HPV Prevention: A New Era in Patient Care

Introduction HPV Vaccine Cross–Protection:Are We There Yet? Kenneth A. Alexander, MD, PhD Burden of HPV in Males Anna R. Giuliano, PhD HPV Vaccine Safety William L. Atkinson, MD, MPH Discussion

Introduction

In the United States, human papillomavirus (HPV) infects approximately 20 million people, one-half of whom are between the ages of 15 and 24 years. More than 6 million new cases of HPV infection occur each year. In June 2006, the FDA licensed for use in females aged 9 to 26 years a vaccine indicated for the prevention of cervical cancer, genital warts, and cervical, vaginal, and vulvar precancerous or dysplastic lesions caused by HPV types 6, 11, 16, and 18. The efficacy of this vaccine against these 4 HPV types has been demonstrated in several clinical trials. Because the HPV vaccine is so highly efficacious for the prevention of diseases due to vaccine-type HPVs, we can ask whether HPV vaccines confer protection against non-vaccine HPV types. Finally, as with any vaccine that acheives widespread use, ongoing safety concerns, both real and imagined, must be addressed both partially and accurately.

To provide an update on the answers to these questions that are being posed in the highly dynamic field of HPV vaccinology, Vindico Medical Education organized an educational symposium in New York in November 2008. Prominent HPV researchers and epidemiologists shared the latest data on these important concerns, focusing on cross-protection provided by the extant vaccines, the burden of HPV infection in males, and the safety of the HPV vaccine.

I thank the participants, both for their contributions to the symposium and for their roles in the development of this monograph. Readers of this monograph will achieve a more comprehensive and up-to-date understanding of the overall burden of HPV infection and the importance of its prevention.

Kenneth A. Alexander, MD, PhD
Course Chair

Kenneth A. Alexander, MD, PhD Course Chair Associate Professor of Pediatrics Chief, Pediatric Infectious Diseases University of Chicago Chicago, Illinois Anna R. Giuliano, PhD Chair, Department of Epidemiology and Genetics Program Leader, Risk Assessment, Detection, and Intervention Program H. Lee Moffitt Cancer Center Tampa, Florida William L. Atkinson, MD, MPH Medical Epidemiologist National Center for Immunization and Respiratory Diseases Centers for Disease Control and Prevention Atlanta, Georgia

HPV Vaccine Cross-Protection: Are We There Yet?

Kenneth A. Alexander, MD, PhD

Currently, only 1 human papillomavirus (HPV) vaccine licensed by the FDA. Gardasil (HPV quadrivalent [types 6, 11, 16, and 18] vaccine, recombinant) is a quadrivalent vaccine administered as three 0.5-mL doses at 0, 2, and 6 months. A bivalent vaccine has been approved in several other countries, but is not yet licensed in the United States.

The quadrivalent vaccine is produced in the yeast Saccharomyces cerevisiae; the bivalent vaccine is produced in the SF9 insect cell line. For both vaccines, the expressed L1 capsid protein (the protein outside of the virus) self-assembles into geometric virus-like particles (VLPs), which are highly immunogenic. The VLPs are purified and adsorbed onto an aluminium-containing adjuvant. Lacking other viral proteins, as well as viral DNA and RNA, the HPV vaccine is not a live, attenuated, or killed vaccine. Accordingly, it is incapable of transmitting any conditions associated with the HPVs, such as anogenital warts, cervical dysplasia, or cervical cancer.

Almost 200 HPV viral strains have been identified, 7 of which are responsible for approximately 85% of all HPV-related cervical cancers (Table 1).¹ Strains 16 and 18 are the most oncogenic and are associated with approximately 70% of cases of cervical cancer and high-grade cervical dysplasia. Each of the remaining strains accounts for less than 5% of cervical cancers.

Table 1. Oncogenic HPV Strains * Included in HPV Vaccine SOURCE: Muñoz N, et al. N Engl J Med. 2003; 348:518-527. de Villiers EM, et al. Virology. 2004; 324: 17-24.

