The Potential Impact of Prophylactic HPV Vaccination on Oropharynx Cancer

Abstract

Oropharyngeal cancer (OPC) is significantly increasing in incidence in the United States. Given that these epidemiologic trends are driven by HPV, the potential impact of prophylactic HPV vaccines on the prevention of OPC is of interest. To date, the primary evidence supporting the approval of current prophylactic HPV vaccines are large phase III clinical trials focused on prevention of genital disease (cervical and anal cancer, as well as genital warts). These trials reported 89-98% vaccine efficacy for prevention of both premalignant lesions and persistent genital infection. However, these trials were designed before the etiologic relationship between HPV and oropharyngeal cancer was established. There are differences in the epidemiology of oral and genital HPV infection, such as differences in age and gender distributions, which suggest that the vaccine efficacy shown in genital cancers may not be directly translatable to the oropharynx. Evaluation of vaccine efficacy is challenging in the oropharynx because no premalignant lesions analogous to cervical intraepithelial neoplasia (CIN) in cervical cancer has been identified. In order to truly investigate the efficacy of these vaccines in the oropharynx, additional clinical trials with feasible endpoints are needed.

The incidence of head and neck squamous cell carcinoma (HNSCC), an entity historically caused by tobacco and alcohol exposure, has decreased over the past 30 years.¹ However, the incidence of oropharyngeal cancer (OPC), a subset of HNSCCs, has risen significantly.^1-3 This dramatic rise in OPC is driven by human papillomavirus (HPV). Approximately 70-90% of newly diagnosed OPCs in the United States are HPV-related.⁴ These HPV-related cancers are often seen in younger, healthier patients, many of whom are non-smokers. In the face of this changing epidemiology, there is increasing interest in the potential impact of HPV vaccines on the prevention of OPC. To date, the three available prophylactic HPV vaccines have been rigorously evaluated only in the context of anogenital HPV infection, premalignancy and malignancy. Therefore, the vaccines are FDA-approved for prevention of anogenital cancer and warts, but not for oropharyngeal cancer. The current body of evidence is insufficient to determine the efficacy of these vaccines in the context of oral HPV infection or oropharyngeal cancer. This review provides an overview of the epidemiology of oral HPV infection, potential impact of the vaccine, and discusses the limitations to our current understanding of oral HPV infection and oropharyngeal cancer in the context of vaccines.

Epidemiology of oral HPV infection

HPV infection, a sexually transmitted virus, is the primary cause of cervical, anal and a growing subset of oropharyngeal cancers. HPV is the most common sexually transmitted infection,⁵ and a majority of the population will have evidence of at least one infection in a lifetime.⁶ Oral HPV infection is associated with increased lifetime sexual exposure to the virus, which can be measured by increased number of lifetime oral or vaginal sexual partners.⁷ Based on prevalence data representative of the United States population, 6.9% of individuals have an oral HPV infection of any type detectable in the oral cavity or oropharynx.⁷ HPV16, which is responsible for the overwhelming majority of OPCs, is only found in 1% of individuals in the United States.⁷

Demographic and behavioral factors associated with increased prevalence of infection, albeit at one time point, provide a glimpse into potential risk factors for incidence and duration, the determinants of prevalent infection. Oral HPV infection is 2.3 times more common in men compared to women, after adjusting for other risk factors including age, race, smoking and lifetime sexual partners. These gender differences could be attributed to a lower antibody response to HPV in men,⁸ or higher transmissibility of infection through oral sex performed on a woman, which is suggested by studies that show higher oral infection rates in heterosexual men compared to homosexual men.^9,10

In multiple studies, a bimodal age distribution of oral HPV has been noted with the first peak at ages 30-34 years and a second peak at 60-64 years of age (Figure 1a).^3,7 Explanations for the first peak include a surge in oral HPV infection after sexual debut and at peak sexual activity. The second peak may be attributed to potential reactivation of dormant infections with immunosenescence, increased HPV exposure in divorced or widowed populations, and/or increased rates of persistent infection which increase overall prevalence.^11-13

Figure 1. HPV infection prevalence in oral and cervical samples by age.

