Velicer, Christine PHD*; Zhu, Xingshu PHD†; Vuocolo, Scott PHD‡; Liaw, Kai-Li PHD*; Saah, Alfred MD§
Persistent infection of the anogenital tract with human papillomavirus (HPV) is associated with genital warts, intraepithelial neoplasia, and invasive cancers, such as cervical cancer.1–4 Each year, approximately 493,000 new cases of cervical cancer are diagnosed internationally and approximately 273,000 women die from the disease, making it the second most frequent cancer in women and the second leading cause of cancer-related death worldwide among women 15 to 44 years old.5 A quadrivalent HPV (types 6, 11, 16, and 18) L1 virus-like particle vaccine (Merck and Co Inc, Whitehouse Stations, NJ) was approved in 2006 in the United States and elsewhere for use in girls and women between the ages of 9 and 26 years. The vaccine is prophylactic, designed to prevent HPV 6, 11, 16, and 18-related infection and associated genital diseases when administered to females naïve to HPV types contained in the vaccine. Among women previously exposed to some of the vaccine HPV types before vaccination, the vaccination prevents HPV-related infection and disease caused by the types to which the subject is both seronegative and polymerase chain reaction (PCR) negative at the time of vaccination.6 A prophylactic bivalent HPV (types 16 and 18) vaccine is also licensed and available in some countries.
While the prevalence and incidence of HPV genital infections in girls and women up to age 26 has been studied extensively,7–10 the rates of HPV infection and HPV serology status of older women have not been well-characterized in the literature. Women 27 to 45 years old may represent a population who could benefit from an HPV vaccine if they are both susceptible to incident genital HPV infections and at risk of acquiring new infections due to HPV vaccine types. Moreover, measuring HPV infection rates among women up to age 45 can also assist in the development of effective cervical cancer screening strategies and provides baseline data for evaluating changes in population rates of HPV-related anogenital infection following HPV vaccine introduction.
In order to characterize the prevalence and incidence of HPV anogenital infection, this report presents baseline and interim data from a randomized, double-blind, placebo-controlled, multicenter trial of the safety, immunogenicity, and efficacy of the quadrivalent HPV vaccine in women from age 24 to 45 (Protocol 019).11 The objectives were to estimate the prevalence of anogenital infection of HPV types 6, 11, 16, and 18; the seroprevalence of anti-HPV 6, 11, 16, and 18; the incidence of anogenital HPV infection; and risk factors associated with incident infections among women in this age group.
A total of 3819 women were enrolled in the trial between June 18, 2004 and April 30, 2008. Subjects were enrolled from 7 countries (Colombia, France, Germany, Philippines, Spain, Thailand, and the United States) through community and academic health centers and primary health care providers. A total of 3730 women are included in this epidemiologic analysis, of which 1858 were in the placebo arm. Among all women (3819) enrolled in the trial, 2 were excluded from this analysis because they were missing both swab and serology data at baseline, and 87 were excluded because they met the trial’s exclusion criteria after randomization. The reasons for these exclusions were: (1) presence of immune disorder or received immunosuppressives before month 7 trial visit, and (2) received immune globulin or blood products before month 7 visit, or used corticosteroids before month 7 visit. These 87 exclusions were considered appropriate for this epidemiologic analysis in order to focus the generalizability to the women meeting the trial’s inclusion criteria. Serum and anogenital swab samples were collected at day 1 and months 7, 12, 24, 36, and 48. In addition, anogenital swab samples were also collected at months 18, 30, and 42. Anogenital swabs included endo/ectocervical, perineal, and perianal specimens, but not intraanal specimens. Complete health and behavior surveys were collected at each trial visit.
Trial enrollment was open to sexually active women who used effective contraception and had an intact cervix. Women were not enrolled if they were pregnant, had purulent cervicitis, treatment-related cervical surgical procedures, and/or a cervical biopsy within the prior 5 years, or a history of genital warts, vulvar intraepithelial neoplasia, or vaginal intraepithelial neoplasia. Additional enrollment criteria are described elsewhere.11 All participants signed informed consents following review of protocol procedures, and the study was conducted in conformance with applicable country or local requirements regarding ethical committee review, informed consent and other statutes or regulations regarding the protection of the rights and welfare of human subjects participating in biomedical research.