The available vaccines include oncogenic HPV types 16 and 18; accordingly, they should reduce infection with HPV 16 and HPV 18 by up to 70%. Adding more antigens against other oncogenic HPV types would confer broader protection against cervical malignancies; a 9-valent vaccine is currently being evaluated. While we await release of higher valency HPV vaccines, it is worth asking whether our current vaccines confer protection against non-vaccine HPV types.

The notion that HPV VLP vaccines would confer cross-protection against non-vaccine HPV types arises in part from the fact that vaccine efficacy against vaccine-type HPVs is very high. Furthermore, because HPV type is determined on the basis of homology between viral (surface) L1 proteins, it is conceivable that vaccine VLPs would confer protection against virus types with structurally similar L1 proteins. Because the L1 protein sequence provides the basis for designating HPV type,² cross-protection may be achieved against viruses that are in the same group. For example, oncogenic virus type HPV-16 is in Group 9, which also includes HPV-31, -33, -35, -52, -58, and -67. HPV-18, the other oncogenic virus type in the vaccine, is classified in Group 7 with HPV-39, -45, -59, -68, -70, and -85.

Studies Examining Cross-Protection Against Non-Vaccine Oncogenic HPV Types

Four studies have been reported that investigated vaccine-induced cross-protection against non-vaccine oncogenic HPV types. Two of these studies examined cross-protection induced by the bivalent HPV vaccine, and 2 studies examined cross protection induced by the quadrivalent HPV vaccine. Primary and secondary endpoints varied among the studies, and included incident and persistent infection, cervical intraepithelial neoplasia (CIN) 2/3, and adenocarcinoma in situ. Analyses included pre-specified and post hoc comparisons.

Understanding the advantages and clinical usefulness of endpoints in cross-protection studies is important to allow appropriate interpretation of results. As a study endpoint, incident infection has the advantage of allowing identification of the causal virus(es), typically through the use of polymerase chain reaction (PCR) technology, which yields a definitive yes/no answer (i.e, a specific virus type is present, or it is not). Incident infection is also a common endpoint that occurs rapidly in the context of a clinical trial. However, persistent, not incident, infections are associated with cancer risk. In fact, most incident infections regress within 6 months. Thus, persistent infection, because it correlates with cancer risk, is a logical study endpoint. Furthermore, like incident infections, the presence or absence of a specific HPV type at any time point is readily determined by PCR. Cytological abnormality is also a reasonable study endpoint. Obviously, HPV vaccines are intended to prevent cervical cytological abnormailities. Thus, cytological abnormalities are the most “real-world” study endpoints. However, while cytological abnormalities are accepted as the most relevant study endpoint, they present a problem of attribution in patients with mixed HPV infections. In a patient with a mixed infection, on the other hand, accurately attributing the abnormalities to the true causative virus type can be problematic. For example, if a woman is infected with both HPV-16 and HPV-52, it would be logical to ascribe the cytological abnormalities to the more aggressive HPV-16 virus; however, it cannot be definitively proven that HPV-16 is indeed the virus causing the observed cytopathology.

A final issue that distinguishes cross-protection studies is the form of statistical analysis used. Specifically, cross-protection effects can be analyzed as an activity against each specific virus type. When data are analyzed by individual virus type, small case numbers may not provide adequate power to achieve statistical significance. If viruses are analyzed together as a group, statistically meaningful data can be acquired more rapidly. However, if there are differences in the response to individual virus types within a group, efficacy of specific virus types can be obscured, and a diminished overall efficacy value can result.