Prevalence of oral HPV infection using NHANES 2009-2012 on men and women ages 14-69.⁷ Cervical prevalence rates are adapted from adjusted NHANES data from women ages 14-59 in 2003-2004.¹⁹

In addition to age and gender, other risk factors associated with oral HPV infection are smoking, heavy alcohol use and sexual behaviors.⁷ Oral HPV prevalence is 2-3 times higher in current smokers,⁷ with a significant dose-response relationship.¹⁴ There is also a dose-response relationship between increasing number of lifetime sexual partners (oral and vaginal) and oral HPV infection in adjusted models.⁷ Other risk factors include immunosuppression (such as HIV infection), single status, and potentially deep kissing. Consistent with prevalence data, risk factors (in limited natural history studies) associated with incident infections include smoking, single status,¹³ heterosexual orientation in males,¹⁰ performing oral sex, and HIV infection.¹⁵

Unfortunately natural history data which would elucidate factors affecting acquisition, clearance and/or persistence of oral HPV infection, and virally- mediated transformation of normal epithelium to pre-cancer are presently limited. Evidence to date suggests that most people who do acquire oral HPV infection generally clear the infection within 6-12 months,^15,16 and 80% of individuals who acquire an oral HPV infection clear the infection by 1 year, regardless of HPV type-specific infection.¹⁵ However, some infections persist, for reasons yet to be identified, which may lead to higher risk for oncologic disease.¹⁵ Male gender, older age and current smoking have been found to be significantly associated with higher risk of HPV persistence.¹⁶ Other factors that may be associated with persistent infection include high oral HPV viral load¹⁷ and concurrent persistent cervical infection, although longterm natural history data are not available.¹⁸

In comparison, the prevalence of any cervical HPV infection in United States women is significantly higher (26.8% for ages 14-59), and 15.2% of women have high-risk cervical HPV infections, the most common being HPV16.¹⁹ The peak prevalence of cervical HPV infection is 44.8% in women ages 20-24, a few years following sexual debut. The age distribution does show a small peak in older populations,²⁰ again around ages 60-65, but this is significantly less prominent than that observed in oral infections (Figure 1B).²¹ Median clearance time of cervical infections is about 9-12 months, with 90-95% of infections being cleared by 2 years.²¹ In cervical infection, HPV16 is associated with increased persistence,²² however, age, smoking and HIV infection do not appear to affect persistence.²³

Oral HPV infection and head and neck cancer

The relationship between HPV and the upper aerodigestive tract was first appreciated in juvenile respiratory papillomatosis, a disease entity in which infants are exposed to female genital infections during birth.^24,25 Subsequent investigations revealed presence of HPV DNA through PCR analysis in head and neck tumors,^26,27 especially those arising from Waldeyer’s ring.²⁴ The presence of a known oncogenic virus within these tumors suggested an etiologic connection, however epidemiologic evidence of causation was not available until 2000-2001.^28,29 In 2001 a large nested case-control study in the Netherlands suggested that HPV-related tumors may be distinct from tobacco-related tumors. Serum samples available from a mean of 9.4 years prior to diagnosis of head and neck cancer were compared to matched controls and HPV16 seropositivity was two-fold higher in HNSCC patients.²⁹ In subgroup analysis by anatomic site, HPV seropositivity was most strongly associated with increased odds of oropharynx cancer (OR 14.4, 95%CI 3.6 to 58.1). This study also showed that HPV16 seropositivity to L1 and L2 detected 10-15 years prior to cancer diagnosis could be associated with disease risk. Another large nested case-control study suggested that HPV-related tumors may be distinct from tobacco-associated disease, with fewer TP53 mutations and improved disease-specific survival.²⁸ Most recently, genomic analysis of prospectively collected oral rinse samples collected in a nested case control study also showed a significant association between presence of oral HPV16 and incident cases of oropharyngeal cancer (OR, 22.4; 95% CI, 1.8-276.7) at a median of 3.9 years follow up.³⁰

