Answers to ’critical questions’ may personalize treatment of HPV–associated head, neck cancer
Incidence of cancers that develop in the back of the throat, including the base of the tongue and tonsils, traditionally had been driven by tobacco and alcohol use.
Now, more than 70% of these cancers — known as oropharyngeal cancers (OPC) — are linked to HPV infection.
An estimated 16,400 new cases of HPV–associated OPCs are diagnosed in the United States each year, according to the CDC. Eighty percent of them occur in men.
“By the year 2020, it is projected that the oropharyngeal cases we see in this country will [surpass] cervical cancer as the number-one HPV–related cancer in the United States,” Nabil F. Saba, MD, FACP, director of the head and neck medical oncology program and professor in the department of hematology and oncology at Emory University, told HemOnc Today. “This has been deemed an epidemic and is a problem of magnificent proportions.”
Patients with HPV–positive OPC tend to have better outcomes than those with HPV–negative disease. Because of the favorable prognosis, researchers have focused on finding effective treatments that spare patients from long-term toxicity.
The ability to tailor treatments through de-escalation — as well as an emphasis on education about the importance of HPV vaccination — present opportunities to control the epidemic.
HemOnc Today spoke with medical oncologists, cancer researchers, and head and neck cancer specialists about the need for HPV vaccine uptake, how de-escalation of chemotherapy or radiation therapy may offer more personalized treatment, and the need to study and treat HPV–related OPC differently than other head and neck cancers.
Learning from the past
HPV became the leading risk factor for OPC following a sharp decline in tobacco- and alcohol-related head and neck cancers over the last 25 years, according to Anthony J. Cmelak, MD, professor of radiation oncology at Vanderbilt University School of Medicine and senior director of radiation oncology satellites at Vanderbilt-Ingram Cancer Center.
“We’ve known the infection can cause cervical cancer in women, but only over the last 2 decades have we recognized that HPV type 16 and 18 are the high-risk strands and are now the primary drivers for oropharyngeal cancers,” Cmelak told HemOnc Today.
Prevalence of head and neck cancers increased from 30,000 total cases in 1996 to 50,000 in 2016, Cmelak said. Prior to 1990, only 20% of cases were related to HPV. Between 1990 and 2000, however, HPV accounted for 50% of cases.
It is difficult to determine the primary cause of this trend because of the nature of how the virus develops and spreads, according to Krzysztof Misiukiewicz, MD, associate professor of medicine, hematology and medical oncology, and assistant professor of otolaryngology at Icahn School of Medicine at Mount Sinai.
“We know that it takes at least 10 years from exposure of the virus to develop cancer,” Misiukiewicz told HemOnc Today. “There are no proven reasons why this prevalence is occurring, but some ideas include migration — there is increased incidence of the infection in other areas of the world — lack of public awareness, and moving away from the traditional model of sexual partners.”
The latter includes an increase in the number of sexual partners.
“We are seeing individuals having more sexual partners at a younger age,” Cmelak said. “HPV–related head and neck cancer also is more common among younger men, who we have found to be a little more promiscuous.”
An analysis by Viens and colleagues from the CDC confirmed that, between 2008 and 2012, most cases of HPV–related OPC occurred among younger, nonsmoking Caucasian men.
Of the average 38,793 HPV–associated cancers diagnosed annually during that time — a rate of 11.7 per 100,000 people — most (n = 11,771) were cervical carcinomas. However, of the average 15,738 yearly cases of OPC, 12,638 occurred in men and 3,100 in women.
Most men with oropharyngeal tumors were aged 50 to 59 years (n = 4,627) or 60 to 69 years (n = 4,047).
“The epidemic of HPV–related head and neck cancer includes the baby boomer population,” Charu Aggarwal, MD, MPH, assistant professor of hematology oncology in Perelman School of Medicine at University of Pennsylvania, told HemOnc Today. “Many 50-year-olds have it, and it is directly related to oral sex. We may see this increase over the next 10 to 20 years in older populations.”
Women and nonwhite individuals also have been affected.
A retrospective cohort study by D’Souza and colleagues, published last year in JAMA Oncology, evaluated whether HPV caused a similar proportion of oropharyngeal cancers among women, Asians, Hispanics and blacks from 1995 to 2012 as it did among white men.
Results showed the proportion of p16–positive OPC increased significantly among women (29% to 77%; P = .005) and men (36% to 72%; P < .001), as well as among whites (39% to 86%; P < .001) and nonwhites (32% to 62%; P = .02).