The Ability of the Bivalent Vaccine to Prevent Incident Infection by Non-Vaccine Viral Types

A long-term follow-up study of a randomized, placebo-controlled trial of the bivalent HPV-16/18 vaccine enrolled 776 women who fulfilled eligibility requirements that specified seronegativity for the vaccine virus types and who were negative for high risk HPV DNA at the beginning of the initial efficacy trial.³ More than 98% seropositivity was maintained for HPV-16/18 for up to 4.5 years after immunization. Vaccine efficacy against incident infection (defined as 1 minus the ratio between the infection rate detected at a single encounter in the vaccinated and placebo groups) was 96.9% (95% CI: 81.3%, 99.9%) against HPV-16/18. Efficacies of 94.2% (95% CI: 63.3%, 99.9%) and 54.5% (95% CI: 11.5%, 77.7%) were also achieved against the high-risk non-vaccine virus types HPV-45 and HPV-31, respectively. Lesser effect was observed with HPV-33, HPV-52, and HPV-58, with efficacies ranging from 8.6% to 18.6%. These data show that the bivalent vaccine conferred significant protective efficacy against incident infection caused by 2 non-vaccine viral types in addition to the types contained in the vaccine.

The Ability of the Bivalent Vaccine to Prevent Persistent Infection by Non-Vaccine Viral Types: The PApilloma TRIal against Cancer In Young Adults (PATRICIA)

An event-triggered interim analysis of this international, phase III, double-blind, placebo-controlled, randomized trial was performed after a mean follow-up of 14.8 months, when 23 cases of CIN 2+ were detected in the 18,525 women comprising the total vaccinated efficacy cohort.⁴ The primary objective of the study was to determine vaccine efficacy against HPV-16/18-associated CIN 2+, based on DNA detection in the lesion. Other objectives included 6- and 12-month efficacy against persistent infection with other HPV types. In contrast to the previous study, in PATRICIA, women with prevalent oncogenic HPV infections who often had several HPV types, and those with low-grade cytological abnormalities at study entry were included in PATRICIA.

In addition to efficacy greater than 90% for the primary endpoint, significant cross-protection against persistent infection with HPV-31, -45, and -52 was achieved at 6 months (59.9%, 36.1%, and 31.6% efficacy, respectively), at which time there were still more than 6000 women in each group (Table 2). Data on persistent infection at 12 months included approximately 3500 women per group. Although the trend toward fewer persistent infections associated with specific HPV types in the vaccine group remained, due to the small number of cases statistical significance was not achieved. However, in a pooled analysis, vaccine efficacy of 27.1% (97.9% CI: 0.5%, 46.8%) against 12 non-vaccine HPV types was statistically significant (P=.0174).

Table 2. Six month Persistent Infection with Oncogenic HPV Types Following Bivalent HPV Vaccination * Data are % (97.9% CI) SOURCE: Paavonen J, et al. Lancet. 2007; 369: 2161–2170.

Results of this interim analysis demonstrate, therefore, that the bivalent vaccine confers significant cross-protection against 6-month persistent infection by HPV types 31, 45, and 52, and may confer cross-protection against 12 month persistent infections.

Cross-protection with the Quadrivalent Vaccine: The Females United to Unilaterally Reduce Endo/Ectocervical Disease (FUTURE) Studies

The FUTURE I and FUTURE II trials enrolled almost 18,000 women who were randomized to receive the quadrivalent vaccine or adjuvant placebo.⁵ Supportive analyses were performed on a subset of subjects who were HPV-naïve (negative PCR and sero-negative to HPV-6, -11, -16, and -18—the vaccine types, and negative PCR to non-vaccine types -31, -33, -35, -39, -45, -51, -52, -56, -58, -59) at the time of enrollment. This population approximates the primary population targeted for HPV vaccination. These subjects must have received at least 1 dose of vaccine, with a normal Papanicolaou (Pap) test at day 1. Study endpoints were persistent infection and CIN 2/3 or adenocarcinoma in situ. Persistent infection was defined as detection of the same HPV type at 2 consecutive visits 6 or more months apart (±1 month), or presence of cervico-genital disease caused by the relevant HPV type, with DNA from the same HPV type found at a visit immediately prior to or after the biopsy. Combined analysis after an average follow-up of 3 years showed 45% efficacy for prevention of persistent infection with HPV-31/45, and 28% efficacy for HPV-31/33/45/52/58 (Table 3). In addition, the quadrivalent vaccine had 62% and 43% efficacy against CIN 2/3 or adenocarcinoma in situ related to HPV-31/45 and HPV-31/33/45/52/58 virus types, respectively.⁵

Table 3. Cross-protection to Non-Vaccine HPV Types by the Quadrivalent HPV Vaccine SOURCE: Brown D and the Future Study Group. The 47th Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC), Chicago, Illinois, September 17 – 20, 2007.

