Prevention and screening of HPV malignancies.

Abstract

It is estimated that 5% of the global cancer burden, or approximately 690,000 cancer cases annually, is attributable to human papillomavirus (HPV). Primary prevention through prophylactic vaccination is the best option for reducing the burden of HPV-related cancers. Most high-income countries (HICs) have introduced the HPV vaccine and are routinely vaccinating adolescent boys and girls. Unfortunately, although they suffer the greatest morbidity and mortality due to HPV-related cancers, many lower- and middle-income countries (LMICs) have been unable to initiate and sustain vaccination programs. Secondary prevention in the form of screening has led to substantial declines in cervical cancer incidence in areas with established screening programs, but LMICs with absent or inadequate screening programs have high incidence rates. Meanwhile, HICs have seen incidence rates of anal and oropharyngeal cancers rise owing to the limited availability of organized screening for anal cancer and no validated screening options for oropharyngeal cancer. The implementation of screening programs for individuals at high risk of these cancers has the potential to reduce the burden of cervical cancer in LMICs, of anal and oropharyngeal cancers in HICs, and of anal cancer for highly selected HIV positive populations in LMICs. This review will discuss primary prevention of HPV-related cancers through vaccination and secondary prevention through screening of cervical, anal, and oropharyngeal cancers. Areas of concern and highlights of successes already achieved are included.

INTRODUCTION

Human papillomavirus (HPV) is etiologically linked to cancers of the cervix, vulva, vagina, anus, oropharynx, and penis, with 690,000 cancer cases, equivalent to 5% of the world’s cancer burden, attributable to HPV in 2018.¹ Worldwide, cervical cancer is the fourth most common cancer among women and accounts for 80% of all HPV-related cancers.¹ Most cervical cancer cases occur in low- and middle-income countries (LMICs), primarily because of lack of access to screening. In contrast, high-income countries (HICs) have experienced declining cervical cancer incidence rates along with a significant increase in oropharyngeal and anal cancers; consequently, these cancers now account for a greater proportion of HPV-associated cancers in HICs than cervical cancer.¹

Along with greater access to cervical cancer screening, prophylactic HPV vaccination in LMICs has the potential to substantially lower the HPV-related cancer burden worldwide. The World Health Organization (WHO) has called for the elimination of cervical cancer (≤4 per 100,000 women per year) as a public health problem.² To accomplish this by the end of this century, the WHO has identified three goals to be reached by 2030: 1) 90% of girls should be vaccinated by age 15 years; 2) 70% of women should be screened at least twice for cervical cancer by age 35–45 years; and 3) 90% of women with cervical disease should receive treatment.² In addition, most HICs have included boys and young adults in their HPV vaccine strategies to reduce transmission to females and to address rising rates of anal and oropharyngeal cancers, of which approximately 49% and 83%, respectively, occur in men.³ Expanding screening programs for anal cancer and developing screening paradigms for oropharyngeal cancer remain opportunities, particularly in HICs and within populations at higher risk. This review will discuss primary prevention of HPV-related cancers through vaccination and secondary prevention through screening of cervical, anal, and oropharyngeal cancers. Areas of concern and highlights of successes already achieved are included.

PREVENTION

Vaccination

Primary prevention of HPV-related cancers can be achieved through vaccination against HPV infection. Currently, three vaccines have been licensed by the Food and Drug Administration (FDA) and European Medicines Agency. A quadrivalent vaccine that targets HPV types 6, 11, 16, and 18 was approved by the FDA in 2006, followed by approval of a bivalent vaccine targeting HPV types 16 and 18 in 2009. In 2014, the FDA approved a nonavalent vaccine that targets an additional five oncogenic types (31, 33, 45, 52, and 58).⁴ Together, HPV types 16 and 18 are responsible for 71% of cervical cancers, while types 31, 33, 45, 52, and 58 are responsible for an additional 16%.⁵ Cancers of the oropharynx and anus are primarily caused by HPV16.⁶

High vaccine prices and supply chain problems have led to challenges for LMICS in the implementation of HPV vaccination programs. As of 2020, 85% of HICs had introduced the HPV vaccine, in contrast to only 30% of LMICs; however, it is expected that more LMICs will implement vaccination programs in the next few years as a result of lowered vaccine prices and support from the Global Alliance for Vaccines and Immunization.^7,8 In the United States, vaccination rates have remained relatively low compared to other HICs, especially those with school-based programs, many of which have achieved >70% coverage rates.^9–11 Barriers to vaccination in the United States are generally related to lack of knowledge of the vaccine, perceived lack of benefit, social influences, and safety concerns,¹² largely propagated online by anti-vaccine groups. Although this is of great concern, public education interventions delivered by authoritative sources targeted to parents and minority groups have been shown to increase vaccine uptake.¹³