Today, it is established that HPV is etiologically responsible for a subset of HNSCC tumors.³¹ In oropharyngeal cancers, HPV 16 is the most common high-risk infection. HPV 16 accounts for 90% of HPV-related OPCs (Figure 2A). Other HPV type-specific infections found in OPCs include HPV31, 33 and 35. These tumors exhibit distinct clinical characteristics with improved prognosis compared to HPV-unrelated tumors,^32,33 and a unique genetic profile with lower mutational burden.³⁴ However, the progression from persistent infection to invasive carcinoma remains unknown for oropharyngeal cancer, including identification of a suitable precursor lesion. HPV infection has been reported with some premalignant lesions of the oral cavity^35,36 and oropharynx.³⁶ However, reliable detection of these premalignant lesions poses a significant challenge,³⁷ perhaps because most HPV-related tumors arise from tonsillar crypts rather than surface epithelium.³⁸

Figure 2. Distribution of type-specific HPV infections found in cervical cancer and oropharyngeal cancers.

Oropharyngeal cancer data is pooled from studies with type-specific HPV detection performed.^2,33 For cervical cancer, data is adapted from Munoz et al.⁴⁰ for tumors with a single HPV infection.

In cervical cancer, several studies have established persistent HPV infection as a necessary step to development of cervical intraepithelial neoplasia (CIN), a precursor lesion to cervical cancer.³⁹ In contrast to OPC, HPV 16 only represents 60% of type-specific infections in cervical cancer, with HPV18 being the next most common type specific infection (Figure 2B).⁴⁰ While cervical cancer has the benefit of large natural history studies with long-term follow up, retrospective studies have also been used to demonstrate that inadequately treated cases of advanced CIN (CIN3) progress to invasive carcinoma in 30-50% of cases by 30 years, a majority occurring within 10 years.⁴¹

Current prophylactic HPV vaccines

Currently there are three FDA-approved prophylactic HPV vaccines (Table 1). In 2006, Gardasil (Merck) became the first prophylactic vaccine to become FDA-approved for females aged 9-26 for prevention of cervical cancer and genital warts. Later, approval was expanded to males for indications of genital warts (2009) and anal malignancies (2010). Cervarix (GlaxoSmithKline) approval followed in 2009 for females. Most recently, Gardasil-9 was approved in 2014 for both males and females. All three vaccines are approved for a dosing schedule of 3 doses over 6 months, although recent trials suggest potential evidence of non-inferiority for shorter dose schedules, which is the subject of new trials under development.^42,43

Table 1.

FDA-approved HPV vaccines.

Vaccine	HPV subtypes covered	Female	Male
Cervarix (bivalent)	16, 18	9-25	-
Gardasil (quadrivalent)	6, 11, 16, 18	9-26	9-26
Gardasil-9 (nine-valent)	6, 11, 16, 18, 31, 33, 45, 52, 58	9-26	9-15

All vaccines currently utilize virus-like particles of the HPV L1 protein. While Cervarix is only designed to protect against HPV subtypes 16 and 18, there is some evidence of cross-reactivity to other high-risk types.⁴⁴

Biologic efficacy of vaccines in cervical infection and cancer

The biologic efficacy of HPV vaccines in prevention of cervical infection and cancer was established through large blinded randomized phase III clinical trials performed in women ages 15-26 (Table 2). FUTURE I/II studies evaluated the quadrivalent Gardasil ,⁴⁵ while the PATRICIA⁴⁶ and Costa Rica HPV Vaccine Trial (CVT)^47,48 evaluated bivalent Cervarix .

Table 2. Summary of HPV vaccine clinical trials performed in cervical cancer populations.

Vaccine efficacy is summarized for prevention of persistent infection and disease (CIN2+) for each study. Efficacy data shown is limited to participants who received all 3 scheduled doses and showed no evidence of HPV exposure prior to vaccination.