Although direct comparisons among studies are complicated by differing samples and endpoints, comparing the 6-month persistent infection data from the bivalent vaccine studies with those obtained from the quadrivalent vaccine studies suggests that the two vaccines induce similar degrees of cross-protection against non-vaccine oncogenic HPV types. For example, the 45% efficacy against persistent infection with HPV-31/45 achieved with the quadrivalent vaccine compares with the efficacy seen with the combined HPV-31 with HPV-45 data from the bivalent vaccine study. Similarly, combined efficacy across the 5 grouped HPV types in the bivalent vaccine study is calculated at 27%, which is similar to the 28% efficacy against persistent infection reported for the group of HPV 31/33/45/52/58 virus types in the quadrivalent vaccine study.

That both the quadrivalent and bivalent vaccines induce significant cross-protection against non-vaccine oncogenic HPV types is encouraging, and reflects the impressive immune responses induced by both vaccines. Nonetheless, several important issues, such as the duration of cross-protection, and role of vaccine-induced cross-protection at sites other than the uterine cervix, remain to be studied.

Putting the Numbers in Place: If We Aren’t There Yet, Are We On the Way?

What will the benefits be of HPV vaccine-induced cross protection in the real world? Based on the data reviewed here, a 60% cross-protection efficacy against CIN 2/3 or adenocarcinoma in situ caused by HPV-31/45 can be achieved with the extant vaccines. Together, these 2 HPV types account for approximately 8.5% of CIN 2/3 and adenocarcinoma in situ (Table 1). Accordingly, 60% of the 8.5% of cases, or approximately 5% of cases, could be prevented using currently-available HPV vaccines. Applying this 5% reduction to the 11,000 cases of cervical cancer per year in the US, between 500 to 600 cases of cervical cancer could be prevented. Furthermore, with 200,000 cases of CIN2/3 occurring each year in the United States, a 5% reduction would eliminate approximately 10,000 cases of CIN 2/3. Thus, viewed in the light of the high prevalence of HPV-associated diseases, the modest, though real, benefits of vaccine-induced cross-protection against non-vaccine oncogenic HPV types amounts to real savings of women’s lives, and in prevention of HPV-associated morbidity and medical expenses.

References

1. Muñoz N, et al. N Engl J Med. 2003; 348:518-527.
2. de Villiers EM, et al. Virology. 2004; 324: 17-24.
3. Harper DM. The Lancet. 2006; 367:1247-1255.
4. Paavonen J. Lancet. 2007; 369: 2161–2170.
5. Brown D and the Future Study Group. The 47th Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC), Chicago, Illinois, September 17 – 20, 2007.

Burden of HPV in Males

Anna R. Giuliano, PhD

Why men? An interest in human papilloma virus (HPV) has typically occupied gyn-oncologists and gynecologists, who share a concern about preventing the worldwide burden of cervical cancer. To accomplish this requires an understanding of the natural history of HPV infection in females, followed by implementing a vaccine prevention strategy. Males were previously considered only from the perspective of the burden of infection, and how that may relate to transmission events, infection, and disease in their female partners. Although the published literature contains little information on the natural history of infections in males, there has been a significant paradigm shift in the last 5 years in how infection in males is perceived.

The International Agency for Research on Cancer (IARC), which is the cancer group within the World Health Organization (WHO), forms consensus panels that are responsible for making statements about carcinogenicity. In 1995, for example, sufficient evidence existed in the published literature to allow the WHO to make the statement that HPV types 16 and 18 cause cervical cancer.¹ The literature at that time was not sufficient, however, to allow statements about other HPV types or cancer at other sites. Ten years later, in 2005, the IARC convened another consensus panel—the Worldwide Experts on HPV and Disease—which was mandated to determine whether adequate evidence had subsequently been accumulated from which statements could be made about other HPV types in relation to cervical cancer etiology.² As a result of their investigation, 11 additional HPV types were added to the list of oncogenic HPV types associated with cervical cancer.