Several large international randomized controlled trials have shown that the HPV vaccine has high efficacy for the prevention of HPV infection and of cervical intraepithelial neoplasia (CIN), the precursor to cervical cancer.^14–17 These trials have shown at least 90% efficacy for the prevention of high-grade CIN and up to 100% efficacy for the prevention of high-grade vaginal and vulvar lesions.^14–17 Additionally, vaccination has been shown to prevent vaccine type anal infections.^18,19 Although not designed to evaluate oral HPV infection, the Costa Rica Vaccine Trial estimated the efficacy of the bivalent vaccine against prevalent oral HPV infection to be 92%.²⁰

Pooled data from the CVT and the PATRICIA trial, both evaluating the bivalent vaccine, provided evidence of the non-inferiority of one- and two-dose schedules compared to the originally approved three-dose schedule against cervical HPV16/18 infections.²¹ A formal evaluation of the efficacy of one dose is currently being conducted.²²

Based on the evidence of non-inferiority of fewer doses, the WHO and the Centers for Disease Control and Prevention (CDC) recommendations were modified from three to two doses if vaccination is initiated prior to age 15, although the recommendations for three doses did not change for those 15 years or older or for immunocompromised individuals.²³ Superior immunogenicity in females younger than 15 years compared with those 15 and older was demonstrated in a multi-center randomized controlled trial conducted in Denmark, Finland, Greece, and Estonia.²⁴ Although all 773 females receiving the bivalent vaccine seroconverted to HPV16/18 positivity from a previously seronegative state, those aged 10–14 years had two-fold higher antibody titers than those aged 15–25 years, which likely resulted in a longer duration of antibody persistence.²⁴ In addition to being less likely to have had previous exposure to HPV, adolescents have a greater immune response before age 15, strengthening the argument for vaccination before the 15^th birthday. Nevertheless, high immunogenicity, efficacy, and safety has been shown in adults up to age 45 years,²⁵ and in 2019, the CDC recommendation was expanded to include catch-up vaccination of individuals aged 27 through 45 years who were not adequately vaccinated previously but who may derive a benefit from vaccination as determined through shared clinical decision-making with their provider.⁴

Data from randomized controlled trials and post-marketing surveillance have shown the HPV vaccines to be safe overall.^26,27 The most common adverse events have been injection-site pain, syncope, dizziness, headache, and fatigue, which all resolved spontaneously with no significant differences observed between vaccine and control groups.²⁶ Additionally, serious adverse events have been rare, with no differences observed between groups.²⁶ The Vaccine Adverse Event Reporting System (VAERS) was searched for adverse events related to the nonavalent vaccine and 7,244 reports were found for the period from 2014 to 2019, during which time approximately 28 million doses were administered. Only syncope and errors related to vaccine administration (e.g., wrong vaccination schedule for recipient’s age) were higher than would be expected, and overall, the safety profile was consistent with that of the quadrivalent vaccine.²⁷

Evidence demonstrates the HPV vaccine has been a public health success. In countries that have instituted vaccination programs, the prevalence of infection with the HPV vaccine types decreased by 83% in girls aged 13–19 years and by 66% in girls aged 20–24 years after 5 to 8 years of vaccine implementation.²⁸ Additionally, the incidence of CIN2+ decreased significantly among girls younger than 25 years and anogenital wart diagnoses decreased significantly in both girls and boys who were younger than 20 years. Changes in prevalence have been especially apparent in countries with high coverage and vaccination of multiple cohorts.²⁸ In the United States, the prevalence of oral vaccine-targeted HPV was 88% lower among vaccinated men and women aged 18–33 years compared with unvaccinated individuals.²⁹ Moreover, the prevalence of oral vaccine-targeted HPV in unvaccinated men aged 18–59 years decreased by 37% between 2009–2010 and 2015–2016, whereas the prevalence of HPV types not targeted by vaccines remained unchanged, suggesting herd immunity as a result of vaccination.³⁰ Lower oral and anal HPV prevalence has also been observed among vaccinated compared to unvaccinated men who have sex with men (MSM), demonstrating vaccine effectiveness rates of 59% among those age 18 years or younger and 18% among those older than 18 years.³¹

Modelling studies have shown that vaccine coverage as low as 30% results in herd immunity.³² In countries with girls-only vaccination programs, evidence of herd immunity for both girls and boys has been observed. For example, in Australia, where 17- to 19-year-old heterosexual boys had already shown low prevalence of vaccine types (2.6% in 2014–2015 in an unvaccinated cohort) as the result of a girls-only vaccine program initiated in 2007, the introduction of a gender-neutral school-based vaccine program in 2013 did not significantly decrease the prevalence further (0.7% in 2016–2017 in a cohort among which 55% had received ≥1 dose; p = 0.371), indicating a herd effect due to vaccination of girls only had already occurred.³³ Among both vaccinated and unvaccinated girls in the United States, the prevalence of vaccine-targeted HPV types has similarly decreased in both cohorts.³⁴