Trial	Cohort	Vaccine	Control	Follow-up	Persistent viral infection efficacy (≥ 6 months)	Disease efficacy (CIN2+)
FUTURE I/II	Women 15-26	Gardasil (quad) N=7864	Placebo N=7865	4 years	Not evaluated	98.2%
PATRICIA	Women 15-25	Cervarix N=7338	Hepatitis A vaccine N=7305	4 years	91.4-94.3%	92.9%
CVT	Women 18-25	Cervarix N=2643	Hepatitis A vaccine N=2697	4 years	90.2-93.1%	89.5%
9-Valent trial	Women 16-26	Gardasil (nine) N=5948	Gardasil (quad) N=5943	4 years	Risk reduction 96.0*	Risk reduction 96.3-96.7*

^*For the 9-valent trial, risk reduction for disease or infection associated with HPV subtypes 31,33,45,52,58 (types added to 9-valent vaccine) is shown rather than absolute efficacy.^49,51,62

The design of a primary endpoint for these studies posed a challenge due to both practical and ethical barriers for evaluating the main endpoint of interest: prevention of invasive cervical cancer.⁴⁹ First, after cervical HPV infection, development of invasive carcinoma is a relatively rare event which may take a decade or longer to develop. Thus design of a clinical trial to evaluate this endpoint requires vast enrollment, long follow-up period and significant cost. Furthermore, because women enrolled in these clinical trials would have regular follow-up, identification of premalignant lesions (CIN) would require intervention before progression to invasive carcinoma.

Given these concerns, regulatory agencies, including the FDA advisory committee, recommended that prevention of premalignant lesions of CIN II or higher (CIN2+) could be used as a sufficient surrogate primary endpoint to evaluate vaccine efficacy in preventing cancer.⁵⁰ In addition, the CVT study also investigated vaccine efficacy for prevention of persistent viral infection (at 6 months and 12 months) as a surrogate endpoint.

These studies uniformly established high efficacy of the vaccine (89.5-98.2%) in prevention of premalignant cervical lesions (CIN2+) associated with cervical HPV type-specific infections covered by the vaccine for women who were previously unexposed to cervical HPV (Table 2). Efficacy decreased in all studies when women who had previous exposure to HPV infection (efficacy 54.8-60.7%) were included in analysis.⁴⁹ Additionally, vaccine efficacy was notably reduced when lesions associated with other HPV type-specific infections, not just those covered by the vaccine, were included (42.9-64.9%).⁴⁹ These data provided a rationale for the development of the 9-valent vaccine (Gardasil-9, Merck) which recently showed high efficacy in prevention of CIN2+ lesions and persistent infections HPV 31, 33, 45, 52, and 58 compared to the quadrivalent vaccine (Table 2).⁵¹

Follow-up studies from these trials have used immunogenicity to gauge antibody response and stability of immune response to the vaccine. Antibody titers one month after vaccine administration rise to levels 100 times higher than those after natural infection.⁴⁹ Following this peak, titers generally decline over two years before stabilizing, but remain at least 10 times higher than the natural infection level.⁴⁴ Both bivalent and quadrivalent vaccines have shown sustained immunogenicity for up to 8 years after initial vaccination.^43,52-54 When comparing both vaccines, Cervarix induces substantially higher titer responses, particularly in younger patients.⁴⁴ Although higher antibody responses would seem to be optimal, the minimum threshold required to achieve vaccine protection against genital infection is unknown.⁴⁹ The range of immunogenicity with known vaccine efficacy against both persistent infection and precancerous cervical disease has served as a basis for subsequent studies to establish non-inferiority for new vaccines,⁵¹ populations,^55-57 and dosing schedules.^42,43

Studies in other populations

Men

While most HPV vaccine studies were performed in women, studies in men may shed important insight as OPCs now account for 78.2% of newly diagnosed HPV-related tumors in men (14.4% anal, 7.4% penile).⁵⁸ Additionally, approximately 80-85% of OPCs are diagnosed in men.^3,32 In a randomized trial, the quadrivalent vaccine showed 90.4% vaccine efficacy for prevention of external genital lesions in men (including penile, perianal, or perineal intraepithelial neoplasia (PIN) or cancer) for covered HPV type-specific infections in HPV naïve subjects.⁵⁹ For persistent infection (≥6 months), the observed efficacy was 78.7% for HPV16 and 96% for HPV18.⁵⁹ Similar to studies in women, efficacy was lower when subjects with prior HPV exposure were included. A sub-analysis of men who have sex with men (MSM) determined vaccine efficacy for prevention of anal intraepithelial neoplasia (AIN) and anal infection to be 77.5% and 93.8-100%, respectively.⁶⁰ Results of these studies led to FDA approval of Gardasil for males.

