then put the patient on a 4-to-6-week treatment course. At the time of surgical resection, [we are] able to get posttreatment tissue to examine the tumor microenvironment after a 4-to-6-week window. Then, we can separate patients by response and [determine whether a] patient is a responder or a nonresponder, [which is] the key. Because HPV-related cancers occur in these natural orifice locations, we can look at whether HPV-specific immune responses are altered [in] a responder or a nonresponder.

Although patients with HPV-associated cancers respond to immunotherapy, we don’t know if that’s because of an antiviral response or some other antigen that the immune system is reacting to in the tumor. Sometimes there are DNA alterations separate from the virus, which is a nonself antigen. [Our bodies] respond to viruses all the time as we’ve learned through the pandemic.

Cellular immunotherapies, using tumor infiltrating lymphocytes that you can isolate from a resected cancer specimen [are another promising therapy]. For example, T-cell receptor transduction, where an HPV-specific T-cell receptor [is added] into the cellular therapeutic, [which is] expanded outside of the body in 1 of our GMP clean rooms. [Next, with] an FDA approved regimen, you take a million cells and expand them into a billion, all of them expressing an HPV-specific T-cell receptor, and then transfer those cells back as a living therapy.

[The engineered cells are similar to chimeric antigen receptor (CAR)] T cells, but instead of the CAR, it’s a T-cell receptor specific for the HPV antigen. Typically [it is] 1 of the oncogenes, E6 or E7. [These cells] cross different tumor locations because they target HPV wherever it may occur. [Data from] clinical trials at the [National Cancer Institute] and elsewhere are showing good responses in patients who are refractory to anti–PD-1 immunotherapies.

Regarding head and neck cancer, how has treatment evolved for HPV-related disease?

HPV-related head and neck cancers back in the 1970s or 1980s [accounted for approximately] 15% to 20% [of all head and neck cancers], but now it’s [about] 50%. Now, when we treat head and neck cancers, approximately half are HPV-related and half are HPV-unrelated and caused by tobacco carcinogen exposure.

There is a subset of [patients who fall into] an intermediate-risk group of HPV-related head and neck cancers [who] tend to do better [with] conventional therapies, [such as] chemotherapy, radiation and surgery. Yet, some of the patients with HPV-related head neck cancer also smoke, so I don’t want to distinguish [patients as] universally smoking or HPV, [because] there are some intermediate-risk factor populations.

Nonetheless, what we’ve learned with immunotherapy is that OS tends to be about the same for [patients who are] HPV-positive and HPV-negative, but the response is deeper and quicker [in patients who are] HPV-positive. [This is] probably because [they] have viral nonself antigens that the immune system has generated a T-cell response to, and then the tumor has escaped, whether that’s through the PD-1 pathway or some other suppressive mechanism.

For patients with the HPV-unrelated head neck cancers, eventually [they] get the same overall survival with anti–PD-1, but it takes longer. In the phase 3 CheckMate 141 trial that I led, the curves separate for patients treated with PD-1 within a month or 2 in the HPV-positive patients, [but] it’s more like 6 to 8 months in the HPV-negative [population]. It takes a little time to generate a priming event for the immune system and generate anti-tumor activity against the more HPV-unrelated antigens, probably due to mutations and smoking exposure.

The response is less frequent, early on, and less rapid with HPV-negative [disease], but eventually they catch up. The overall survival in CheckMate 141 showed a hazard ratio of 0.59 [for patients with HPV-negative disease] vs 0.6 [for those with HPV-positive disease], so a 39% to 40% reduction in risk of death.1

What are some of the sequencing hurdles that exist in the treatment landscape for this patient population?

We now have FDA approval for anti–PD-1 monotherapy in second line, or with chemotherapy in first line for recurrent metastatic [head and neck cancer]. PD-1 therapy, anti–PD-1 antibodies, nivolumab, and pembrolizumab [Keytruda] are approved for [patients with] recurrent and metastatic disease. The next frontier is previously untreated, locally advanced disease, which is how [approximately] 75% of patients present.

This is the real frontier because these patients are curable. The question is, can we add immunotherapy on top of standard of care? [One standard of care is] surgery, usually with a transoral robotic procedure, which was approved by the FDA in 2009. [Now, we are asking] if we can add immunotherapy on top of standard postoperative radiation with or without chemotherapy. Or, with nonsurgical standard-of-care approach, can we add immunotherapy to high-dose cisplatin chemotherapy and radiation?

Unfortunately, results of the randomized phase 3 Javelin 100 trial [NCT02603432], which was presented at [the European Society for Medical Oncology Annual Congress] 2021, were negative when we added the antibody avelumab [Bavencio], a PD-L1 specific antibody, [to chemoradiation]. [The study was conducted in patients with] high-risk head and neck cancers that were locally advanced and previously untreated. This included those with both HPV-negative and HPV-positive disease, smokers, and [patients] with bulky tumor burden [who were considered] intermediate risk.

When avelumab was added to chemoradiation, the sequence was avelumab first, then concurrent [avelumab] during chemoradiation, [followed by] a year of maintenance or adjuvant avelumab. There was no survival benefit, either progression-free or overall survival. That’s unexplained, so we don’t know why. It could be biology, meaning anti–PDL-1 was negatively affecting the treatment or the chemoradiation was reducing the efficacy of the immunotherapy. There are some other developmental trials ongoing where we separate the chemoradiation from the immunotherapy to see if we can harness the benefits of both, which we’ve seen independently, but for some reason did not see additive benefit when they were combined.

That’s where the field is going. Additional trials in the coming year or 2 will report out to validate the use of pembrolizumab. [This includes data from] KEYNOTE-412 [NCT03040999], [looking at whether] there is benefit when you add concurrent immuno-oncology to chemoradiation. Other trials [with] chemoradiation first, and then separately sequentially give 1 year of immuno-oncology therapy. We anxiously await those results.

What are the current challenges and unmet needs for patients in this paradigm?

Like other tumor types, we’re gratified that immuno-oncology agents, particularly the PD-1 inhibitors and more recently PD-1 CTLA4, benefit a subset of patients, but we don’t know how to select those patients. We don’t know how to personalize immunotherapy. The preoperative neoadjuvant trials give us an early response assessment and the ability to identify the responders to monotherapy so that we can add other agents.

[Determining the proper sequence of] neoadjuvant therapies with multiple off-the-shelf agents that can be added on is where the field needs to go. It would be ideal to figure this out at baseline, but sometimes we have to treat with monotherapy and then have a biomarker signal that tells us to add in another combination targeted agent to maximize efficacy for this exciting new class and new modality of therapies.


  1. Ferris RL, Blumenschein Jr G, Fayette J, et al. Nivolumab vs investigator’s choice in recurrent or metastatic squamous cell carcinoma of the head and neck: 2-year long-term survival update of CheckMate 141 with analyses by tumor PD-L1 expression. Oral Oncol. 2018;81:45-51. doi:10.1016/j.oraloncology.2018.04.008