For the detection of HPV, type 6, 11, 16, and 18 DNA, endo- and ectocervical and external genital swab samples were processed using a DNA purification method (Qiagen Technology Kit) and prepared for multiplex PCR, based on real-time fluorescent PCR. This strategy allows the simultaneous detection of 3 gene products (L1, E6, and E7) for a given HPV type in 1 reaction. The HPV type-specific primer pairs (based on the published HPV L1, E6, and E7 sequences), are used to specifically amplify a portion of each gene simultaneously.
Detection of anti-HPV 6, 11, 16, and 18 in the sera of subjects was performed with a competitive Luminex-based immunoassay (cLIA; developed by Merck Research Laboratories using technology from the Luminex Corporation12). Antibody titers were determined in a competitive format in which known, type-specific, phycoerythrin-labeled neutralizing monoclonal antibodies competed with the subject’s serum for binding to conformationally sensitive HPV 6, 11, 16, and 18-specific neutralizing epitopes. Fluorescent signals from this assay are inversely proportional to the subject’s neutralizing antibody titers for a specific HPV type. Seropositivity in the assay is defined as serum anti-HPV levels above the serostatus cut-off for the relevant cLIA. The cutoffs for the HPV-6, -11, -16, and -18 cLIAs are 20, 16, 20, and 24 mMU/mL, respectively.
Prevalence of Anogenital Infection and Seroprevalence.
The prevalence of HPV anogenital infection at baseline in the placebo and vaccine arms of the trail was based on detection of HPV DNA in cervical and/or external anogenital swabs, as measured by the type-specific HPV 6, HPV 11, HPV 16, and HPV 18 PCR assay. For women who received the quadrivalent vaccine, the baseline swabs were included in the analysis if they were collected before administration of the first dose of vaccine. These anogenital swabs do not represent cumulative HPV exposure, but rather, a single-cross-sectional measure of HPV prevalence. Seroprevalence was based on detection of anti-HPV 6, 11, 16, and 18 in sera using assays described above.
Incidence of HPV Anogenital Infection.
The incidence rates (per 100 person-years) and cumulative incidence rates of HPV incident and persistent incident infections were calculated among 1827 women in the placebo arm who were negative for HPV at baseline (i.e., negative HPV results for serology, cervical, and external genital swab samples for at least 1 HPV type). Unlike the baseline (prevaccination) prevalence calculations that included the placebo and vaccine arms, as described above, only the placebo arm was used in prospective analyses, because inclusion of vaccine recipients would have resulted in incidence rates not representative of the natural history of HPV infection in the absence of vaccination. An incident infection for a given HPV type was defined as detection of HPV DNA in cervical or external genital swabs (i.e., anogenital swabs) collected on at least 1 postbaseline visit. A persistent infection was defined as detection of HPV DNA on cervicovaginal specimens collected on at least 2 consecutive postbaseline visits. There was no minimum time requirement between these visits, though visits were typically 6 months apart, with a range of 4 to 8 months. The rates of recurrent detection and recurrent persistent detection were also calculated in a similar manner as incident and incident persistent infections, except the starting population was women seropositive but PCR negative at baseline for the relevant HPV types.
Person-time calculations started at the date of the negative baseline exam and continued until the date of the incident or persistent infection. Censoring occurred at the last follow-up visit available for each participant. Occasionally, there was missing data for a particular exam date. If there was more than one consecutive missing clinic exam in which swab data should have been available, the censor date was the last exam with swab data available before the consecutive missing exams. If a woman had missing swab data for a single exam date and the missing exam was followed by a negative swab result at the next exam, the missing exam swab was assumed to be negative for the given HPV type. If a woman had missing swab data for a single exam date, followed by a positive result at the next exam, the missing exam swab was assumed to be positive. The date of the missing exam was assumed to be the midpoint between the 2 exam dates with available data.
All cumulative incidence estimates were obtained from a Kaplan-Meier function. Less than 50% of study participants had follow-up time beyond month 30 of the trial. Therefore, cumulative risk was estimated up to month 30. Confidence intervals for incidence rates were calculated using a binomial exact distribution.
