CHAPTER 94: HUMAN PAPILLOMAVIRUS INFECTIONS

Investigation of human papillomavirus (HPV) infection began in earnest in the 1980s after Harold zur Hausen postulated that infection with these viruses was associated with cervical cancer. It is now recognized that HPV infection of the human genital tract is extremely common and causes clinical states ranging from asymptomatic infection to genital warts (condylomata acuminata); dysplastic lesions or invasive cancers of the anus, penis, vulva, vagina, and cervix; and a subset of oropharyngeal cancers. This chapter describes the epidemiology of HPV in general and as a pathogen, the natural history of HPV infections and associated cancers, strategies to prevent HPV infection and HPV-associated disease, and treatment modalities.

Molecular overview

HPV is an icosahedral, nonenveloped, 8000-base-pair, double-stranded DNA virus with a diameter of 55 nm. Like those of other papillomaviruses, HPV’s genome consists of an early (E) gene region, a late (L) gene region, and a noncoding region that contains regulatory elements. The E1, E2, E5, E6, and E7 proteins are expressed early in the growth cycle and are necessary for viral replication and cellular transformation. The E6 and E7 proteins cause malignant transformation by targeting the human cell cycle–regulatory molecules p53 and Rb (retinoblastoma protein), respectively, for degradation. Translation of the L1 and L2 transcripts and splicing of an E1^E4 transcript occur later. The L1 gene encodes the 54-kDa major capsid protein that makes up the majority of the virus shell; the 77-kDa L2 minor protein constitutes a smaller percentage of the capsid mass.

More than 125 HPV types have been identified and are numerically designated according to a unique L1 gene sequence. Approximately 40 HPV types are regularly found in the anogenital tract and are subdivided into high-risk and low-risk categories on the basis of the associated risk of cervical cancer. For example, HPV-6 and HPV-11 cause genital warts and ~10% of low-grade cervical lesions and are thus designated low risk. HPV-16 and HPV-18 cause dysplastic lesions and invasive cancers of the cervix and are considered high risk.

HPV targets basal keratinocytes after microtrauma has exposed these cells to the virus. The HPV replication cycle is completed as keratinocytes undergo differentiation. Virions are assembled in the nuclei of differentiated keratinocytes and can be detected by electron microscopy. Infection is transmitted by contact with virus contained in these desquamated keratinocytes (or with free virus) from an infected individual.

Immune response to HPV infection

A cell-mediated immune response plays an important role in controlling the progression of natural HPV infection. Histologic examination of lesions in individuals who experience regression of genital warts demonstrates infiltration of T cells and macrophages. CD4+ T cell regulation is particularly important in controlling HPV infections, as evidenced by the higher rates of infection and disease among immunosuppressed individuals, particularly those who are infected with HIV. Specific T-cell responses may be measured against HPV proteins, the most important of which appear to be the E2 and E6 proteins. In women with HPV-16 cervical infection, a strong T-cell response to HPV-16-derived E2 protein is associated with a lack of progression of cervical disease.

Natural HPV infection of the genital tract gives rise to a serum antibody response in only 60–70% of individuals because there is no viremic phase during infection. Significant, although incomplete, protection against type-specific reinfection is associated with the presence of neutralizing antibodies. Serum antibodies likely reach the cervical epithelium and secretions by transudation and exudation. Therefore, protection against infection is related to the amount of neutralizing antibody at the site of infection and lasts as long as levels of neutralizing antibodies are sufficient.

General population

HPV is transmitted by sexual intercourse, by oral sex, and possibly by touching of a partner’s genitalia. In cross-sectional and longitudinal studies, ~40% of young women have evidence of HPV infection, with peaks during the teens and early twenties—soon after first coitus. The number of lifetime sexual partners correlates with the likelihood of HPV infection and the subsequent risk of HPV-associated malignancy. HPV infection may develop in a monogamous person whose partner is infected. Most HPV infections become undetectable after 6–9 months. However, with prolonged follow-up and frequent sampling, the same HPV types may again be detected weeks or months after becoming undetectable. Whether such episodic detection indicates viral latency followed by reactivation or reinfection with an identical HPV type is still debated.