Researchers found few nonoropharyngeal head and neck cancers were related to HPV; of 623 cases, only 62 were positive for p16.
Based on these findings, the increase in HPV–associated OPC “was not restricted to white men, but was a consistent trend for women and men, as well as for white and nonwhite racial groups,” researchers wrote.
Need for screening, staging methods
Despite greater prevalence, researchers still seek to define the best means for detecting HPV–positive OPC.
Unlike the pap smear for cervical cancer, no screening techniques exist for the disease.
“It makes perfect sense to find a way to screen but we have failed to do so, partly because the oropharynx is difficult to visualize and is not an easy area to inspect,” Saba said. “We haven’t identified a clearly visible lesion that could be a precursor to HPV–related cancer.”
The p16 immunohistochemistry (IHC) test is the most common tool used to detect HPV genotypes due to its high sensitivity.
“p16 is a surrogate marker that tells us if there is a high chance that the cancer is going to be HPV positive,” Misiukiewicz said. “We then conduct HPV testing via polymerase chain reaction, which tests not only for the presence or absence of the virus, but which subset you have.”
Other techniques include in situ hybridization for high-risk strains, and quantitative polymerase chain reaction for viral E6 mRNA and DNA.
In a pooled analysis, Prigge and colleagues showed the combination of p16 IHC and HPV DNA polymerase chain reaction testing enhanced specificity and maintained high sensitivity for the diagnosis of HPV–induced OPC.
A study by Qian and colleagues evaluated the use of p16 and beta-catenin by IHC among 101 OPC tissues, and HPV status by HPV DNA in situ hybridization. The combination of the three biomarkers was comparable in sensitivity to p16 alone (98.8% vs. 100%), but appeared more specific (88.9% vs. 81%).
Researchers also have evaluated a much simpler method to detect oral HPV — a saliva test.
In a study published last year in JAMA Oncology, Agalliu and colleagues found patients with HPV16 in saliva samples were 22 times more likely to develop OPC (OR = 22.4; 95% CI, 1.8-276.7) and 7.1 times more likely to develop any head and neck cancer (OR = 7.1; 95% CI, 2.2-22.6) than individuals without oral HPV16.
“These findings inform our thinking about the etiology of oropharyngeal cancer — specifically, oral HPV preceding its diagnosis,” Aimée R. Kreimer, PhD, senior investigator in the NCI’s division of cancer epidemiology and genetics, said in a press release. “However, it would be premature to consider the detection of oral HPV as a screening test.”
This method may help to not only predict disease risk, but also determine treatment intensity in already-diagnosed patients, Cmelak said.
“There are some preliminary data from forthcoming studies that show detectable HPV DNA in saliva is a useful tool for determining cure or persistence of disease that hasn’t been eradicated with treatment,” he said. “This may be a helpful marker for us to consider adjuvant treatment for harder-to-treat patients who still harbor HPV DNA after standard treatment.”
Although researchers seek to determine the presence of HPV in head and neck cancers to better characterize a patient’s disease, HPV is not included in head and neck cancer staging systems, and no tumor registries exist that routinely record HPV status in these patients.
Further, research from Zanotti and colleagues — presented at this year’s ASCO Annual Meeting — showed many physicians do not use HPV testing to help guide treatment. The results — based on 2016 survey data from 152 physicians in the United States, France, Germany and the United Kingdom — showed 54% of physicians were unaware of the type of test used to determine HPV status.
Of 2,193 patient cases, only 42% underwent HPV testing.
These results reflect a core challenge for head and neck cancer clinicians: There are no treatment guidelines for this specific population of patients, who often receive the same treatments as patients without HPV.
“There are a lot of treatments available for cancer in general, and we apply those principles broadly to HPV–related cancer,” Aggarwal said. “We know that, by virtue of having HPV–related disease, patients will have a better prognosis. What we don’t know however, is how to treat these patients differently.”
Thus, it is important to study the HPV–related head and neck cancer population separately, according to Cristina Rodriguez, MD, medical oncologist at Seattle Care Cancer Alliance, associate professor in the department of medicine at University of Washington, and HemOnc Today Editorial Board member.
“There have been very large clinical trials that have been completed or are ongoing trying to answer critical questions about how to treat these patients,” Rodriguez said in an interview. “With the completion of these trials, we will begin to get some answers about how we can best treat this population.”
In 5 to 10 years, HPV–related OPC likely will be treated quite differently than non-HPV–related disease, Aggarwal said.
“In the future, we won’t be just staging them differently, but we will be thinking about treating them differently,” she said.