The panel also determined that there was sufficient published evidence to conclude that HPV, especially HPV-16, causes cancer at other sites. Specifically, they found that in women, in addition to cancer of the cervix, vulvar and vaginal cancer are caused by HPV-16. In men, HPV-16 causes penile cancer, and in both men and women it is associated with anal and oropharyngeal cancer. The incidence of oropharyngeal cancer in males is 5 times that in females. In addition, although the data were not definitive, there was some suspicion in 2005 that HPV possibly causes laryngeal cancer and non-melanoma skin cancer. It is expected that this list will expand further in the next 10 years.

HPV is, therefore, now known to be a cause of cancer in males, and its burden in the United States has been estimated. The American Cancer Society (ACS) makes annual predictions based on population growth and rates of cancer observed in past years. Considering only new cancer cases in men that are believed to be caused by HPV, the ACS expected almost 11,000 cases in 2008 (Table 1).³ This is similar to the CDC cancer statistical data for 2004 that reported 11,892 new cases of cervical cancer in the United States,⁴ although including HPV-related vulvar, vaginal, and anal cancer in women brings the total incidence rate of these HPV-related cancers in females to approximately 14,000. These data show that HPV-related cancers are a significant problem in men as well as in women.

Table 1. Estimate of Annual Number of New Cases of HPV-Related Cancers in US Men: 2008 SOURCE: ACS. American Cancer Society. Cancer facts and figures 2008. Available at: http://www.cancer.org/downloads/STT/2008CAFFfinalsecured.pdf. Kreimer AR, et al. Cancer Epidemiol Biomarkers Prev. 2005;14:467-475. Ryan DP, et al. N Engl J Med. 2000;342:792-800. Daling JR, et al. Int J Cancer. 2005;116:606-616.

Trends in the incidence rates of 2 cancers are particularly interesting. Data from the National Cancer Institute (NCI) Surveillance Epidemiology and End Results (SEER) database revealed a significant increase in the incidence of anal cancer (2.1% annual percentage change [APC]) in both men (2.5% APC) and women (1.9% APC) from 1975 to 2005 (Figure 1).⁵ The increase was greater in the last 10 years of that interval; that is, from 1996 to 2005 the overall APC was 2.6%. The 2001 to 2005 age-adjusted incidence rates per 100,000 population for men (1.4) and women (1.8) reflected a narrowing gap between genders. Significantly, approximately 90% of anal cancers are caused by HPV-16/18.

Figure 1. Age-adjusted Incidence Rate of Primary, Malignant Anal Cancer by Gender and Year of Diagnosis There was a significant increase in the incidence rate of anal cancer in both men and women from 1975 to 2005. Source: Ries LAG, et al. SEER Cancer Statistics Review, 1975-2005, National Cancer Institute. Bethesda, MD.

Another HPV-related cancer undergoing major changes in incidence rate is oropharyngeal cancer. These cancers are frequently associated with smoking and alcohol use; in fact, there has been a significant decrease in smoking-related oropharyngeal cancers in the United States in response to anti-tobacco legislation and policies.⁶ However, HPV-16 and, to a lesser extent, HPV-18, are also responsible for oropharyngeal cancers, and these types are on the increase. Specifically, there was a 20% increase in HPV-related oropharyngeal squamous cell carcinoma among males from 1973 to 2003. This increase, should it continue, suggests that the incidence rate could reach that currently reported for invasive cervical cancer in the United States; that is, approximately 5 to 6 cases per 100,000 population. The base of the tongue and the tonsils have a strong association with HPV; in fact, up to 60% of these cancers are caused by HPV.

In addition to cancer, HPV causes genital warts (condyloma) in both sexes. In fact, 90% of genital warts are caused by HPV types 6 and 11, which are HPV types included in the quadrivalent vaccine. United Kingdom data show a significant rise, similar to cancer incidence rates, in the incidence rate of condyloma over the past 3 decades, increasing between 1971 and 2000 almost 7-fold in males (from 37/100,000 to 257/100,000), and more than 10-fold in females (from 19/100,000 to 195/100,000).⁷ In addition, the incidence rate in males is significantly higher than that in females.