Following the call to action by the WHO, the feasibility of and best strategies for cervical cancer elimination have been investigated. In a modelling study conducted by Brisson et al. to determine the feasibility of cervical cancer elimination in LMICs, 60% of LMICs could eliminate cervical cancer (defined as ≤4 cases per 100,000 women-years) with girls-only vaccination at 90% coverage by the end of the century.²⁸ To achieve 100% elimination, twice-in-a-lifetime screening with up to 90% coverage by 2045 in the highest-incidence countries, in addition to 90% vaccination coverage, would be required because vaccination alone would not achieve elimination in countries with very high incidence (>25 cases per 100,000).³⁵

As a leading example of what is achievable, Australia began a school-based program in 2007 to vaccinate all females aged 12–13, with catch-up vaccination to age 26 years; males aged 12–13, with catch-up to age 15 years, were subsequently added in 2013.³⁶ By 2017, Australia had achieved 80% and 76% coverage of girls and boys by age 15 years, respectively.³⁷ A modelling study has projected that cervical cancer will be eliminated in Australia by 2028 (<4 cases per 100,000) and will be as low as <6 cases per 100,000 women by 2020.³⁶ Furthermore, an incidence rate of <1 case per 100,000 could be achieved by 2066 with screening for HPV every 5 years, assuming continued coverage with the nonavalent vaccine at current levels.³⁶ A substantial reduction in the burden of non-cervical HPV-related cancers as a result of HPV vaccination is also projected.³⁸

Other strategies

Although vaccination remains the most effective strategy against HPV infection and HPV-related disease, other methods show various levels of protection. Consistent condom use decreases the risk of sexually transmitted infections, including HPV. However, because HPV is transmitted via skin-to-skin contact and condoms do not cover the entire genital area, the data for HPV prevention have been less consistent.³⁹ Regardless, the National Cancer Institute has concluded that there is adequate evidence from case-control and cohort studies for the protective effect of condom use in reducing cervical cancer risk.⁴⁰

Male circumcision is known to reduce the risk of contracting HIV, but it also reduces the risk of HPV infection and penile cancer.^41,42 In addition to its direct benefit for males, circumcision may also prevent HPV transmission to sex partners, a finding confirmed in prospective clinical trials.^43,44

SCREENING

CERVICAL CANCER

Cervical cancer is particularly well-suited to screening because the period from initial infection with an oncogenic HPV type to the development of cancer spans many years, and invasive cancer is preceded by an identifiable precursor lesion.⁴⁵ CIN is categorized based on increasing degree of severity as CIN1, CIN2, and CIN3. Screening followed by appropriate management of cervical dysplasia has led to a dramatic reduction in the incidence of cervical cancer.⁴⁶ The majority of invasive cancers occur in women who have not been screened, and as a consequence of lack of access to screening or treatment of cervical dysplasia, LMICs are disproportionately affected by the cervical cancer burden.⁴⁷

Available screening tests

Pap test cytology.

Cytologic examination of exfoliated cervical cells is the primary method of cervical cancer screening and major reductions in cervical cancer morbidity and mortality in locations with well-organized, high-coverage screening programs have been observed.⁴⁸ Although cytologic testing has high specificity, the sensitivity is low. In a meta-analysis of randomized controlled trials conducted in Europe and North America the estimated average sensitivity of cytologic testing was 53% (range, 19% to 77%), with 96% specificity to detect CIN2 or worse (CIN2+).⁴⁹ The low sensitivity of cytologic testing can be overcome by more frequent testing, and testing every 1–3 years is typically recommended. While this is feasible in countries with adequate resources, frequent screening is challenging in low-resource settings.

Visual inspection.

Due to its simplicity, immediate results, and low cost, visual inspection of the cervix, using either acetic acid (VIA) or Lugol’s iodine (VILI), is an attractive point-of-care option for low-resource settings. A pooled analysis of 26 studies conducted in LMICs evaluated the accuracy of VIA and found a sensitivity of 80% and specificity of 92%.⁵⁰ However, VIA has a low positive predictive value (10%), especially among inexperienced providers.⁵¹ Visual inspection with Lugol’s iodine offers higher sensitivity and similar specificity as VIA (91% and 85%, respectively) but is more costly and is not as easily obtained.⁵² In low-resource settings where screening is limited, a one-time screen with VIA has been shown to reduce the incidence of cervical cancer by 25% and mortality by 35% within 6 years compared to unscreened women.⁵³ WHO has recommended a screen-and-treat strategy using VIA followed by treatment of screen-positive women in the absence of other options for cervical cancer screening in resource-limited settings.⁵⁴ As with cytologic testing, the main limitation of VIA/VILI is the high inter-rater variability, and accuracy is highly dependent on the training of providers.^51,55

HPV DNA.