Guidelines will be critical to ensure the best outcomes for these patients.
“We have to come up with standards so everyone will be following them and doing the same thing,” Misiukiewicz said.
Treatment guidelines for HPV–related head and neck cancer will be based on one key distinction of this disease: These patients tend to respond better to chemotherapy and radiation therapy, leading to better prognosis.
This is particularly true among nonsmokers, who survive 20% to 30% longer than individuals with a smoking-related cancer, Cmelak said.
Based on this knowledge, researchers are developing trials to test treatments that will maximize outcomes and minimize toxicity through de-escalation, also called deintensification.
Because HPV–positive patients tend to live longer, they could experience detrimental effects from therapies for years. For example, radiation can cause severe swallowing impairment, aspiration, feeding-tube dependence, stricture, xerostomia and hypothyroidism. Concurrent cisplatin also increases acute toxicity and late noncancer mortality.
Long-term results of the RTOG 91-11 trial — conducted in patients with stage III or stage IV glottic and supraglottic squamous cell cancer — showed death not attributed to cancer occurred more frequently among patients treated with concomitant chemotherapy (30.8%) than induction chemotherapy (20.8%) or radiation alone (16.9%).
“Patients at 10 years of follow-up were starting to die from noncancer-related issues more frequently if they had cisplatin, so the survival advantage at 10 years was beginning to wash away,” Cmelak said. “Now, we are trying to get away from the platinum-based standard.”
There are many layers to de-escalation, and it is not simply a removal of chemotherapy, Saba said.
“The focus of deintensification is relying on a certain type of treatment,” he said. “The challenge is to figure out which way will be less toxic for patients and produce fewer side effects.”
De-escalation options being explored include reducing radiation dose, therefore relying on chemotherapy; changing or reducing the chemotherapy agent; giving chemotherapy alone; and relying on new systemic agents.
“The right way to do de-escalation at this point is in a clinical trial setting so we can contribute to our understanding of this disease,” Rodriguez said.
Ongoing clinical trials investigating de-escalation for HPV–positive OPC include:
- RTOG 1016, which includes more than 800 patients randomly assigned to cetuximab– or cisplatin–radiotherapy;
- NRG HN002, a randomized phase 2 trial designed to compare dose-reduced cisplatin chemoradiotherapy with accelerated radiotherapy alone;
- the De-ESCALaTE trial in the United Kingdom, which is assessing morbidity and quality of life among patients randomly assigned to standard cisplatin plus radiotherapy vs. cetuximab plus radiotherapy; and
- ECOG 3311, a phase 2 trial investigating transoral surgery followed by adjuvant therapy.
It may be another 10 years before these trials establish the best modality of deintensification, Saba said.
“It will take lots of effort and many more clinical trials before we get to that stage of clearly knowing what would be acceptable deintensification without putting patients at higher risk for recurrence,” he said.
In addition to identifying the best means of de-escalation, determining which patients are best suited for this approach remains a challenge.
In the phase 2 ECOG 1308 trial, Marur and colleagues evaluated whether complete clinical response to induction chemotherapy could identify patients for whom a reduced radiation dose may be appropriate.
In total, 56 of 80 enrolled patients achieved a primary-site complete clinical response to chemotherapy, 51 of whom went on to receive cetuximab (Erbitux, Eli Lilly) — an anti-EGFR antibody — with 54 Gy intensity-modulated radiation therapy. Among those patients, 80% achieved 2-year PFS and 94% achieved 2-year OS.
At 12 months, a smaller proportion of patients who received the lower radiation dose had impaired nutrition (10% vs. 44%; P = .025) or difficulty swallowing solids (40% vs. 89%; P = .011).
These outcomes showed “patients can get less treatment and still have good outcomes,” Misiukiewicz said.
Better identifying patients appropriate for de-escalation “is the most critical question,” Rodriguez said.
“The current treatments are not easy and result in acute toxicities,” she added. “If we can spare a certain population from that, it would not only contribute to better outcomes, but would be more valuable for cost-effective treatment.”
Rationale for immunotherapy
Because existing treatments are so toxic, another research goal focuses on the development of new treatments with fewer toxic effects.
“The radiation and chemotherapy treatments can cause swallowing difficulties and taste problems, exactly the reasons why we need to come up with better treatments that are less toxic,” Aggarwal said.
Researchers have looked to immunotherapies to fulfill this need.
“Some immunotherapies are showing good response rates in patients with recurrent or metastatic disease, and there are a percentage of patients who have prolonged response, sometimes lasting years,” Cmelak said.