These data provide convincing evidence of a considerable burden posed by HPV-related cancer disease as well as non-cancer disease in males. The facts enhance the importance of directing attention toward reducing infection in males not only because of the role men play in transmitting infection to females, but also because of the disease HPV causes in the males themselves.

Natural History of the Disease in Males

A series of studies is underway in the United States as well as internationally, predominantly in Latin America, endeavoring to eliminate the information gap surrounding the natural history of HPV infection in males.^8,9 When the assumption was that men were only transmitters of the disease, it was also presumed that the natural history of the disease would be different in men than it is in women. Recent epidemiological studies show, however, that HPV behaves similarly in both sexes. Although the 12-month cumulative risk of 0.29 in males is some what smaller than the 0.41 risk in females, the Kaplan-Meier cumulative risk curves do not differ markedly (Figure 2).

Figure 2. Cumulative Risk of Any HPV Infection Among Men and Women The cumulative risk of HPV infection is similar in both sexes. SOURCE: Giuliano AR, et al. J Infect Dis. 2002.186:462-469; Giuliano AR, et al. J Infect Dis. 2008;198:827–835. Reprinted with permission from the Journal of Infectious Diesases.

Figure 3. Duration of Any HPV Infection Among Men and Women Infections are cleared in males at a similar rate as they are in females. SOURCE: Giuliano AR, et al. J Infect Dis. 2002;186:462-469. Giuliano AR, et al. J Infect Dis. 2008;198:827–835. Reprinted with permission from the Journal of Infectious Diesases.

In addition, infections are cleared in males at a similar rate compared with females; that is, approximately 50% of males will clear an HPV infection in 5.9 months, compared with a 9 month median time to clearance for females (Figure 3).^8,9 These risk and clearance data indicate that the natural history of HPV infection does not vary considerably between genders.

Data from an ongoing international study investigating overall and age-specific prevalence of HPV reveal an infection rate of approximately 60% in males, with most infections from non-oncogenic HPV types.⁹ In fact, the prevalence of oncogenic types is very similar to that observed in women. However, unlike in women, the prevalence of oncogenic types in men may increase with age; that is, men are at risk for infection and disease throughout their lifespan.¹⁰ Like women, the presence of antibody to natural infection increases with age in men, providing a marker of total lifetime exposure to HPV.

Safety and Efficacy of the Quadrivalent Vaccine Against External Genital Lesions in Males

An ongoing randomized, double-blind, international study began enrollment in 2004 to evaluate the quadrivalent HPV vaccine in young men for safety, immunogenicity against HPV infection, and efficacy against external genital lesions, which comprised a composite endpoint of both condyloma as well as preneoplastic lesions of the external genital skin, including penile intraepithelial neoplasia. The study is underway in 18 countries in Asia, Europe, the Americas, Australia, and Africa, with multiple sites in each country. Similar to studies in females, this study of males involved 3 doses of vaccine or placebo administered at 0, 2 and 6 months, with a planned 36-month follow-up, which is shorter than what is typically included in trials with females. Two subject groups were enrolled—the primary group comprised approximately 3400 heterosexual men aged 16 to 23 years. The second group included approximately 600 men who have sex with men (MSM) aged 16 to 26 years, who underwent anal canal sampling for both infection and disease.

The per protocol analysis included subjects who received all 3 doses of vaccine, were antibody and polymerase chain reaction (PCR) DNA-negative to the vaccine HPV types at both day 1 and month 7, and had no protocol violations. Endpoints were not evaluated until after the 7-month visit. Of the 4055 subjects enrolled and who received at least one injection, more than 2000 were eligible for per protocol analysis related to HPV-16 and HPV-18, and combined HPV-6 and HPV-11.