Although HPV is the causative agent of several cancers, most attention has focused on cervical cancer—the second most common cancer among women worldwide, which affects more than 500,000 women and kills more than 275,000 women annually. More than 85% of all cervical cancer cases and deaths occur in women living in low-income countries, especially in sub-Saharan Africa, Asia, and South and Central America. A quarter-century of evidence shows that HPV causes nearly 100% of cervical cancers. HPV infection is the most significant risk factor for cervical cancer; relative risks range from 10 to 20 and exceed 100 in prospective and case-control studies, respectively. The time from HPV infection to cervical cancer diagnosis may exceed 20 years. Cervical cancer peaks in the fifth and sixth decades of life among women living in developed countries but as much as a decade earlier among women living in resource-poor countries. Persistent carriers of oncogenic HPV types are at greatest risk for high-grade cervical dysplasia and cancer. Why only certain HPV infections eventually lead to malignancy is not clear. Biomarkers that can predict which women will develop cervical cancer are not available. Immunosuppression in general plays a significant role in re-detection/reactivation of HPV infections, while other factors such as smoking, hormonal changes, Chlamydia infection, and nutritional deficits promote viral persistence and cancer.

The International Agency for Research on Cancer concludes that HPV types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, and 59 are carcinogenic in the uterine cervix. HPV-16 is particularly virulent and causes 50% of cervical cancers. Worldwide, HPV-16 and HPV-18 cause 70% of cervical squamous cell carcinomas and 85% of cervical adenocarcinomas. Oncogenic types other than HPV-16 and HPV-18 cause the remaining 30% of cervical cancers. HPV-16 and HPV-18 also cause nearly 90% of anal cancers worldwide. Although oncogenic HPV infection is necessary for the development of cervical malignancy, only ~3–5% of infected women will ever develop this cancer, even in the absence of cytologic screening.

In addition to cervical and anal cancer, other HPV-associated cancers include vulvar and vaginal cancer, which are associated with HPV in 50–70% of cases; penile cancer, which is caused by HPV in 50% of cases; and oropharyngeal squamous cell carcinoma (OPSCC). Over the past two decades, an epidemic of OPSCC related to oncogenic infection with HPV (primarily HPV-16) has developed. Annual rates of OPSCC among men in the United States have been increasing from a low of 0.27 case/100,000 in 1973 to 0.57 case/100,000 as of 2004; rates in women have remained relatively stable at ~0.17 case/100,000 per year. The increase in the incidence of OPSCC is greatest among white men 40–50 years of age. Nearly 14,000 new cases were diagnosed in the United States in 2013. Annual rates of OPSCCs of the base of the tongue and the tonsil have increased dramatically—i.e., by 1.3% and 0.6%, respectively. Fewer data are available from developing countries about OPSCCs.

Effects of HIV on HPV-associated disease

HIV infection accelerates the natural progression of HPV infections. HIV-infected persons are more likely than other individuals to develop genital warts and to have lesions that are more recalcitrant to treatment. HIV infection has been consistently associated with precancerous cervical lesions, including low-grade cervical intraepithelial neoplasia (CIN) and CIN 3, the immediate precursor to cervical cancer. Women with HIV/AIDS have significantly higher rates of cervical cancer and of subsets of some vulvar, vaginal, and oropharyngeal tumors than women in the general population. Studies indicate a direct relation between low CD4+ T lymphocyte counts and the risk of cervical cancer. Some studies show a reduced likelihood of HPV infection and precancerous lesions of the cervix in HIV-infected women receiving antiretroviral therapy (ART). The incidence of cervical cancer among HIV-infected women has not changed significantly since ART was introduced, possibly because of preexisting oncogenic HPV infections.

The burden of HPV-related cancers is expected to increase in HIV-infected patients, given the prolonged life expectancies made possible by ART. For women living in developing countries where cervical cancer screening is not widely available, this situation may have significant consequences. Thus, elucidating the interactions of HIV infection and cervical cancer with cofactors such as diet, other sexually transmitted infections, and environmental exposures is a research focus with potentially enormous implications for women living in low- and middle-income countries.