Risk Factors for Incident HPV DNA Infection.
Among subjects who had negative swabs and serology at baseline, the risk of a single incident HPV type 6, 11, 16, and/or 18 (hereafter referred to as “6/11/16/18”) anogenital infection (vs. no incident HPV infection of any vaccine HPV type) associated with participant baseline characteristics was assessed via age-adjusted hazard ratios from Cox proportional hazards models. The baseline characteristics considered were: age, marital status, age at first intercourse, lifetime number of sex partners, number of new sex partners in last 6 months, residency in the United States, Chlamydia trachomatis and/or Neisseria gonorrhoeae infection, number of prior pregnancies, and Pap test results. Presence of C. trachomatis and/or N. gonorrhoeae infection at baseline was assessed using the Roche multiplex AMPLICOR C. trachomatis/N. gonorrhoeae PCR test from the liquid-based Thin Prep Pap fluid collected at baseline for each woman.
Prevalence of Anogenital Infection at Baseline
At baseline, anogenital infection with HPV 6/11/16/18 (i.e., 6, 11, 16, and/or 18) was detected in 7.8% of study subjects, 2.1% for types 6 and/or 11, and 6.1% for 16 and/or 18 (Table 1). The prevalence for types 6, 11, 16, and 18 were 1.9%, 0.2%, 4.4%, and 2.1%, respectively. None of the trial participants in this analysis was infected with all 4 vaccine HPV types (6, 11, 16, and 18) at baseline. Prevalence varied by country, with the highest overall prevalence for any HPV type noted in France (16.0%) and Germany (13.7%). The lowest prevalence was found in the Philippines (2.8%) and Thailand (2.9%). The prevalence of anogenital HPV 6/11/16/18 infection in the overall study population was highest among 24- to 29-year-old women (Fig. 1A).
Seroprevalence at Baseline
At baseline, 29.7% of the combined study population was anti-HPV 6/11/16/18 seropositive, whereas 17.5% was anti-HPV 6 and/or anti-HPV 11 seropositive, and 18.3% was anti-HPV 16 and/or anti-HPV 18 seropositive (Table 1). Seroprevalence for types 6, 11, 16, and 18 were 14.9%, 4.9%, 14.8%, and 5.4%, respectively. Seroprevalence varied by country, with the highest overall seroprevalence noted in the United States (47.5%) and France (36.0%) and lowest in the Philippines (17.7%) and Thailand (23.3%). The seroprevalence in the study population was generally stable across the age range for any of the HPV types measured (Fig. 1B).
Combined Anogenital Infection Prevalence and/or Seroprevalence
The combined prevalence of any HPV 6/11/16/18 infection or seropositivity in the study population at baseline was 32.8% (Table 2 and Fig. 1C). Prevalence of infection and/or seropositivity was 18.5% for types 6 and/or 11 and 21.5% for 16 and/or 18. Of those subjects who were HPV 6/11/16/18 DNA and/or seropositive, the majority (23.2%) were positive for exactly 1 of these types and only 0.4% were positive to all 4 types. The combined HPV DNA and/or seroprevalence was highest among 24 to 29 year olds (Fig. 1C), though these estimates were relatively stable across age groups.
Association Between Infection Prevalence and Pap Test Results
Women who had prevalent anogenital HPV infections at baseline were also more likely to have abnormal Pap results (data not shown). In particular, women with the most severe cytologic atypia as determined by Pap testing had the highest HPV prevalence. For example, approximately 71% of women with high-grade squamous intraepithelial lesion (HSIL) at baseline also had an HPV 16 or 18 anogenital infection, whereas only 4.8% of women with a normal Pap test result had an HPV 16 or 18 infection.