In the active vaccine group, 3 cases of external genital lesions (EGL) that were related to a vaccine HPV type developed compared with 31 cases occurring in the placebo arm, a highly significant 90.4% (95% CI: 69.2%, 98.1%; P<0.001) efficacy (Table 2).¹¹

Table 2. Development of External Genital Lesions after Quadrivalent HPV Vaccination in Males * randomized subjects who received at least 1 injection and had follow-up after month 7 † per 100 person years KEY: EGL – extra genital lesions including external genital warts, penile/perianal/perineal intraepithelial neoplasia (PIN), penile, perianal, or perineal cancer, CI – confidence interval SOURCE: Giuliano AR. Data presented at EUROGIN. 2008.

Should Males Be Vaccinated Against HPV?

The data reviewed here indicate that the burden of HPV-related diseases in males is high and that the quadrivalent HPV vaccine is highly efficacious in reducing the incidence of external genital lesions in men aged 16 to 26 years. All data must be considered in decisions to implement male vaccination; that is, decisions must be based on the burden of disease and its epidemiology, which will vary among countries, in addition to safety and efficacy. For example, the incidence of penile cancer in Brazil is 5 times higher

than it is in the United States. In addition, the cost-effectiveness of the vaccine must be determined, for which models are being created that may produce estimates within 6 to 12 months. Finally, the feasibility of a vaccination program must be analyzed. Will it be easier for pediatricians to vaccinate both boys and girls, or will constraints be encountered reaching boys? These aspects must be considered when developing policies related to HPV vaccination in males.

References

1. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans. Volume 64. 1995. Available at: http://monographs.iarc.fr/ENG/Monographs/vol64/index.php.
2. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans. Volume 90. 2005. Available at: http://monographs.iarc.fr/ENG/Monographs/vol90/index.php.
3. ACS. American Cancer Society. Cancer facts and figures 2008. Available at: http://www.cancer.org/downloads/STT/2008CAFFfinalsecured.pdf.
4. CDC. 2008. Cervical Cancer Statistics. Available at: http://www.cdc.gov/cancer/cervical/statistics/.
5. Ries LAG. SEER Cancer Statistics Review, 1975-2005, National Cancer Institute. Bethesda, MD, http://seer.cancer.gov/csr/1975_2005/ posted to the SEER web site 2008.
6. Chaturvedi A, et al. J Clin Oncol. 2008; 26:612-619.
7. CDR Wkly (Online). 2001; 11:35. Available at: http://222.hpa.org,uk/crd/archives/2001/cdr3501.pdf.
8. Giuliano AR, et al. J Infect Dis. 2002;186:462-469.
9. Giuliano AR, et al, J Infect Dis. 2008;198:827–835.
10. Dunne EF, et al. Manuscript in preparation.
11. Giuliano, AR. Presented at EUROGIN. 2008.

HPV Vaccine Safety

William L. Atkinson, MD, MPH

The quadrivalent human papillomavirus (HPV) vaccine contains non-infectious, L1 major capsid protein virus-like particles (VLPs) for 4 HPV types: 2 oncogenic (HPV types 16 and 18) and 2 genital (HPV types 6 and 11). The vaccine is currently licensed for females aged 9 through 26 years and is administered as 3 intramuscular doses over 6 months. More than 20 million doses of the vaccine have been distributed.

The value of the vaccine not only depends on data supporting its efficacy, but also requires evidence of its safety. Clinical trial data from more than 11,778 vaccinated women and a comparison group of 9,686 placebo recipients provided valuable information about the more common adverse events (AEs) that were experienced after vaccination.¹ Detailed safety data were acquired from diary cards kept for 14 days following vaccination by 8878 females aged 9 to 23 years (5088 vaccine, 3790 placebo).

In this post-vaccination interval, local reactions were manifested as pain in 83.9% of subjects receiving the HPV vaccine, swelling in 25.4%, and erythema in 24.7%. Local reactions occurred significantly more frequently in females receiving the HPV vaccination compared with those receiving either an adjuvant or saline placebo (Table 1). Systemic reactions, such as fever, nausea, and malaise, occurred similarly between vaccine and placebo recipients, in fewer than 10% of the subjects.