Older women

Studies in older women may inform of age-related immunogenic changes after vaccination. The age of these study populations approaches the median age of oropharyngeal cancer diagnosis. One study evaluated Gardasil in women up to age 45 and found vaccine efficacy for persistent infection to be 91.8% for women ages 24-34, and 88.6% for women ages 35-45, without prior exposure to HPV.⁶¹ In immunogenicity studies for Cervarix, while antibody titers were slightly lower in women ages 46-55 compared to 15-25, the peak titers were still 57-84 times higher than those elicited by natural infection and plateau levels at two years remained 8-16 fold higher than natural infection levels.⁵⁴

Children

Immunogenicity studies have allowed for “immunobridging,” which utilize measurement of immune response to vaccines (through anti-HPV antibodies levels) to establish non-inferiority of vaccine efficacy.⁶² This allows for bridging of vaccines to populations for whom evaluating primary disease endpoints are not feasible, such as in children who are the target recipients of these vaccines (in order to provide the vaccine to a fully HPV naïve population). Ethical considerations are significantly greater when designing clinical trials that include minors, thus shorter trials with minimal intervention are desired. In addition, children are unlikely to acquire other primary end-points (i.e. HPV infection or precancerous lesions) until they reach adulthood. These studies established safety in pediatric populations and showed that immune responses were robust in boys and girls ages 10-15, with antibody titers 1.7-2.7 times higher when compared to young adults.^55-57 In addition, antibody titers remain stable for up to 8 years in pediatric populations.⁵³ As we think about establishing vaccine efficacy for oral HPV infection and oropharyngeal cancer, determination of immunogenicity levels that can protect against oral HPV infection may permit immunobridging and defining efficacy in pediatric populations.

What do we know about HPV vaccines in HNSCC?

Currently, few studies have been conducted to evaluate the impact of the HPV vaccine on oral HPV infection or oropharyngeal cancer. The convergence of our relatively new understanding of the etiologic connection between HPV and oropharyngeal cancer, the sample size considerations and consequent cost implications means that clinical trials were not previously designed for the study of oral HPV infection in the context of vaccines.

Additional data collected from the CVT study cohort has provided some indirect evidence for the potential efficacy of the vaccine in preventing oral infection.^63,64 At the 4-year follow-up after vaccination, oral rinses were obtained and showed the prevalence of oral HPV overall (0.7% vs. 1.3%) and of HPV 16 and/or 18 (0.03% vs. 0.5%) was lower in the vaccinated group compared to the control group, with an estimated vaccine efficacy of 93.3% for HPV 16/18.⁶³ Because the study was not originally designed to evaluate efficacy in oral HPV infection, no baseline oral HPV infection data was available. Furthermore, there were few prevalent or incident oral HPV infections in this cohort. The study was conducted only in women, while oral HPV infection and oropharynx cancer are more common in men.⁶⁵ Lastly, this study provided point prevalence data, but with only one time point was not able to evaluate vaccine efficacy for prevention of persistent infection, the recommended endpoint by the World Heath Organization International Agency for Research on Cancer (WHO IARC).⁶²