Given that HPV is a necessary cause of cervical cancer, HPV testing is now recommended in the United States and some European countries as an alternative to cytologic testing for the primary screening of women aged 30 years or older.^56,57 Although cytologic testing is still the most widely used screening test, testing based on the detection of viral DNA or E6 and E7 mRNA for oncogenic HPV types is increasingly being incorporated into screening programs. Compared to cytologic testing, HPV testing offers increased sensitivity for the detection of high-grade cervical lesions or cancer with the limitation of lower specificity.⁵⁸ Ultimately, increased detection of preinvasive disease at the first screening round leads to prevention of invasive cancer through early intervention in preinvasive disease. In a pooled analysis of four randomized controlled trials in Europe comparing cytologic testing with HPV test-based screening, HPV testing resulted in the detection of more high-grade lesions and, ultimately, 60%–70% fewer invasive cervical cancers with the greatest number of cancers averted after 2.5 years or more of follow-up and among women with a negative screening test at entry.⁵⁹

Triage of screen-positive women

When HPV testing is used for primary screening in women 30 years of age or older, triage of screen-positive women is necessary to avoid overtreatment of lesions that would have regressed on their own. Triage of HPV+ women should consist of reflex cytology and/or HPV16/18 genotyping combined with referral for colposcopy for those with atypical squamous cells of uncertain significance (ASCUS) or worse on cytology or who are positive for HPV16/18.^56,60 Repeat HPV testing after one year for HPV+ women with normal cytology increases the sensitivity for detection of CIN3+ by 30% compared to cytology alone.⁶¹ Alternatively, molecular assays such as dual staining for p16-INK4A and Ki-67 increase the sensitivity of cytologic testing for HPV-screen-positive women.⁶² Women younger than 30 years are likely to have transient HPV infections, thus reducing the specificity of HPV testing and leading to over-referral for follow-up; therefore, HPV testing is utilized for women at least 30 years old.⁶³

Role of self-sampling in HPV testing

An advantage of HPV testing is that it allows for self-sampling. While a clinician-collected sample is required for cytologic testing, either a clinician- or patient-collected sample can be used for HPV testing. Self-sampling has repeatedly been found to have HPV test accuracy comparable to clinician-collected samples.⁶⁴ The opportunity to self-collect samples may allow more women to get regular screenings and has the potential to increase coverage in under-screened populations.⁶⁵

Current guidelines

Guidelines for cancer screening aim to maximize benefit while minimizing harm. For cervical cancer, this means detecting and treating high-grade cervical lesions to prevent progression to invasive cancer while avoiding treatment of lesions that would not progress to cancer even if left untreated.⁶⁶ Current guidelines for major organizations are detailed in Table 1. For women considered at low risk of cervical cancer, screening every 2 to 3 years with cytologic testing or every 5 years with HPV testing does not result in a higher cancer rate compared to annual testing, but it does reduce unnecessary colposcopy referrals and unnecessary treatment of lesions that would have regressed on their own. Two modeling studies found that annual screening did not significantly reduce mortality compared with screening every 2 or 3 years.^59,67 In contrast, HIV-positive women, considered at high risk of cervical cancer, require more frequent screening.⁶⁸

Table 1.

Cervical cancer screening guidelines of various organizations.

Organization	Age group	Screening test	Screening interval	Recommendation	Other considerations
World Health Organization 65	30–49	Cytology	3–5 years	Acceptable if HPV testing unavailable	Screening should not begin before age 30 years Screen at least once in lifetime Screen-and-treat approach recommended VIA may be used as an alternative in low-resource settings HIV-positive women should be screened as soon as possible after HIV status becomes known.
		HPV test	5 years	Preferred
European Guidelines for Quality Assurance in Cervical Cancer Screening 55,66	<30	Cytology	3–5 years	Cytology alone	Screening within an organized population-based program Screening should not begin before age 25 years Screening should be stopped at age 60–65 years No recommendation regarding primary HPV testing in women aged 30–34 years because of insufficient evidence for/against in this age group
	35–65	HPV test	5–10 years	HPV testing alone; cytology triage for HPV+ women
United States Preventive Services Task Force (USPSTF) 54	21–29	Cytology	3 years	HPV testing not recommended	Screening should stop at age 65 if adequate negative screening Women without a cervix and no history of high-grade cervical disease do not need to be screened More frequent screening for immunocompromised women or those who were exposed to diethylstilbestrol
	30–65	Cytology	3 years	Cytology alone
		HPV test	5 years	Alone or cotesting with cytology
American Cancer Society (endorsed by the National Comprehensive Cancer Network) 67	21–29	Cytology	3 years	HPV testing not recommended	Screening should be stopped at age 65 years for women with a history of negative screening Screening should continue for at least 20 years following CIN2+ diagnosis Women without a cervix and no history of CIN2+ should not be screened
	30–65	Cytology	3 years	Screening with cytology alone
		HPV test	5 years	HPV cotesting with cytology (preferred)