Table 1. Percentage of Subjects with Adverse Events Following Any Dose of HPV Vaccine among Females 9 to 23 Years Old SOURCE: CDC. MMWR. 2007; 56(RR-2): 12-14.

Although this was a large trial with more than 20,000 subjects, it was nonetheless inadequate to detect rare AEs, which would require millions of doses. This information can be obtained only during post-marketing surveillance. The Vaccine Adverse Event Reporting System (VAERS) is a passive surveillance system operated jointly by the FDA and CDC that receives, analyzes, and reports AE information on licensed vaccines.² However, because the denominator is not known; that is, the total number of persons vaccinated, specific AE rates cannot be determined. Accordingly, the data allows hypotheses to be generated from which detailed investigations may be launched, if warranted. An example of VAERS at work was the identification of a “signal” that suggested an association of intussusception with the first licensed rotavirus vaccine in the late 1990s.

From June 8, 2006, when the quadrivalent HPV vaccine received FDA approval, until the end of August, 2008, VAERS received 10,326 reports of adverse reactions experienced in individuals who received the HPV vaccine, which was approximately 20% of all VAERS reports received during that interval.³ Of these AEs, 94% were considered to be non-serious, and included local reactions, fatigue, and syncope. The latter can, however, be serious.

Reports of syncope after vaccination of young adults and adolescents have been increasing, with the rate almost doubling since 2004. In fact, there have been 463 reports of syncopal episodes to VAERS between 2005 and 2007.² Most of the reports were related to females aged 11 to 18 years, not all of whom received the HPV vaccine. This increased incidence in girls may be related to the fact that they receive more injections than boys. Accordingly, a similar increase of syncope episodes in boys may occur if HPV vaccine is approved for males in the future.

Serious injuries related to syncope have occurred, including 1 death from an intracerebral bleed in a boy aged 12 or 13 years who received the meningococcal conjugate vaccine. Seventy percent of these syncopal episodes occur within 15 minutes of the vaccination; accordingly, adolescents and young adults should remain seated and under observation for 10 to 15 minutes after receiving the vaccine, to facilitate preemption of injuries that might occur as a result of a syncopal episode. When syncope occurs, the patient should be observed until symptoms resolve. Staff should be aware of symptoms that might precede syncope—for example, weakness, dizziness, and pallor—and should be familiar with treatment protocol.

As of August 2008, 27 deaths were reported in females who recently received the HPV vaccine (Table 2).³ A common pattern associated with these deaths was not found, and the cause of death in investigated cases was explained by factors other than the vaccine.

Table 2. Reports of Death Following HPV Vaccination SOURCE: CDC. 2008; Available at: www.cdc.gov/vaccinesafety/vaers/gardasil.htm.

Cases of Guillain-Barré Syndrome have been reported among vaccine recipients, although the rate in patients receiving the HPV vaccine is not higher than what would occur in the specific population in general; that is, about 1 to 2 cases per 100,000 people in their teens. There have also been a number of thromboembolic events reported to VAERS, including deep venous thrombosis and pulmonary emboli. Most of these individuals had other known risk factors for thromboembolic disease, particularly oral contraceptive use. These events are being investigated in detail using a more population-based adverse event reporting system, the Vaccine Safety Data Link. To date, an increased risk for thromboembolic disorders with the HPV vaccine has not been demonstrated.

The data to date support that local reactions occur more commonly in HPV vaccine recipients compared with placebo recipients. Systemic reactions occur in similar frequency following both active and placebo injections. Serious adverse events reported following HPV vaccine appear to be coincidental rather than causal; that is, there is no clear evidence that any of these events were caused by the vaccine. As with all new vaccines, careful monitoring will continue. In conclusion, the existing data indicate that the HPV vaccine provides safe as well as effective protection against major HPV types.

References

1. CDC. MMWR. 2007;56(RR-2):12-14.
2. CDC. MMWR. 2008;57:457-460.
3. CDC. Reports of Health Concerns Following HPV Vaccination. 2008. Available at: www.cdc.gov/vaccinesafety/vaers/gardasil.htm.