NEW YORK – There are many ways to divide head and neck cancer patients into subgroups. Some patients have tumors that can be removed surgically; some have tumors characterized by risky strains of the human papillomavirus (HPV); some have cancers that reappear or metastasize; and other have cancers that stay put.
But, as immunotherapy increasingly makes its way into experimental and established treatment regimens for head and neck squamous cell carcinoma (HNSCC), investigators are hunting for the tumor and immune features that will predict patients' treatment responses and, ultimately, survival.
Building on results reported in metastatic forms of the disease, a Sanford Cancer Center-led team recently published data from a Phase IB safety and efficacy trial of pembrolizumab (Merck's Keytruda) in combination with chemotherapy and radiation in 59 individuals with locally advanced HNSCC, demonstrating that the addition of checkpoint blockade immunotherapy did not dramatically bump up adverse events or hinder treatment in this population.
"You can add this molecule during treatment and activate the immune system, and it doesn't seem to add any significant toxicity to our standard treatment," explained Steven Powell, a medical oncologist and hematologist with the Sanford Cancer Center in Sioux Falls, South Dakota, and first author on a Journal of Clinical Oncology paper on the trial that came out this spring.
Though five of the 59 patients experienced immune-related side effects that prompted them to stop taking pembrolizumab, all but one patient completed a full course of radiotherapy. "Radiation is probably the most important part of this treatment. Radiation provides the most benefit," said Powell, who specializes in head and neck cancer, as well as lung cancer and melanoma. "The most important part of our study is that we didn't have any patients with significant treatment delays, and all but one patient completed all their planned doses of radiation, which is very good."
With data demonstrating the safety of the regimen, the researchers have now turned to genomic techniques to interrogate tumor and immune tissue samples from patients in the Phase IB trial, looking for clues to who might respond best to treatment. The researchers are employing a variety of techniques, from exome and RNA sequencing to search for tumor neoantigens and immune signatures of response, to conducting T-cell receptor sequencing to look at tumor-immune interactions.
The genomic analyses are supported, in part, by a five-year National Institutes of Health grant valued at nearly $11.7 million that was awarded in 2016 to support the head and neck cancer research and other cancer research at Sanford.
"That work is ongoing in my lab right now," Powell said. "We're looking at the genetics of the T-cell receptors to see if there are clues or different mutations … or different signatures that may lead to a better response to this treatment."
The data will become increasingly informative over the coming years, as they continue to collect clinical data from individuals receiving the pembrolizumab-chemoradiation combo, he added, noting that samples are also available from HNSCC patients treated with conventional chemoradiation approaches for comparison.
As those results roll in, the team hopes to build on known prognostic patterns in head and neck cancer. In particular, patients with HPV-positive tumors tend to fare better than their HPV-free counterparts.
The JCO analysis — though small and not designed to compared the HPV-positive and -negative subgroups — recapitulated this pattern: the team reported an 85 percent complete response rate at the end of treatment for the 34 patients in the HPV-positive subgroup, while the end-of-treatment response rate came in at 78.3 percent for the 25 patients with HPV-negative HNSCC patients.
"When we designed the study, the first step was to include anybody who needed chemotherapy and radiation and add in pembrolizumab," Powell said. "But as we got through it, we decided to separate it and to focus on a separate group of HPV-related cancers and an HPV-negative."
Broadly speaking, he noted, the cure rate for difficult-to-resect HPV-negative head and neck cancers is close to 50 percent, but can vary from 70 percent to 90 percent or higher for those with HPV-positive tumors, depending on whether cases are deemed low- or intermediate-risk based on other factors such as smoking history, tumor size, or lymph node involvement.
Given the good outcomes for low-risk HPV-positive head and neck cancer cases, much of the drug development research focus has been on the HPV-negative cases and the intermediate-risk HPV-positive cases, including efforts to push potential immunotherapy approaches into earlier treatment stages.
But there is still a need to improve treatment of HPV-positive head and neck cancer. Despite the survival advantage associated with that subgroup, patients may be diagnosed too late for a surgeon to resect the tumor.
"Unfortunately, many patients, especially patients with HPV-associated disease, present with cancers that are too advanced to remove surgically," Powell said, explaining that there are instances where it is physically possible to remove a tumor, but not without significant cost to an individual's quality of life.
Though treatment options and opportunities for surgical interventions are improving, oncologists sometimes turn to chemotherapy and radiation to treat certain tumors in the oropharynx, pharynx, and larynx to avoid leaving patients reliant on an electrolarynx or having to remove much of their tongue.