Effectiveness of cervical cancer screening

Observational studies have shown the effectiveness of cervical cancer screening programs. In a systematic review of cervical cancer screening programs in Europe, Jansen et al. found wide variation in mortality reduction in the 10 included studies from Northern and Western Europe.⁴⁸ Mortality was reduced 41%–92% among those attending organized screening compared to non-attendees and 17%–79% among women who were invited compared to those not receiving an invitation (opportunistic screening). In Finland, a 66% mortality reduction was observed in a case-control study of women aged 25–69 years that compared attendees to non-attendees.⁶⁹ In a cohort study from Denmark, mortality from cervical cancer was reduced by 87% among women aged 23–59 years attending with 3-year screening intervals compared to non-attendees.⁷⁰ Testing for oncogenic HPV types has also resulted in a reduced incidence of cervical cancer. A 40% reduction in cervical cancer incidence was observed in the arms that included HPV testing in a pooled analysis of four European randomized controlled trials.⁵⁹ An Indian randomized controlled trial found the incidence of stage II or greater cervical cancer and the mortality due to invasive cancer was reduced by 50% within 8 years among unscreened women who received one HPV test and with adequate follow-up of screen-positive women.⁶⁷ Screening intervals of more than 5 years for women with a negative HPV test have been shown to be safe.^71,72

The greatest declines in cervical cancer incidence have been observed in HICs, particularly those with organized screening programs.^73,74 Despite either stable or increasing cervical cancer risk in women born after 1940 or 1950 (likely due to changes in sexual behavior and increased exposure to HPV), age-standardized incidence rates decreased in HICs with effective screening but not in countries with no organized screening.⁷³ For example, Vaccarella et al. estimated that cervical cancer screening programs may have prevented more than 60,000 cases, equivalent to almost 50% of the expected cervical cancer cases, in the 50 years since screening began in Nordic countries (Denmark, Finland, Norway, and Sweden).⁷⁴ In the United States, it is estimated that screening with cytology has led to the prevention of almost 500,000 cases.⁷⁵

Even with vaccination, primary screening with HPV testing will be necessary because the positive predictive value of cytology will suffer owing to the lower prevalence of premalignant lesions.⁷⁶ Randomized controlled trials are ongoing to determine the optimal screening interval for women who have been vaccinated.⁷⁷ However, the lack of vaccine registries in some countries will necessitate the continued screening of women for whom vaccination status cannot be obtained.⁷⁸

ANAL CANCER

As the incidence of anal cancer continues to rise, screening for high-risk individuals is recommended, although no national consensus guidelines exist. The incidence of anal cancer varies by risk group with individuals living with HIV, MSM, women with a history of HPV+ gynecological precancer or cancer, solid organ transplant recipients, and individuals with autoimmune diseases being at highest risk.⁷⁹

Available screening tests

Anal cancer screening strategies are derived from those used for cervical cancer screening. The most common initial approach is anal cytologic testing with referral to high-resolution anoscopy (HRA) for those with a concerning finding on cytology. Unlike cervical cytologic screening, anal swabs are typically obtained without visual guidance.⁸⁰ The accuracy of liquid-based anal cytologic testing varies according to risk group. A meta-analysis that included 30 observational studies found 86% sensitivity and 52% specificity for identifying high-grade anal intraepithelial (AIN2+) lesions versus normal tissue among 4,074 patients screened for anal cancer.⁸¹ Sensitivity was 88% among HIV-positive patients and 91% among MSM.⁸¹ Among 278 women with a history of gynecological precancer or cancer, sensitivity was 71% with 73% specificity for detection of a high-grade squamous intraepithelial lesion (HSIL) or cancer; however, sensitivity improved to 92% among immunosuppressed women compared to 60% in non-immunocompromised women.⁸² Given this variability, most efforts target high-risk populations to increase test performance.^81,82

Adding HPV testing to anal cytology increases sensitivity but at the cost of decreased specificity.^83,84 Kimura et al. compared different strategies for detection of HSIL using HRA as the gold standard.⁸⁴ The sensitivity of cytology alone was 73%, which was increased to 85% when HPV testing was added. However, specificity was decreased from 73% to 43% with HPV testing. Therefore, the authors concluded that the optimal screening algorithm was anal cytologic testing followed by triage, with HPV testing for patients with normal cytology or ASCUS and subsequent referral to HRA for those with oncogenic HPV detected (80% sensitivity and 71% specificity).⁸⁴ In a study that included female attendees of the Anal Neoplasia Clinic of the University of Puerto Rico Comprehensive Cancer Center, cytologic and HPV testing combined had 100% sensitivity and 16% specificity compared with 85% and 39%, respectively, for cytology alone for detection of HSIL, as confirmed using HRA with biopsy.⁸³