Pembrolizumab (Keytruda, Merck) — an anti–PD-1 antibody — has demonstrated durable antitumor activity in patients with recurrent metastatic head and neck cancer resistant to platinum and cetuximab therapy.
Bauml and colleagues conducted a phase 2 trial in which 171 patients received 200 mg pembrolizumab every 3 weeks. Sixteen percent of patients responded, with an 8-month median response duration. Response rates did not differ between patients with or without HPV.
In the CheckMate 141 trial — which included patients with recurrent or metastatic head and neck cancer — those who received the anti–PD-1 antibody nivolumab (Opdivo, Bristol-Myers Squibb) achieved longer median OS (7.5 months vs. 5.1 months; HR = 0.7; 95% CI, 0.51-0.96) and a higher rate of 1-year OS (36% vs. 16.6%) than those who received standard single-agent systemic therapy.
“In Checkmate 141, initial results suggested HPV–positive patients had greater benefit, but when we looked closer, the study shows nivolumab to be effective regardless of HPV status,” Saba said. “The challenge for the future is how best to understand how immunotherapy should be used in HPV–positive vs. HPV–negative disease.”
The better toxicity profile has prompted researchers to evaluate immunotherapies in the first-line setting.
“There are discussions about how to incorporate them into standard regimens or even substitute these for chemotherapy,” Cmelak said.
Ongoing trials that seek to better define the use of immunotherapy in head and neck cancer include CheckMate 714, in which researchers will evaluate the addition of ipilimumab (Yervoy, Bristol-Myers Squibb) to nivolumab; NCT03085719, which will investigate the safety and durability of high- and low-dose radiation with pembrolizumab; and NCT02002182, which will investigate whether ADXS11-001 (Advaxis) — an experimental vaccine — stimulates the body’s defense system against HPV–positive OPC before transoral surgery.
Aggarwal and colleagues also developed INO-3112 (Inovio Pharmaceuticals) — a therapeutic DNA vaccine that consists of HPV16/18 plasmids combined with interleukin (IL-12) — specifically for HPV–positive disease.
Researchers presented results from 22 patients at this year’s ASCO Annual Meeting. Findings showed the vaccine increased CD8–positive infiltration into tumor in two patients (1.6- to 3.6-fold) and decreased FOXP3–positivity in three patients (1.8 to 2.1-fold). Eighteen patients demonstrated high HPV16/HPV18–specific T-cell activity.
“We were able to see generation of HPV–directed immune responses in both blood and tissue, which is very novel,” Aggarwal said. “The next step for that trial is to develop the vaccine in combination with a PD-L1 inhibitor for metastatic disease.”
It is exciting to see immunotherapies have activity in head and neck cancers, Rodriguez said.
“The current wave of clinical trials is closely examining how we can integrate these immunotherapy agents into current standards for therapy,” she said. “Over time, we will understand what benefits these add to our current treatments.”
Lack of vaccine uptake
The increasing prevalence of HPV–related OPC — and the questions about how to treat it — would be less of a concern with more widespread use of HPV vaccination.
Routine HPV immunization has been recommended for girls since 2006, and for boys since 2011. However, vaccination rates remain low.
A report from the National Vaccine Advisory Committee — published in 2016 in Public Health Report — showed initiation rates for the HPV vaccine series were 57.3% for girls and 34.6% for boys. Only 40% of girls and 15% of boys completed the recommended three-dose series.
“Uptake has increased over the past 10 years, but it is still not at [the Healthy People 2020 target of] 80% series completion for males and females,” Electra D. Paskett, PhD, Marion N. Rowley professor of cancer research, and director of the division of cancer prevention and control in the College of Medicine at The Ohio State University, told HemOnc Today.
In October 2016, the CDC and Advisory Committee on Immunization Practices recommended a two-dose vaccine series when initiated at age younger than 15 years. The groups made the recommendation after clinical trial data showed adolescents aged 9 to 14 years experienced similar or higher immune response to the vaccine after two doses than young adults who received three doses.
“The reduction to two shots now for those under 15 years of age possibly will help increase the completion rates,” Paskett said.
The FDA has approved three vaccines to prevent HPV infection: Gardasil (Merck), Gardasil 9 (Merck) and Cervarix (GlaxoSmithKline).
A study by Gillison and colleagues — presented at the ASCO Annual Meeting — showed men and women aged 18 to 33 years who received at least one dose of the HPV vaccine were less likely than unvaccinated individuals to develop oral HPV16/18/6/11 infections (0.11% vs. 1.61%; P = .008). The vaccine reduced prevalence by 17% overall, by 25% among women and by 6.9% among men.