Similar to that of cervical cancer, the incidence of anal cancer is strongly influenced by HIV infection. HIV-infected men who have sex with men (MSM) and HIV-infected women have much higher rates of anal cancer than HIV-uninfected populations. Specifically, the incidence has been found to be as high as 130 cases/100,000 among HIV-positive MSM as opposed to only 5 cases/100,000 among HIV-negative MSM. The advent of ART has not impacted the incidence of anal cancer and high-grade anal intraepithelial neoplasia in the HIV-infected population.

More information on screening, prevention, and treatment in the HIV-infected population can be found at the Department of Health and Human Services website (aidsinfo.nih.gov/guidelines).

HPV infects the female vulva, vagina, and cervix and the male urethra, penis, and scrotum. Perianal, anal, and oropharyngeal infections occur in both genders. Figures 94-1, 94-2, 94-3, 94-4 show vulvar, penile, penile and scrotal, and perianal warts, respectively. Genital warts are caused primarily by HPV-6 or HPV-11; their surface is either smooth or rough. Penile genital warts are usually 2–5 mm in diameter and often occur in groups. A second type of penile lesion, keratotic plaques, is slightly raised above the normal epithelium and has a rough, often pigmented surface. Vulvar warts are soft, whitish papules that either are sessile or have multiple fine, finger-like projections. These lesions are most often located in the introitus and on the labia. In nonmucosal areas, lesions are similar in appearance to those in men: dry and keratotic. Vulvar lesions can appear as smooth, sometimes pigmented papules that may coalesce. Vaginal lesions appear as multiple areas of elongated papillae. Biopsy of vulvar or vaginal lesions may reveal malignancy, which is not always reliably identified by clinical examination.

Subclinical cervical HPV infections are common, and the cervix may appear normal on examination. Cervical lesions often appear as papillary proliferations near the transformation zone. Irregular vascular loops are present beneath the surface epithelium. Patients who develop cervical cancer arising from HPV infection may present with a variety of symptoms. Early carcinomas appear eroded and bleed easily. More advanced carcinomas present as ulcerated lesions or as an exophytic cervical mass. Some cervical carcinomas are located in the cervical canal and may be difficult to see. Bleeding, symptoms of a mass lesion in late stages, and metastatic disease that may manifest as bowel or bladder obstruction due to direct extension of the tumor have also been described.

Patients with squamous cell cancer of the anus have more variable presentations. The most common presentations include rectal bleeding and pain or a mass sensation. Of patients who are diagnosed with anal cancer, 20% may present with no specific symptoms at the time of diagnosis; rather, the lesion is found fortuitously.

Vaccines effective in preventing HPV infection and HPV-associated disease represent a major development in the last decade. The vaccines use virus-like particles (VLPs) that consist of the HPV L1 major capsid protein. The L1 protein self-assembles into VLPs when expressed in eukaryotic cells (i.e., yeast for the Merck vaccine or insect cells for the GlaxoSmithKline vaccine; see below). These VLPs contain the same epitopes as the HPV virion. However, they do not contain genetic material and cannot transmit infection. The immunogenicity of HPV vaccines relies on the development of conformational neutralizing antibodies to epitopes displayed on viral capsids.

Several large trials have demonstrated the high degree of safety and efficacy of HPV vaccines. The evidence to date has shown high and sustained efficacy against disease caused by those HPV types represented in the vaccines (HPV-6, -11, -16, and -18 in the Merck vaccine and HPV-16 and -18 in the GlaxoSmithKline vaccine). However, no therapeutic effect against active infection or disease has been found for either vaccine.

