Human papillomavirus (HPV) is a common sexually transmitted infection (STI).1,2 Persistent infection with an oncogenic HPV type is a necessary cause for cervical cancer and its immediate precancerous lesions.3–5 Most epidemiologic studies detect prevalent HPV infections using sensitive polymerase chain reaction (PCR) assays on cervical cytologic samples,6 and most HPV infections clear spontaneously, resulting in negative HPV DNA tests. However, an antibody response is elicited in some women, resulting in HPV seropositivity.6
The availability of prophylactic HPV vaccines heightens the importance of understanding the serologic response to HPV. Increasingly, HPV serology is being used as a measure of response to HPV vaccine7,8; however, more information is needed on the advantages and limitations of serologic assessments. Although HPV serology provides information about HPV infection and immune response, it is not clear whether seropositivity as measured with the available assays is a general measure of HPV exposure, current HPV infection and/or immunity.
Not all HPV-infected women develop antibody response to natural infection and some seroconvert at very low antibody levels, often below the assays' detection limit.9–11 Thus, seronegativity measured with the available assays may not imply susceptibility to future infection or nonexposure, and HPV seropositivity in unvaccinated women may not imply full protection. Therefore, it is crucial to rigorously evaluate the utility of HPV seroassays in epidemiologic and clinical studies as they relate to infection, immunity, and susceptibility. In this analysis, we investigated HPV seroepidemiology cross-sectionally among unvaccinated women (18–25 years of age) at the age of peak exposure to HPV.
MATERIALS AND METHODS
Data are from the enrollment, prevaccination phase of the ongoing publicly funded, community-based randomized phase III HPV16/18 vaccine trial in Costa Rica (CVT). The study has been described elsewhere.12,13 Briefly, the main objective of the trial is to evaluate the efficacy of a prophylactic HPV16/18 vaccine to prevent HPV16/18 infection and related precancerous lesions compared to women receiving the control hepatitis A vaccine. Enrollment took place between June 2004 and December 2005. Eligible women were between 18 and 25 years of age; residents of Guanacaste and Puntarenas, Costa Rica; in good general health; and, agreed to use birth control during the vaccination period.
A total of 7466 women provided written, informed consent at enrollment (30% of those prescreened, 59% of those eligible). The enrollment (prevaccination) demographic, behavioral, and sexual risk factor data were obtained from an initial interview administered by trained female interviewers. Participants were asked about their demographics, contraceptive use, reproductive, menstrual and smoking history, and sexual practices, such as lifetime number of sexual partners, and age at first intercourse. Additionally at the enrollment visit, a pelvic exam was performed on consenting sexually experienced women (n = 5871); exfoliated cervical cells were collected for liquid-based cytology (ThinPrep; Cytyc Corporation, Marlborough, MA), and tested for HPV DNA, Chlamydia trachomatis, and Neisseriagonorrhoeae. All testing was done masked to the results of randomization arm, and other test results.
Protocols were approved by the Institutional Review Boards (IRBs) of INCIENSA and the US National Cancer Institute.
HPV DNA Detection
HPV DNA detection and genotyping was performed at DDL Diagnostic Laboratories (Voorburg, Netherlands), as described previously.14,15 Briefly, total DNA was isolated from a PreservCyt aliquot drawn before ThinPrep preparation, DNA was eluted in 100 μL of water, and10 μL of extracted DNA was used for PCR amplification with the SPF10 primer sets, as described earlier.14,15 The samples were run through an HPV DNA enzyme immunoassay (DEIA) to obtain an optical density (OD) reading, and categorized as HPV DNA negative, positive, or borderline. Samples with borderline SPF10-DEIA values were retested with repeat PCR and DEIA.
The same SPF10 amplimers were used on SPF10-DEIA-positive samples to identify HPV genotype by reverse hybridization on a line probe assay (LiPA) (SPF10HPV LiPA25, version 1; Labo Bio-Medical Products, Rijswijk, Netherlands), which detects 25 HPV genotypes as follows: 6, 11, 16, 18, 31, 33, 34, 35, 39, 40, 42, 43, 44, 45, 51, 52, 53, 54, 56, 58, 59, 66, 68/73, 70, 74.
Because CVT uses a bivalent HPV16/18 vaccine, as part of the testing algorithm and to maximize detection for these types, HPV16 and 18 type-specific PCR (TS-PCR) primer sets were used to selectively amplify HPV16 and HPV18 from specimens tested SPF10 DEIA-positive, but LiPA25 HPV16 and/or HPV18 negative.14 Amplimers from the TS-PCRs were detected by DEIA similar to the method used for SPF10 amplimer detection.14–16
To prevent laboratory contamination, reagent preparation, DNA isolation, and PCR and post-PCR analysis were performed in 3 separate, certified, and audited laboratories. Contamination rates are actively measured and recorded, and contamination incidence has been below 1%.