While limited epidemiologic data is available on vaccine efficacy for oral HPV infection, there is some promising biologic evidence. In animal studies, both vaccination and passive antibody transfer provided protection against oral infection and development of oral lesions from canine oral papillomavirus (COPV).⁶⁶ Efficacy of passive immunity shows the important role of IgG neutralizing antibodies in providing protection against oral infection,⁶⁷ and oral mucosal IgG is largely derived from the serum.⁶⁸ Oral IgG to HPV is detectable, though at a lower rate, in individuals who are seropositive for HPV^69,70 as well as vaccinated individuals.⁷¹ In addition to IgG, the other main mucosal antibody present in the oral cavity is secretory IgA (sIgA),⁷² and high prevalence of oral IgA has been observed in women with history of CIN.⁷³ Given robust serum immunogenic response to the vaccine in multiple populations,^55-57 presence of oral IgG and sIgA and subsequent protection against oral infection is likely. However, little data are currently available evaluating oral mucosal antibody response to vaccination and its correlation to vaccine efficacy.

Other studies have used predictive modeling to understand the potential benefit of HPV vaccination in the context of oropharyngeal cancer. One such modeling study showed that even with 50% vaccine uptake and 50% vaccine efficacy, vaccinating young boys specifically for the prevention of oropharyngeal cancer would be cost-effective.⁷⁴ Studies in cervical cancer project that given established trends and efficacy, significant decreases in incidence and mortality from cervical cancer would be observed by 2050.⁷⁵ Given the higher median age of diagnosis of oropharyngeal cancer compared to cervical cancer (58 vs. 48 years),⁷⁶ reduction in incidence and mortality rates for oropharyngeal cancer would not likely be observed until at least 2060.³ In addition to assumptions regarding vaccine efficacy and vaccine uptake, this projection also assumes sustained vaccine protection and no contribution of the second peak of oral HPV infection seen in older populations.³

What questions should we be asking next?

How much do the differences between oral and genital HPV infection matter in relation to vaccine efficacy? If a higher level of immunogenicity is required for protection against oral HPV infection, this may inform recommendations for dosing schedules, particularly in men. If the increased incidence of oral HPV infection at an older age is related to immunosenescence, would a vaccine booster be needed at older age? Should type-specific HPV infections unique to the OPC tumors be considered in future vaccine development?

Many questions remain, but the most important issue to address is how to investigate and establish efficacy of vaccines in the oropharynx. What is a feasible primary endpoint for such a study? How closely can we established the relationship between persistent infection and invasive cancer? Is clearance of persistent infection an adequate surrogate endpoint, given that no precursor lesion can currently be identified?

Primary endpoints for trial design in oral HPV

The WHO IARC and National Cancer Institutes (NCI) recently convened to discuss acceptable primary endpoints for future evaluation of prophylactic HPV vaccines.^62,77 A decade ago, CIN2+ was recommended as the primary endpoint to establish efficacy of disease prevention.⁵⁰ However, our body of knowledge has grown significantly since this time. Another challenge of trial design is that given the established efficacy of the vaccine, randomized placebo controlled trials can no longer be ethically performed.⁶²

The main primary endpoints for vaccine trial design are (1) disease endpoint (prevention of precancerous or invasive cancer), (2) viral endpoint (prevention of persistent infection), and (3) immunogenicity (comparison of immunologic response). Disease endpoints are the gold standard as the primary goal of HPV vaccine is not prevention of infection but prevention of disease. While the ideal trial design would evaluate vaccine efficacy for prevention of invasive cancer, the time between oral HPV infection and cancer progression is likelyprohibitively long. Unfortunately, no suitable premalignant lesion, such as CIN, which could serve as a surrogate disease endpoint is currently appreciated in the oropharynx.³ Even malignant tumors can be a diagnostic challenge, and HPV-related OPCs are often late stage with frequent nodal metastasis at the time of diagnosis.⁷⁸

A viral endpoint, i.e. prevention of persistent infection, may be ideal as it has several advantages. Persistent infection is a more common event and can objectively be determined within a shorter time frame; this allows for significantly smaller and shorter trials to establish efficacy.⁶² However, the validity of a viral endpoint is contingent upon a strong etiologic connection between persistent infection and disease, and therefore its ability to predict disease efficacy.