Self-sampling for anal cancer screening may not be as effective as for cervical cancer. In a study comparing self- with clinician-collected samples among MSM with no prior experience collecting samples, the sensitivity for detection of HSIL was lower for self-collected samples (75% vs. 90% in the HIV-positive group and 48% vs. 62% in the HIV-negative group).⁸⁵ However, others have shown that among MSM who routinely received screening, the sensitivity of self-sampling was similar to clinician-collected samples.⁸⁶

Digital anorectal examination (DARE) is a simple procedure for the detection of anal abnormalities that is associated with minimal discomfort and generally acceptable to patients.⁸⁷ Additional studies are needed to determine the sensitivity and specificity of DARE, but in a phase II clinical study of MSM, self anal exam or partner anal exam after instruction by a clinician had a sensitivity and specificity of 75% and 94%, respectively, for detection of masses ≥3 mm in size.⁸⁸ Moreover, it was found to be acceptable to MSM and has the potential to be a feasible and cost-effective adjunct screening tool.^87–89

HRA is the gold standard for detection of anal dysplasia and cancer; it is performed in a manner similar to cervical colposcopy, with application of acetic acid or Lugol’s iodine to allow visualization of abnormal lesions. However, because it takes specialized training and is time-consuming, it is typically reserved for use following an abnormal cytology result.⁹⁰

High-risk groups

According to a recent meta-analysis, among HIV+ individuals, MSM are at particularly high risk, with an incidence of 85 per 100,000 person-years, followed by 32 per 100,000 for non-MSM males and 22 per 100,000 for females with HIV; among HIV- MSM, the incidence was 19 per 100,000 person-years.⁷⁹ For women with a history of gynecological cancer or precancer, the incidence ranged from 6 per 100,000 person-years for women with a history of cervical precancer to 48 per 100,000 person-years for women with a history of vulvar cancer.⁷⁹ For solid organ transplant patients, the incidence was 13 per 100,000 person-years, and for individuals with autoimmune disease, the incidence was highest among those with systemic lupus erythematosus at 10 per 100,000 person-years.⁷⁹ Higher incidence rates are also observed for HIV-positive non-MSM, HIV-positive women, HIV-negative MSM, and women with a history of vulvar or cervical cancer.⁷⁹

Although anal cancer incidence is increasing among women, focusing on those with a previous diagnosis of HPV+ gynecological precancer or cancer or those who are immunocompromised is likely to be the most cost-effective strategy. Studies of women with a previous diagnosis of gynecological precancer or cancer have demonstrated a high rate of anal abnormalities.^91,92 Among 75 women with HSIL or cancer of the cervix, vagina, or vulva, 16 (22%) were found to have abnormal anal cytology, with six (8%) ultimately diagnosed with anal HSIL on cytology or biopsy.⁹¹ HPV16/18 testing was positive in 15 women (20%), including three of the six (50%) diagnosed with HSIL.⁹³ In another study of 317 women aged ≥40 years with a history of HSIL or cancer of the cervix who underwent anal cytologic and HPV testing, 96 (30%) were found to have abnormal anal cytology.⁹² Thirty women were found to have an anal squamous intraepithelial lesion on HRA, with two-thirds being HSIL.⁹² Additionally, HIV-positive women have a greater than 10 times increased risk of anal cancer compared with HIV-negative women, and among these women anal cytologic and HPV testing both have high sensitivity (83% for cytology and 75%–77% for HPV testing, depending on test used, p>0.02) with modest specificity (67% vs. 50%, respectively).⁹⁴

Lack of consensus for screening recommendations

There is a lack of national consensus guidelines for anal cancer screening, although guidelines from various societies, groups, and institutes have been issued, as recently summarized in a systematic review, with most targeting HIV-positive individuals (the American Society of Colon and Rectal Surgeons, New York State Department of Health AIDS Institute, HIV Medicine Association of the Infectious Diseases Society of America, European AIDS Clinical Society, Northwest Pennsylvania Rural AIDS Alliance, Spanish AIDS Study Group/Grupo de Estudio de SIDA, and the German AIDS Society).⁹⁵ Additionally, the American Society of Colon and Rectal Surgeons recommends screening for other high-risk individuals, including MSM and women with a history of HPV+ gynecological precancer or cancer,⁹⁶ and the American Society of Transplantation Infectious Disease Community of Practice recommends screening of solid-organ transplant patients.⁹⁷ Most groups recommend annual screening, although guidelines for upper or lower age limits are lacking.⁹⁵ For patients with normal cytology and a negative HPV test, a screening interval of 3 years may be sufficient, based on the observation that the 3-year probability of HSIL was 7.5% among patients who were HPV negative at baseline (81% were HIV positive).^83,98 Since the majority (>90%) of MSM are HPV positive, primary HPV testing is not recommended in this group.⁹⁹ Referral for HRA is recommended for patients with abnormal cytology.^95,100 Although most groups recommend cytology as the primary screening test, a few groups recommend DARE with referral for HRA for suspicious lesions.⁹⁵ Guidelines on the management of abnormal lesions are also needed, as none have been issued by the CDC, USPSTF, or American Cancer Society owing to insufficient evidence favoring screening. The Anal Cancer HSIL Outcomes Research study (NCT02135419) is an ongoing trial to evaluate the effect of HSIL treatment (topical or ablative) vs. active surveillance on anal cancer incidence in HIV-positive men and women.