Hybrid Capture 2.
Residual PreservCyt specimens were tested with Hybrid Capture 2 (HC2) (Qiagen, Gaithersburg, MD) at the University of Costa Rica HPV laboratory. HC2 is a commercially available, FDA-approved HPV DNA test which collectively targets 13 oncogenic HPV types: (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68), without distinguishing the HPV type present.15 Values above a positive cutoff point of 1.00 relative light unit per positive control (RLU/pc) are considered HC2-positive. The RLU/pc can be used as a semiquantitative measure of HPV viral burden15 particularly when PCR indicates that only a single genotype is present.
HPV Serological Measurements
Serum collected at enrollment was used to determine HPV16 and HPV18 IgG serostatus at GSK Biologicals (Rixensart, Belgium) using a virus like particle (VLP)-based direct enzyme-linked immunosorbent assay (ELISA) that measures polyclonal antibodies as described previously.17,18 Briefly, ELISA microtiter plates were separately coated with 2.7 μg/mL of either HPV16 or HPV18 VLPs, which were produced in a baculovirus expression system. The plates were blocked with PBS containing 4% skim milk with 0.2% Tween-20. Serum samples from participants were serially diluted in the blocking solution starting at 1:100 in 2-fold increments. Following a washing step, serial dilutions of samples, standard, and controls were added to the microtiter plates. After further wash, a peroxidase-conjugated antihuman polyclonal antibody was added, and excess conjugate removed by washing. Enzyme substrate and chromogen were added and color allowed to develop. Reactions were stopped, and OD read at 450/620 nm. ELISA titers were calculated by averaging the values from all dilutions that fell within the working range of the reference curve relative to the OD readings and expressed as ELISA units (EU)/mL. The antibody results were dichotomized using standard cutoff points as determined by GSK and calculated from antibody titer values 3 standard deviations above the geometric mean titers taken from groups of HPV-negative individuals.17,18 Cut points were OD ≥8 EU/mL for anti-HPV16 and OD ≥7 EU/mL for anti-HPV18.17,18
All analyses were conducted separately for HPV16 and HPV18. We excluded 3 women from the analysis because their age was outside the inclusion age range (one 17 year old and two 26 year olds).
Type-specific HPV seroprevalence was defined as the proportion of women with positive HPV serology for that type, relative to the entire population. The association between DNA positivity and seropositivity was evaluated as the proportion of women seropositive, relative to the number of women DNA positive for the same and discordant HPV type.
To investigate the correlates of HPVseropositivity, the outcome of interest was type-specific HPV16 or 18 serostatus. Covariates chosen for evaluation were from the enrollment medical examination and questionnaires. Univariate and multivariate logistic regression was used to estimate crude and adjusted odds ratios (OR and AOR) and 95% confidence intervals. Variables with P-values <0.10 in univariate models were considered in multivariate models. Variables with P values <0.05 were retained in our multivariate model.
We also evaluated pair-wise interactions and correlations between all statistically significant variables in the multivariate model, and evaluated collinearity between hormonal contraceptive and condom use by examining the correlation coefficient (ρ = 0.25), and performing analysis of seropositivity stratified by hormonal contraception and condom use. We did not observe evidence of collinearity, thus included both in the final model. We also evaluated time since sexual debut and age. Because time since sexual debut was a better predictor of seropositivity in the multivariate model, and because of the high correlation between these variables (ρ = 0.67), we removed age from the multivariate model.
Geometric mean titers are presented for single- and multiple-type HPV DNA infections. We note that because the HPV16 and 18 ELISAs are not identical, titers cannot be directly compared. We further evaluated the correlates of HPV16 and 18 seropositivity (separately) using linear regression models with the log-transformed continuous antibody titer levels. Because results were similar to the dichotomous models, we present only the dichotomous models. SAS version 9.1 was used for all analyses.
The median age was 21 years (n = 7463); 5871 (78.7%) self-identified as nonvirgins. Among sexually active women, the median time since sexual debut was 4 years (interquartile range [IQR], 3–7 years); and the median number of lifetime sexual partners was 2 (IQR, 1–3).