Incidence of Anogenital Infection
Incidence rates (per 100 person-years) of HPV 6/11/16/18 anogenital infections were estimated for 4 subpopulations (as described in the methods section) within the group who received placebo. Among women who were seronegative and PCR negative to HPV 6/11/16/18 at baseline, the rate of persistent infections were roughly one-half the rate of incident infections (Table 3 and Fig. 2). The highest rates of incident and persistent infections were observed among 24- to 34-years-old women. The rates of recurrent detection (among women seropositive but PCR negative at baseline) were higher in the older age stratum (35–45 year old women). In women over the age of 35, the rates of recurrent detection equaled or surpassed the rates of incident infections, though confidence intervals for recurrent detection were wide and overlapped with incident infections. In general, rates were higher for HPV 16 and/or 18, infections compared with HPV 6 and/or 11 infections.
Cumulative rates of incident and persistent HPV infection among women who were seronegative and swab negative in the placebo arm at baseline are shown in Figures 3A, B. By month 30 of the trial, ∼10.5% of all HPV naïve women in the placebo group (i.e., women who were seronegative and swab negative to a specific HPV type at baseline; these women could have been positive to another type) developed an incident HPV 6/11/16/18 infection. Approximately, 6.7% developed an incident HPV 16/18 infection and ∼4.5% developed an incident HPV 6/11 infection. By month 30 of the trial, ∼5.1% of women who were HPV naïve at baseline had developed an HPV 6/11/16/18 persistent infection, ∼3.2% developed a persistent HPV 16/18 infection, and ∼2.1% developed a persistent HPV 6/11 infection.
Risk Factors for Incident HPV Anogenital Infection
Among HPV naïve women at baseline, compared to those who did not develop incident infections for any HPV type, women who developed an incident anogenital HPV 6/11/16/18 anogenital infection were more likely at baseline to be younger (mean age: 31.2 vs. 34.6 years). In addition, after adjusting for age, they were more likely at baseline to be to be single or not in their first marriage, to have a higher number of lifetime sex partners, to have a higher number of recent sexual partners (in the 6-month period before baseline), and to be infected with C. trachomatis or N. gonorrhoeae (Table 4). No association was noted between age at first intercourse, US residency, and number of prior pregnancies and the risk of incident HPV 6/11/16/18 infection. The risks of HPV 6/11 infections or HPV 16/18 infections associated with baseline characteristics were comparable to those found for 6/11/16/18 infections (data not shown).
This report of HPV type 6/11/16/18 infection prevalence, incidence, and risk factors for incident infection among women between the ages of 24 and 45 years has 3 main findings. First, a large proportion of women in this age group have a prevalent anogenital infection of at least 1 HPV type and/or are seropositive (∼32.8%) but only (0.4%) had evidence of past or current infection with all 4 HPV types tested. Second, women between the ages of 24 and 45 are at risk for acquiring anogenital HPV infections, as demonstrated by a rate of 10.5% for incident infections and ∼5% for persistent infections over a 30-month period among women seronegative and PCR negative at baseline to the relevant HPV types. Third, factors associated with an increased risk of acquiring an incident HPV infection among women 24 to 45 years old include younger age, marital status other than being in a first marriage, and sex behaviors (such as number of recent sex partners).
Infection Prevalence and Seroprevalence
As far as is known, this is the first study in this age group to combine cross-sectional seroprevalence and prevalence by PCR to estimate the baseline population at risk for an HPV infection of HPV types 6/11/16/18. Virtually, all women (>99%) were naïve to at least one of these HPV types and 77% were naive to 2 or more types via swab and/or serology at enrollment. This suggests that the majority of women are susceptible to at least 1 HPV type within the quadrivalent vaccine. It is also possible that women could have been exposed to HPV vaccine types but cleared their infections. Given that more than 99% of women did not have evidence of HPV infection from all 4 types, even with some degree of misclassification of women as “HPV naïve” who were actually previously infected, these data suggest that many women in this age group are at risk for infections with at least 1 of the 4 HPV types in the quadrivalent vaccine.