OROPHARYNGEAL CANCER

Despite ongoing vaccination efforts, the incidence of HPV-mediated oropharyngeal squamous cell carcinoma (HPV+ OPC) continues to rise rapidly. Over 13,000 cases are diagnosed in the United States annually,¹⁰¹ and the incidence is expected to double in the next decade.¹⁰² While HPV vaccination effectively reduces the prevalence of oral HPV infections,²⁹ just over half of United States adolescents are up to date with the vaccination series, and it will take decades until at-risk cohorts have been vaccinated.^9,102

In light of this, organizations including the USPSTF have encouraged exploration of other modes of HPV+ OPC prevention, including screening for oral HPV16 infections.¹⁰³ (Among the 13 oncogenic HPV genotypes, HPV16 has been identified as the primary screening target because 82% of all HPV+ OPCs are attributable to HPV16.)¹⁰⁴ However, numerous barriers preclude the development and implementation of an HPV+ OPC screening program: 1) lack of consensus regarding high-risk characteristics and thresholds that would define a highly-at-risk group eligible for screening; 2) lack of screening tests that are sufficiently valid for detecting and localizing early-stage disease; 3) poor understanding of the natural history of disease progression to HPV+ OPC, including the existence of a precancerous state; 4) lack of evidence that screened patients will exhibit earlier-stage disease and improved outcomes; and 5) insufficient characterization of the harms of screening.^105,106

Defining a high-risk group eligible for screening

Multiple risk factors for the development of HPV+ OPC have been identified. Predictive demographic and behavioral factors include male sex, middle to elderly age, tobacco use, more lifetime oral sexual partners, and more lifetime any sexual partners.¹⁰⁵ Valid biologic predictors of HPV+ OPC have also been identified and include oral oncogenic HPV infection,¹⁰⁷ serum HPV E antibodies, and circulating HPV DNA.¹⁰⁸ A few investigators have combined clinical and biologic data to categorize risk for oral oncogenic HPV infection and HPV+ OPC.^109,110 However, both risk prediction models require collection of sensitive sexual behavior data and it is unclear if this mode of screening is acceptable to eligible participants. Further, investigators have estimated that among a higher-risk group with an HPV+ OPC incidence rate of 20 per 100,000 individuals per year, an 80% sensitive and 99% specific screening test will require screening 12,500 individuals to detect one case of HPV+ OPC (number needed to screen [NNS]).¹⁰⁶ Prospective validation according to these screening algorithms along with characterization of the impact on the NNS will be necessary to define screening eligibility criteria.

Candidate screening tests

Although several candidate HPV+ OPC screening tests are available (Table 2), a sufficiently valid test or combination of tests has yet to be identified.

Table 2.

Candidate screening tests for HPV+ OPC.

Local/Oropharyngeal

Oral rinse for oral oncogenic HPV DNA detection

Oropharyngeal brush cytology ± HPV DNA detection

Direct clinical oropharyngeal examination

Transcervical oropharyngeal ultrasonography

Regional/Neck

Clinical examination of the neck

Neck ultrasonography

Systemic/Whole Body

HPV E antibodies

Circulating HPV DNA

PET-CT

Abbreviations: E: early proteins; HPV: human papillomavirus; OPC: oropharyngeal cancer; positron emission tomography-computed tomography (PET-CT)

Among screening tests that may detect local oropharyngeal disease, oral oncogenic HPV DNA may be the most promising and was included in a published risk prediction model.¹¹⁰ According to a cancer epidemiology cohort study, individuals with oral HPV16 DNA exhibited 22-fold greater odds of developing OPC.¹¹¹ This biomarker has several drawbacks, however, as 4% of United States adults have an oral oncogenic HPV infection,¹¹² and half will clear their infections within 7 months.¹¹³ Consequently, although reasonably sensitive (72%), oral HPV16 DNA screening is insufficiently specific (92%) – even among higher-risk groups with elevated HPV+ OPC incidence rates.¹⁰⁷ Perhaps even more importantly, since the biomarker is collected by oral rinse and gargle, it fails to localize the subsite, or even laterality, of the disease.