In cervical cancer, significant evidence collected through prospective natural history studies have strongly established persistent HPV infection as an essential step in the development of CIN3+.⁷⁹ In addition, vaccine efficacy for persistent infection was nearly equivalent to efficacy for prevention of disease in both PATRICIA and CVT trials (Table 2). In OPC, evidence establishing this etiologic connection between persistent infection and disease continues to grow, but definitive studies have been limited by lack of a suitable precursor lesion.⁷⁷ Nevertheless, WHO IARC recommended that prevention of persistent oral HPV 16 and/or 18 infection would be an acceptable surrogate endpoint for evaluation of vaccine efficacy.⁷⁷ Persistent oral HPV infection is currently being evaluated using oral rinses,⁶³ which have been optimized⁸⁰ and show high concordance with HPV infection detected by tonsillar brush biopsy.³⁷ In regard to immunogenicity endpoints, as discussed earlier, it is unknown what level of immunogenicity is sufficient for protection against oral HPV infection. Furthermore, would immunogenicity be best evaluated with serum or oral antibody measures? After vaccine efficacy against oral HPV infection has been established, immunobridging trials can be used to establish vaccine efficacy for oral infections and disease in pediatric populations, new vaccines or new dosing schedules as has been done in cervical cancer. In summary for the oropharynx, a viral primary endpoint represents the most feasible option for trial designs, and data on immunogenicity should be collected to inform future studies (Figure 3).

Figure 3. Schematic diagram to compare known and unknown vaccination related endpoints for the progression from HPV infection to malignancy in the cervix and oropharynx.

After vaccination, the first potential endpoint would be antibody response. Antibody titers necessary for prevention of cervical HPV infection have been established, but remain unknown in the oropharynx. In the cervix, peak age of HPV infection is earlier (20-24 years). The average age of CIN2+, the precancer disease surrogate endpoint used in vaccine trials in the cervix, is approximately 10 years later, at median age of 34 years.⁸⁶ The progression of CIN2+ to invasive cancer occurs in 30-50% of cases.⁴¹ Median age of invasive cervical carcinoma is 48 years,⁷⁶ with a latency of greater than 10 years from CIN2+.⁴¹ In comparison, oral HPV infection has a bimodal age distribution. The first peak is 30-34 years of age and the second peak is 60-64 years of age. While persistent oral HPV infection has not been shown to be a viral endpoint predating oropharyngeal cancer, it is expected to have analogous role as in the cervical HPV to cervical cancer progression model. No precursor lesion has been identified as of yet, but the median age of oropharynx cancer is later than cervical cancer at 58 years.⁷⁶ Based on our current knowledge, potential vaccination related endpoints for oral HPV infection and oropharyngeal cancer include antibody response (immunogenicity endpoint), persistent HPV infection (viral endpoint), precancer (disease surrogate endpoint) and invasive cancer (disease endpoint). Unknowns are depicted in grey color.

6) Conclusions – what does the landscape look like ahead?

As of 2014, it is estimated that 21% of eligible teenage boys and 40% of girls have received three doses of the HPV vaccine.⁸¹ While HPV vaccine knowledge and acceptance continues to grow, uptake still significantly lags behind the goal of 80% vaccine uptake by 2020 set by the Department of Health and Human Services Health People 2020 Objectives.⁸² In other countries, vaccine uptake for at least two doses has been able to achieve approximately 70% in school-aged girls in both the UK⁸³ and Australia.⁸⁴ Improved public knowledge on the etiologic connection between HPV and OPCs may increase vaccine uptake in men, as a key barrier in HPV vaccination for males is a “lack of perceived benefit” by both parents⁸² and healthcare providers.⁸⁵ Even if vaccine efficacy against oral infection or vaccine uptake in men is minimal, herd immunity is likely to play a role in decreasing the overall prevalence of HPV in the general population, though these effects are difficult to estimate.⁷⁵ In this changing landscape, there are new challenges in evaluating vaccine efficacy. Placebo-controlled trials can no longer be ethically performed, however, observed changes in current OPC incidence trends can provide indirect evidence for the potential impact of HPV vaccines in OPC.