Specific to anogenital infection prevalence estimates, the measurement of HPV DNA data at baseline represents a single, cross-sectional measure of HPV prevalence, rather than a measure of cumulative exposure. Also, in comparison to other studies of infection prevalence, few other studies have reported results for HPV types 6, 11, 16, and 18 among women 24 to 45 years old. Among studies that reported combined prevalence estimates for types 16/18, our findings (6.1%) were consistent with the range of 4.5% to 8.6% reported by others.13–15 Among studies that reported prevalence for HPV16 separate from HPV18, our findings for HPV16 (4.4%) were within the range in the literature, and our findings for HPV18 (2.1%) were at the upper end of the range found in the literature (0.3%–2.0%).16–21 In comparison to the 1 study, we found that reported HPV6 and 11 prevalence estimates in this age range, our estimate for HPV11 (0.2%) was similar, but our estimates for HPV6 (1.9%) was higher than the prevalence range for HPV6 report in that study (0.3%–0.8%).17
Anti-HPV 6/11/16/18 seroprevalence data are also scarce in women 24 to 45 years old. An Australian study, using the Merck cLIA, reported a seroprevalence range among 30- to 49-year-old women of 26% to 27% (HPV6 or 11), 28% to 30% (HPV16 or 18), and 40% to 41% (types 6/11/16/18).22 These ranges are higher than our aggregate findings but consistent with our findings for the United States and France, possibly indicative of similar behaviors associated with acquisition of infection among women in these countries. Among studies that did not use the Merck cLIA, only 1 study reported HPV6/11 seroprevalence estimates in this age range.23 In a study of Finnish women between the ages of 26 and 31, HPV6/11 seropositivity was 10% to 12%.23 Seroprevalence data from the current report are modestly higher – at 17.5% for HPV6/11. These estimates of seroprevalence for types 16 and 18 were within the range reported in the literature.23–25
For HPV6/11 and 16/18, the 30-month cumulative incidence was 4.5% and 6.7%, respectively. One other large-scale study has reported type-specific incidence rates of HPV anogenital infection among women who are 24 to 45 years old. In that study conducted in Colombia, incidence rates were measured over a 5-year period among 1610 females 15 to 85 years old (unspecified number 24–45 years old) who were HPV negative and had normal cervical cytology results at baseline.26 The 3-year cumulative incidence rates were: 0% to 0.5% (HPV6), 0% to 0.4% (HPV11), 2.1% to 3.5% (HPV16), and 1.5% to 2.8% (HPV18). The 5-year cumulative incidence rates were: 0% to 0.5% (HPV6), 0.4% to 0.7% (HPV11), 4.2% to 5.2% (HPV16), and 2.0% to 3.9% (HPV18). The lower end of these ranges was among 30- to 44-year-old women and the higher end was among 24 to 29 year olds. Our cumulative incidence rates by month 30 for types 6/11 are higher than those reported in the Colombian study, whereas our findings for types 16/18 are similar to the 3-year rates in the Colombian study (assuming women in the Colombian study had minimal overlap in infection with both types 16 and 18, as found in this study). Type-specific incidence density rates (person-years) were not stratified by age in the Colombian study and thus could not be compared to our rates. The Colombian study also reported higher incidence rates at the lower end of the age range of 24 to 45 years. Nonetheless, incident HPV infections, particularly for the high-risk (cancer-causing) types 16 and 18, were reported in women at the upper end of this age range in both studies, demonstrating that women across this range are not only susceptible to, but are also acquiring HPV infections.
The rates of incident and incident persistent infections were estimated among women in the placebo arm of the trial who were seronegative and PCR negative at baseline to the relevant HPV types (i.e., “HPV naïve” women). Because seroconversion occurs in only 50% to 70% of HPV infections that clear,27 it is likely that some women identified as HPV naïve for these incidence estimates would have had some prior HPV exposure, and as such would have been more appropriately classified in the “recurrent detection” subpopulations. The potential impact of this factor on the incidence rates and the rates of recurrent detection is unclear.