Bivalent vaccine (Cervarix)

A bivalent L1 VLP vaccine (HPV-16 and -18), marketed under the name Cervarix (GlaxoSmithKline), is administered by IM injection at months 0, 1, and 6. This vaccine was tested in 18,644 women 15–25 years of age who were residing in the United States, South America, Europe, and Asia. The primary endpoints of the study included vaccine efficacy against persistent infections with HPV-16 and -18. Investigators also assessed the vaccine’s efficacy against CIN of grade 2 or higher due to HPV-16 and -18 in women who had no evidence of infection with these HPV types at baseline; in these women, vaccine efficacy was 94.9% (95% confidence interval [CI], 87.7 to 98.4) against CIN ≥2 related to HPV-16 or HPV-18, 91.7% (95% CI, 66.6 to 99.1) against CIN ≥3, and 100% (95% CI, –8.6 to 100) against adenocarcinoma in situ.

Adverse events were evaluated in phase 3 trials in a subset of 3077 women who received the bivalent vaccine and 3080 women (controls) who received hepatitis A vaccine. Injection-site adverse events (pain, redness, and swelling) and systemic adverse events (fatigue, headache, and myalgia) were reported more frequently in the HPV vaccine group than in the control group. Serious adverse events (mainly injection-site reactions), new-onset chronic disease, or medically significant conditions occurred in 3.5% of HPV vaccine recipients and in 3.5% of women receiving the control vaccine.

The bivalent vaccine is approved in the United States for prevention of cervical cancer, CIN ≥2, adenocarcinoma in situ, and CIN 1 caused by HPV-16 and -18. This vaccine is approved for administration to girls and women 9–25 years of age.

Quadrivalent vaccine (Gardasil)

A quadrivalent L1 VLP (HPV-6, -11, -16, and -18) vaccine, marketed under the name Gardasil (Merck), is administered IM at months 0, 2, and 6. A combined efficacy analysis based on data from four randomized double-blind clinical studies including more than 20,000 participants demonstrated that the vaccine’s efficacy against external genital warts was 98.9% (95% CI, 93.7 to 100). Its efficacy was 95.2% (95% CI, 87.2 to 98.7) in protecting against CIN, 100% (95% CI, 92.9 to 100) against HPV-16- or HPV-18-related CIN 2/3 or adenocarcinoma in situ, and 100% (95% CI, 55.5 to 100.0) against HPV-16- or HPV-18-related vulvar intraepithelial neoplasia grades 2 and 3 (VIN 2/3) and vaginal intraepithelial neoplasia grades 2 and 3 (VaIN 2/3).

Safety data on the quadrivalent HPV vaccine are available from seven clinical trials including nearly 12,000 girls and women 9–26 years of age who received the vaccine and ~10,000 who received placebo. A larger proportion of young women reported injection-site adverse events in the vaccine groups than in the aluminum-containing or saline placebo groups. Systemic adverse events were reported by similar proportions of vaccine and placebo recipients and were described as mild or moderate by most participants. The types of serious adverse events reported were similar for the two groups. Ten persons who received the quadrivalent vaccine and seven persons who received placebo died during the course of the trials; no deaths were considered to be vaccine related.

During the course of the quadrivalent vaccine trials, surveillance data on the development of new medical conditions were collected for up to 4 years after vaccination. No statistically significant differences in the incidence of any medical conditions between vaccine and placebo recipients were found; this result indicated a very good safety profile. A recent safety review by the U.S. Food and Drug Administration and the Centers for Disease Control and Prevention (CDC) examined events related to Gardasil that had been reported to the Vaccine Adverse Events Reporting System. Adverse events were consistent with those seen in previous safety studies of the vaccine. It is noteworthy that rates of syncope and venous thrombotic events were higher with Gardasil than those that have usually been documented for other vaccines.

The quadrivalent vaccine is approved for (1) vaccination of girls and women 9–26 years of age to prevent genital warts and cervical cancer caused by HPV-6, -11, -16, and -18; (2) vaccination of the same population to prevent precancerous or dysplastic lesions, including cervical adenocarcinoma in situ, CIN 2/3, VIN 2/3, VaIN 2/3, and CIN 1; (3) vaccination of boys and men 9–26 years of age to prevent genital warts caused by HPV-6 and -11; and (4) vaccination of individuals 9–26 years of age to prevent anal cancer and associated precancerous lesions due to HPV-6, -11, -16, and -18.