“The vaccine has not yet been shown to be fully effective to reduce incidence of oral cancer, but the study suggests the vaccine is reducing the oral HPV infection, which is likely to reduce the incidence of oral cancer,” Saba said.
Despite the clear link to reduced cancer incidence, lack of vaccination may be linked to poor access.
“Only 18% of the population are getting the vaccine, so access is clearly limited,” Cmelak said, adding that the vaccine is covered by most insurance plans but could be expensive otherwise.
Strides have been made to promote vaccination, but barriers remain. They include concerns about the vaccine’s effect on sexual behavior, low perceived risk for HPV infection and irregular preventive care.
Research shows some parents decide not to vaccinate their sons because of a perceived lack of direct benefit.
Physicians should remember to strongly suggest the HPV vaccine to males because the HPV epidemic is not “just a female problem,” Paskett said.
“This picture needs to change to prevent future cancer cases among men and women,” she said. “We have a vaccine to prevent some cancers. Why aren’t people lining up to get it? It’s what everyone wants, but now people won’t get it.”
Raising awareness about vaccination should be a priority for clinicians, Misiukiewicz said.
“We are hoping HPV vaccination will decrease HPV cancer in general — not just orally, but cervically, too,” he said. “Vaccination will be the main contributor to what happens in the future.”
Without a good screening tool, “the vaccine seems to be the most logical and effective way to reduce the disease,” Saba said.
“As leading cancer centers, we need to work with others to promote vaccinations,” he said. “We need concerted efforts to try to get to where we want to be as far as HPV vaccination.”
A change to public health policy could lead to an uptake in vaccination rates, Rodriguez said.
“Uptake rates in Australia and Canada are superior to those in the United States,” she said. “It is because of the way our system is set up. We rely on a parent having medical insurance and taking their children to be vaccinated, whereas other countries mandate the vaccine and have programs that make it easier for the vaccine to reach the population that needs it.” – by Melinda Stevens
Agalliu I, et al. JAMA Oncol. 2016;doi:10.1001/jamaoncol.2015.5504.
Bauml J, et al. J Clin Oncol. 2017;doi:10.1200/JCO.2016.70.1524.
Bhatia A and Burtness B. J Clin Oncol. 2015;doi:10.1200/JCO.2015.61.2358.
D’Souza G, et al. JAMA Oncol. 2016;doi:10.1001/jamaoncol.2016.3067.
Ferris RL, et al. N Engl J Med. 2016;doi:10.1056/NEJMoa1602252.
Forastiere AA, et al. J Clin Oncol. 2013;doi:10.1200/JCO.2012.43.6097.
Holman DM, et al. JAMA Pediatr. 2014;doi:10.1001/jamapediatrics.2013.2752.
Marur S, et al. J Clin Oncol. 2016;doi:10.1200/JCO.2016.68.3300.
Prigge ES, et al. Int J Cancer. 2017;doi:10.1002/ijc.30516.
Qian G, et al. J Oral Pathol Med. 2016;doi:10.1111/jop.12378.
National Vaccine Advisory Committee. Public Health Rep. 2015;131:17-25.
Viens LJ, et al. MMWR Morb Mortal Wkly Rep. 2016;doi:10.15585/mmwr.mm6526a1.
The following were presented at ASCO Annual Meeting; June 2-6, 2017; Chicago:
Aggarwal C, et al. Abstract 6073.
Alsidawi S, et al. Abstract 6048.
Bhide S, et al. Abstract 6043.
Gillison ML, et al. Abstract 6003.
Zanotti G, et al. Abstract e17526.
For more information:
Charu Aggarwal, MD, MPH, can be reached at firstname.lastname@example.org.
Anthony J. Cmelak, MD, can be reached at email@example.com.
Krzysztof Misiukiewicz, MD, can be reached at firstname.lastname@example.org.
Electra D. Paskett, PhD, can be reached at email@example.com.
Cristina Rodriguez, MD, can be reached at firstname.lastname@example.org.
Nabil F. Saba, MD, FACP, can be reached at email@example.com.
Disclosure: Paskett reports institutional research funding from Merck. Rodriguez reports research funding from Bristol-Myers Squibb, Ignyta and Merck. Saba reports honoraria from Bristol-Myers Squibb, Eli Lilly, Merck and Pfizer, as well as a data and safety monitoring board role for a study sponsored by Pfizer. Aggarwal, Cmelak and Misiukiewicz report no relevant financial disclosures.