In that chemoradiation setting there appears to be a strong rationale for reaching for immunotherapy to help unleash immune activity against tumors being blasted with chemotherapy and radiation, since each treatment seems to boost the expression of PD-L1, one of the checkpoint proteins that can hold back immune responses to tumors. "There's some thought that perhaps that's a mechanism that tumors exploit to try to escape immune clearance," Powell said. "By adding a checkpoint inhibitor during treatment to block that, it may allow the immune system to clear tumors."
Waiting on Phase III data
Powell and colleagues designed their Phase IB trial with a focus on the pembrolizumab-chemotherapy-radiation regimen in locally advanced HNSCC, drawing on lessons from metastatic disease. The US Food and Drug Administration last June approved the pembrolizumab plus chemotherapy and radiation combination as a first-line option for metastatic or recurrent and unresectable forms of HNSCC. The agency based its decision on interim data from the KEYNOTE-048 Phase III trial presented at the American Society of Clinical Oncology annual meeting in 2019 and in a paper in The Lancet later that year.
For that trial, nearly 900 metastatic HNSCC patients were randomized to single-agent pembrolizumab, pembrolizumab plus chemo, or an immunotherapy-free arm with the EGFR inhibitor cetuximab or platinum chemotherapy plus fluorouracil. In the pembrolizumab plus chemo arm, the investigators saw a median overall survival time that stretched out to 13 months compared to a median overall survival time of 10.7 months in the cetuximab-chemo group.
Notably, a KEYNOTE-048 sub-group analysis suggested response to single-agent anti-PD-1 immunotherapy was more pronounced in the metastatic HNSCC patients with higher-than-usual tumor expression of PD-L1.
Powell recalled attending a conference with his colleagues where this data was being presented and that the idea struck him to look at whether this regimen would work in locally advanced disease. "We actually saw this data and immediately started thinking, 'This is great that this is going to help for recurrent, metastatic disease, but how do we get this into the setting to cure more patients?' That's how this evolved," he said.
As Powell and his colleagues now search for predictive response markers among participants of the Phase IB trial, other teams have pushed forward with a series of related clinical trials, including the ongoing KEYNOTE-412 trial — a randomized Phase III trial pitting a pembrolizumab-chemotherapy-radiation combination treatment (followed by maintenance pembrolizumab for close to a year) against chemotherapy and radiation in nearly 800 participants with locally-advanced HNSCC. A Future Oncology paper published this spring outlined the rationale for that trial, along with details of the design.
If the results of this study favor immunotherapy plus radiation, it may shift the standard of care for locally advanced HNSCC, according to Powell, who is anxiously awaiting the data. That trial will likely include some molecular and clinical subgroup analyses, though KEYNOTE-412 investigator Lillian Siu emphasized that such comparisons would more likely be within the already distinct HPV-positive and HPV-negative subgroups, rather than between them, owing to the survival boost already linked to HPV positivity.
"Regardless of whether you're giving them immunotherapy, chemotherapy, whatever it is, [HPV-positive patients] are going to do better," said Siu, a senior scientist, head and neck cancer specialist, and clinical trial specialist at the Princess Margaret Cancer Centre in Toronto.
Broadly speaking, there are distinct biological reasons why tumors from both the HPV-positive and HPV-negative groups may respond to immunotherapy — a topic that Siu and her colleagues explored in a biomarker review published in the Annals of Oncology last year.
The HPV-negative tumors are more apt to turn up in former smokers, for example, who may have a high tumor mutational burden and neoantigen load, while the HPV-positive tumors are, by definition, subject to alterations stemming from a risky HPV subtype. Given these factors, Siu doesn't believe HPV status alone should be used as a predictor of immunotherapy response.
She cautioned that there are typically relatively few actionable mutations in head and neck cancers compared to other cancer types. Still, Siu and the other KEYNOTE-412 investigators are hoping that more predictive markers will turn up in their study and other ongoing trials as the field tries to figure out precisely how to give immunotherapy in locally advanced HNSCC.
A Phase III trial, the JAVELIN Head and Neck 100 trial, which was comparing the anti-PD-L1 immunotherapy avelumab (Pfizer/Merck KGaA's EMD Serono's Bavencio) plus standard chemoradiation to a placebo plus chemoradiation in locally advanced squamous cell carcinoma of the head and neck, did not meet its primary endpoint of progression-free survival and was terminated in March.
Meanwhile, investigators behind the KEYNOTE-689 trial are focused on the possibility of using checkpoint blockade immunotherapy in the neoadjuvant setting for head and neck cancer patients who are eligible for surgery.
"There are a few ways people are looking at it," Siu said. "In the metastatic setting, immunotherapy is credentialed, it's approved. In the locally advanced setting, we're still trying to figure out how best to give it and how best to schedule it, and what drugs [to use]."