Cross-protection of HPV vaccines

Women vaccinated with either of the available vaccines produce neutralizing antibodies against types related to HPV-16 or -18. Analyses of data from clinical trials suggest that both vaccines may offer cross-protection against nonvaccine types. The bivalent vaccine appears more efficacious against HPV-31, -33, and -45 than the quadrivalent vaccine, but differences in study design make direct comparisons difficult. In addition, vaccine efficacy against persistent infections with HPV-31 and -45 appeared to wane over time in the bivalent vaccine trials, whereas efficacy against persistent infection with HPV-16 or -18 remained stable.

Second-generation vaccines

While HPV-16 and -18 cause the majority of cervical cancers worldwide, global data have shown that HPV-31, -33, -35, -45, -52, and -58 are the next most frequently detected types in invasive cervical cancers. Second-generation vaccines that are in development incorporate VLPs of additional oncogenic HPV types (other than HPV-16 and -18), including HPV-31, -33, -45, -52, and -58; efficacy studies are ongoing. If vaccines with these five additional oncogenic types prove to be effective, mathematical models estimate that the level of protection could be raised to 90% of all squamous cell cervical cancers worldwide.

Recommendations for vaccination

The CDC’s Advisory Committee for Immunization Practice recommends administration of the quadrivalent HPV vaccine—with the schedule used in the vaccine trials—to all boys and girls 11–12 years of age as well as to boys/men and girls/women 13–26 years of age who have not previously been vaccinated or who have not completed the full series. For women, Papanicolaou (Pap) smear testing and screening for HPV DNA are not recommended before vaccination. After vaccination, Pap testing is recommended to detect disease caused by other oncogenic HPV types.

After HPV infection occurs, prevention of HPV-associated disease relies on screening. Women residing in developing countries who lack access to cervical screening programs have a higher rate of cervical cancer and a lower rate of cancer-specific survival. Approximately 75% of women living in developed countries have been screened in the past 5 years, whereas the figure is only ~5% among women living in developing countries. Economic and logistic obstacles likely impede routine screening of these populations for cervical cancer.

The primary method used for cancer screening is cervical cytology via Pap smear. The guidelines of the American Society of Colposcopy and Cervical Pathology recommend initiation of cervical cancer screening at age 21, regardless of the age of sexual debut. Women 21–29 years old with a normal Pap smear should have the test repeated every 3 years. Although adolescent and young women often test positive for HPV DNA, they are at very low risk of cervical cancer. Co-testing, or testing for HPV DNA at the time of the Pap smear, is not recommended for women in this age group because the presence of HPV DNA does not correlate with the presence of high-grade squamous intraepithelial neoplasia. Women 30–65 years of age should have a Pap smear every 3 years if testing for HPV DNA is not performed. The screening interval for women in this age group can be extended to every 5 years if co-testing results are negative. HPV testing is not recommended for partners of women with HPV or for screening for conditions other than cervical cancer.

Currently, there is no clear consensus regarding screening for anal cancer and its precursors, including high-grade anal intraepithelial lesions. This lack of clarity is due to an inadequate understanding of optimal treatment for low- or high-grade anal dysplasia found during cytologic screening. The current HIV treatment guidelines suggest that there may be a benefit to screening, but an effect on the associated morbidity and mortality of anal squamous cell cancer has not been consistently demonstrated.

Most sexually active adults will be infected with HPV during their lives. For all patients (vaccinated or unvaccinated), certain behavioral interventions can reduce the risk of acquiring HPV. Physicians can provide their patients with measures that can reduce this risk. The only way to avoid acquiring an HPV infection is to abstain from sexual activity, including intimate touching and oral sex. Practicing safe sex (partner reduction, condom use) may lower the likelihood of HPV transmission. Most HPV infections are controlled by the immune system and cause no symptoms or disease. Some infections lead to genital warts and cervical precancers. Genital warts can be treated for cosmetic reasons and to prevent spread of infection to others. Even after resolution of genital warts, latent virus can persist in normal-appearing skin or mucosa and thus theoretically can be transmitted to uninfected partners. Precancerous cervical lesions should be treated to prevent progression to cancer.