Younger age within the range of 24 to 45 year olds emerged as a strong determinant for acquiring incident HPV infections. Independent of age, the number of lifetime and recent sex partners, marital status, and infection with C. trachomatis and/or N. gonorrhoeae also emerged as risk factors for incident HPV infections. These factors are consistent with prior studies that were not specific to the 24- to 45-year-old age range, as summarized by Burchell et al.28 These determinants also illustrate how women could remain at risk for acquiring HPV infection and related diseases throughout their lives. Changes in sexual behavior during the last 30 years, characterized by rising age at first marriage and an increase in divorce rates, have led to more widespread premarital sex and acquisition of new sex partners around middle age.29 Literature suggests that in the United States, nearly 40% of men and women have married and divorced by age 55, and that over 25% of these people have remarried at least once.30
There are several limitations to this study. First, as previously mentioned, seropositivity may underestimate the cumulative exposure to the relevant vaccine HPV types. Also, the incidence data could vary by country, but the sample sizes did not permit country-specific estimates. Regional variation was observed in prevalence (2.8% in Philippines to 16.0% in France) and seroprevalence (17.7% in Philippines to 47.5% in the United States). It is also possible that the data generated from participants in a clinical trial may differ from an observational study. However, as discussed above, the infection prevalence, seroprevalence, and infection incidence rates estimated from this study are largely consistent with the available literature, suggesting that this may not be a large source of bias. The observation that women 35 to 45 years appear to have an increase in recurrent detection (in comparison to women 24–34) is intriguing but sample sizes were small and confidence intervals overlapped. Future efforts to understand the natural history of recurrent and reactivated HPV infections and their relationship to associated genital disease among women over the age of 24 would be helpful, as would robust, region-specific HPV incidence rates of infection, and associated disease.
The age-stratified, regional, and type-specific estimates of HPV anogenital infection provided in this report will be useful in evaluating the potential benefits of prophylactic HPV vaccines. It is important to establish a baseline understanding of genital HPV prevalence and incidence rates before the establishment of vaccination programs. These HPV infection metrics can assist in the modeling of effective cervical cancer screening strategies and as a reasonable baseline for evaluating changes in the population rates of HPV-related genital infections and diseases following the introduction of prophylactic HPV vaccines.
This study represents the largest and most comprehensive epidemiologic analysis of anogenital infection, seroprevalence, and risk factors for infection of HPV types 6, 11, 16, and 18 specifically among women 24 to 45 years old. The findings demonstrate that women in this age range are susceptible to HPV infections, and some acquire infections with these types.
1. Daling JR, Madeleine MM, Schwartz SM, et al. A population-based study of squamous cell vaginal cancer: HPV and cofactors. Gynecol Oncol 2002; 84:263–270.
2. Ferenczy A, Mitao M, Nagai N, et al. Latent papillomavirus and recurring genital warts. N Engl J Med 1985; 313:784–788.
3. Goffin F, Mayrand MH, Gauthier P, et al. High-risk human papillomavirus infection of the genital tract of women with a previous history or current high-grade vulvar intraepithelial neoplasia. J Med Virol 2006; 78:814–819.
4. Melbye M, Frisch M. The role of human papillomaviruses in anogenital cancers. Semin Cancer Biol 1998; 8:307–313.
5. Ferlay J, Bray F, Pisani P, et al. GLOBOCAN 2002: Cancer incidence, mortality and prevalence worldwide. IARC Cancer Base No 5 Version 2 0 IARC Press Lyon, France; 2002. Available at: http://www-dep.iarc.fr/globocan/database.htm
. Accessed October 15, 2008.
6. The FUTURE II Study Group. Prophylactic efficacy of a quadrivalent human papillomavirus (HPV) vaccine in women with virologic evidence of HPV infection. J Infect Dis 2007; 196:1438–1446.
7. Auvinen E, Niemi M, Malm C, et al. High prevalence of HPV among female students in Finland. Scand J Infect Dis 2005; 37:873–876.
8. Brown DR, Shew ML, Qadadri B, et al. A longitudinal study of genital human papillomavirus infection in a cohort of closely followed adolescent women. J Infect Dis 2005; 191:182–192.
9. Dillner J, Kallings I, Brihmer C, et al. Seropositivities to human papillomavirus types 16, 18, or 33 capsids and to Chlamydia trachomatis are markers of sexual behavior. J Infect Dis 1996;173:1394–1398.
10. Manhart LE, Holmes KK, Koutsky LA, et al. Human papillomavirus infection among sexually active young women in the United States: Implications for developing a vaccination strategy. Sex Transm Dis 2006; 33:502–508.