Oropharyngeal HPV DNA or cytology testing has demonstrated efficacy among patients with already-detected, clinical HPV+ OPC.¹¹⁴ However, to our knowledge, this modality has not detected precancer or cancer in at-risk, screened individuals without a previously identified cancer. Furthermore, it may not successfully access the tonsillar crypts (the presumed site of disease origin) and is therefore not favored.^115,116 Direct oropharyngeal exam with or without biopsy generally consists of transoral visualization and palpation of the oropharynx along with indirect mirror exam or flexible fiberoptic nasopharyngoscopy. The validity of nasopharyngoscopy may be augmented by narrow band imaging.¹¹⁷ The sensitivity of direct examination in isolation is not clearly known, but since approximately 90% of HPV+ OPC patients exhibit a known primary tumor,¹¹⁸ this may serve as the upper boundary of the true estimate. Finally, transcervical oropharyngeal ultrasonography is an emerging and promising screening test that has identified primary tumors as small as 5 mm.¹¹⁹

Clinical examination of the neck is a potential screening test to detect early regional metastatic disease, but according to one institutional study of consecutive head and neck cancer patients, it is only 67% sensitive and 73% specific.¹²⁰ Neck ultrasonography, according to a meta-analysis of eight studies of head and neck cancer patients without palpable neck disease (cN0 patients), is 66% sensitive and 78% specific.¹²¹

Systemic screening tests for HPV E antibodies and circulating HPV DNA exhibit excellent validity¹⁰⁸ but fail to localize disease. In a recent meta-analysis of three case-control studies, HPV16 E6 antibody seropositivity was 56% sensitive and 99% specific for detecting HPV+ OPC. Further, the HPV Cancer Cohort Consortium reports that HPV16 E6 seropositivity precedes HPV+ OPC diagnosis by 6–28 years.¹²² Digital droplet polymerase chain reaction for the detection of circulating HPV DNA is also promising, and, according to one multi-institutional study evaluating five oncogenic genotypes, yielded a sensitivity rate of 89%, with 97% specificity for detection of newly diagnosed HPV+ OPC.¹²³ This same marker had high sensitivity and specificity (100% and 99%, respectively) for the detection of recurrent disease among patients with primary HPV+ OPC undergoing definitive chemoradiation.¹²⁴ Unfortunately, as described above, the favorable performance characteristics of these systemic biomarkers will likely need to be augmented further to satisfactorily lower the NNS.

Finally, positron emission tomography (PET)-CT may be useful in a stepped HPV+ OPC screening program involving biomarkers and imaging. PET-CT trends towards exhibiting the highest sensitivity of all imaging modalities¹²⁵ and is already standard of care for patients with unknown primary HPV+ OPCs. Since approximately 10% of clinical HPV+ OPC patients present with an unknown primary tumor,¹¹⁸ it is reasonable to presume that an even higher proportion of screened HPV+ OPCs will exhibit an unknown primary. PET-CT exhibits notable drawbacks: it is costly, requires patients fast prior to imaging and is less accessible relative to other imaging modalities. Importantly, however, the feasibility of a stepped multicancer blood test followed by whole body PET-CT in 10,006 healthy elderly women has already been demonstrated.¹²⁶ Large-scale trials would be needed to better understand the benefits, harms and costs of this HPV+ OPC screening approach.

Other Barriers to HPV+ OPC Screening

Natural history of HPV+ OPC

Although identification of oral HPV DNA and serum HPV16 E antibodies years and even decades prior to an HPV+ OPC diagnosis supports the existence of a precancerous disease state, the exact nature of disease progression is unknown. Some head and neck pathologists lament that a true precancerous disease state may not exist, given the gaps in the basement membrane of the tonsillar crypt epithelium.¹²⁷ Further, even if an intermediate outcome was unequivocally defined, our inability to reliably localize subclinical disease would inhibit treatment.

Treatment of HPV+ OPC detected by screening

Whether or not an intermediate outcome can be identified and treated, screened patients may exhibit comparably earlier stage disease amenable to de-escalated treatment. Prospective studies of screened versus unscreened individuals will be needed to adjudicate between-group differences in stage at diagnosis, delivery of unimodal or de-intensified therapy, and patient-reported and oncologic outcomes.

Screening harms

The harms caused by screening for HPV+ OPC have not yet been characterized. However, screening for its sister disease – HPV-associated cervical cancer – provides some insight into potential psychological and other harms. These harms include frequent follow-up testing, invasive diagnostic procedures, overdiagnosis, unnecessary treatment of patients with false-positive results, and lesions that may spontaneously regress.⁵⁶

CONCLUSION

Prevention of HPV-related cancers is now possible with the availability of a safe and highly effective HPV vaccine. While most HICs have established vaccination programs, coverage remains low in LMICs. However, with the worldwide vaccination coverage still low, screening for cervical cancers will need to continue, and the access to and compliance with such screening must be expanded. Additionally, with the increased incidences of both anal and oropharyngeal cancers in HICs, increased anal cancer screening for high-risk individuals and the development of a screening strategy for oropharyngeal cancer are also imperative.