11. Munoz N, Manalastas R, Pitisuttihum P, et al. Safety, Immunogenicity, and efficacy of quadrivalent HPV (types 6, 11, 16, 18) recombinant vaccine in adult women between 24 and 45 years of age. Lancet. In press.
12. Opalka D, Lachman CE, MacMullen SA, et al. Simultaneous quantitation of antibodies to neutralizing epitopes on virus-like particles for human papillomavirus types 6, 11, 16 and 18 by a multiplexed luminex assay. Clin Diagn Lab Immunol 2003; 10:108–115.
13. Duttagupta C, Sengupta S, Roy M, et al. Are Muslim women less susceptible to oncogenic human papillomavirus infection? A study from rural eastern India Int J Gynecol Cancer 2004; 14:293–303.
14. Laikangbam P, Sengupta S, Bhattacharya P, et al. A comparative profile of the prevalence and age distribution of human papillomavirus type 16/18 infections among three states of India with focus on northeast India. Int J Gynecol Cancer 2007; 17:107–117.
15. Li LK, Dai M, Clifford GM, et al. Human papillomavirus infection in Shenyang City, People’s Republic of China: A population-based study. Br J Cancer 2006; 95:1593–1597.
16. Forslund O, Antonsson A, Edlund K, et al. Population-based type-specific prevalence of high-risk human papillomavirus infection in middle-aged Swedish women. J Med Virol 2002; 66:535–541.
17. Herrero R, Castle PE, Schiffman M, et al. Epidemiologic profile of type-specific human papillomavirus infection and cervical neoplasia in Guanacaste, Costa Rica. J Infect Dis 2005; 191:1796–1807.
18. Kitchener HC, Almonte M, Wheeler P, et al. HPV testing in routine cervical screening: Cross sectional data from the ARTISTIC trial. Br J Cancer 2006; 95:56–61.
19. Klug SJ, Hukelmann M, Hollwitz B, et al. Prevalence of human papillomavirus types in women screened by cytology in Germany. J Med Virol 2007; 79:616–625.
20. Peto J, Gilham C, Deacon J, et al. Cervical HPV infectionand neoplasia in a large population-based prospective study: The Manchester cohort. Br J Cancer 2004; 91:942–953.
21. Ylitalo N, Sørensen P, Josefsson AM, et al. Consistent high viral load of human papillomavirus 16 and risk of cervical carcinoma in situ: A nested case-control study. Lancet 2000; 355:2194–2198.
22. Newall A, Beutels P, Wood JG, et al. Cost-effectiveness analyses of human papillomavirus vaccination. Lancet Infect Dis 2007; 7:289–296.
23. Lehtinen M, Kaasila M, Pasanen K, et al. Seroprevalence atlas of infections with oncogenic and non-oncogenic human papillomaviruses in Finland in the 1980s and 1990s. Int J Cancer 2006; 119:2612–2619.
24. Chen CJ, Viscidi RP, Chuang CH, et al. Seroprevalence of human papillomavirus types 16 and 18 in the general population in Taiwan: Implication for optimal age of human papillomavirus vaccination. J Clin Virol 2007; 38:126–130.
25. Wang SS, Schiffman M, Shields TS, et al. Seroprevalence of human papillomavirus-16, -18, -31, and -45 in a population-based cohort of 10000 women in Costa Rica. Br J Cancer 2003; 89:1248–1254.
26. Munoz N, Mendez F, Posso H, et al. Incidence, duration, and determinants of cervical human papillomavirus infection in a cohort of Colombian women with normal cytological results. J Infect Dis 2004; 15:2077–2087.
27. Dillner J. The serological response to papillomaviruses. Semin Cancer Biol 1999; 9:423–430.
28. Burchell AN, Winer RL, de Sanjose S, et al. Chapter 6: Epidemiology and transmission dynamics of genital HPV infection. Vaccine 2006; 24(suppl 3):S52–S61.
29. Wellings K, Collumbien M, Slaymaker E, et al. Sexual behaviour in context: A global perspective. Lancet 2006; 368:1706–1728.
30. Stevenson B, Wolfers J. Marriage and divorce: Changes and their driving forces. J Econ Perspect 2006; 21:27–52.