Overall, HPV16 seroprevalence was 26.1%; 30.8% among nonvirgins; 8.5% among virgins. HPV18 seroprevalence was 24.3%; 28.1% among nonvirgins; 10.2% among virgins (Table 1). Among nonvirgins, 50.3% (n = 895/1780) of HPV16 seropositives were HPV18 seropositive; likewise, 55.4% (n = 895/1615) of HPV18 seropositives were HPV16 seropositive. Since a pelvic exam was only performed among sexually experienced participants, cervical DNA analyses presented hereafter are restricted to nonvirgins. Overall HPV DNA prevalence by PCR was 50.1% (n = 2937/5868). HPV16 DNA prevalence was 8.3% (n = 488/5868) and HPV18 was 3.2% (n = 188/5,868). Multiple infections were common, 65.2% (n = 318/488) of HPV16-positives, and 70.7% (n = 133/188) of HPV18-positives were infected with at least another HPV type; 6.2% (n = 30/488) of HPV16 DNA-positives were also coinfected with HPV18.
Of the HPV16 seropositives, 17.1% (n = 305/1779) were HPV16 DNA-positive; 6.5% (n = 105/1614) of HPV18-seropositives were HPV18 DNA-positive. Table 2 presents the association between DNA and serostatus. Among HPV16 DNA-positives, 63.0% (n = 305/484) were HPV16 seropositive; 58.0% (n = 105/181) of HPV18 DNA-positives were HPV18-seropositive. For the discordant pairs, after removing HPV16 DNA-positives 49.4% (n = 76/154) of HPV18 DNA-positives were HPV16 seropositive; after removing HPV18 DNA-positives, 35.5% (n = 159/448) of HPV16 DNA-positives were HPV18 seropositive.
To evaluate the presence of cross-reactivity, we examined HPV16 and HPV18 serostatus among women with a single cervical HPV infection (Table 3). HPV DNA infections were categorized by species: α9 group consisted of HPV31, 33, 35, 52, or 58 (excluding HPV16), α7 group consisted of HPV39, 45, 59, 68, or 70 (excluding HPV18), compared to all others infected with non-α9 and non-α7, respectively. HPV16 seroprevalence was 34.0% among women infected with the α9 group compared to 33.5% among those infected with the non-α9 types (Pearson χ2, P = 0.87). Similarly, HPV18 seroprevalence was comparable among women infected with the α7 types compared to non-α7 types (35.8% vs. 30.6%, respectively, Pearson χ2, P = 0.16).
Figure 1 presents HPV16/18 DNA and seroprevalence by years since sexual debut (1A) and years since sexual debut stratified by lifetime number of sexual partners (1B–1D). Overall and across all lifetime sexual partner categories, seroprevalence of HPV16 and 18 was lowest in women who initiated sexual activity recently, and increased significantly with increasing years since sexual debut. Across all years since sexual debut, HPV16 and 18 seroprevalences were greater than respective DNA prevalence.
Table 4 presents the univariate and multivariate associations of selected factors with HPV16 and 18 seropositivity. After simultaneously adjusting for all variables included in the multivariate model, there was an increase in prevalence of HPV16 seropositivity with lifetime number of sexual partners, number of pregnancies, and HC2/abnormal cytology result. HPV16 seropositive women were significantly more likely to have ever used injectable contraceptives, be current smokers, have had an STI, and be HC2-positive with worse cervical lesions. Women who reported ever using a condom were less likely to be seropositive (compared to never users). Similar to HPV16 seropositives, HPV18 seropositives (Table 4) were significantly more likely to have had a higher number of lifetime sexual partners, a higher viral load/abnormal cytology, and a current and/or past STI. In contrast to HPV16 seropositivity, hormonal contraceptive use, condom use, number of pregnancies, and smoking history were not significantly associated with HPV18 seropositivity.
Table 5 presents HPV16/18 antibody titers. Again, because the HPV16 and 18 ELISAs are not identical, HPV16 and HPV18 titers cannot be directly compared. Among nonvirginal seropositives, HPV16 and 18 geometric mean titers (GMTs) were 41.7 and 24.2, respectively. HPV16 GMTs were similar among women infected with HPV16 and at least another type (multiply infected) compared to women singly infected with HPV16 only (P: 0.52, t test). While there was no statistically significant difference, anti-HPV18 titers were higher in women with a single HPV18 infection than in women infected with HPV18 plus at least one other HPV type (P: 0.17, t test).
In this study of 18- to 25-year-old women in Costa Rica, 30.8% and 28.1% of sexually experienced women were HPV16 and HPV18 seropositive, respectively. In contrast, HPV DNA prevalence was 8.3% for HPV16 and 3.2% for HPV18. Thus, whereas HPV16 and HPV18 seroprevalences were similar, HPV16 DNA prevalence was 2.5 times higher than HPV18 DNA prevalence. A possible explanation for this finding is that the HPV18 ELISA used in this study may be more cross-reactive for genetically related non-HPV18 types than the HPV16 ELISA is for genetically related non-16 types. Our observations, however, that HPV16 seroprevalence was comparable among women infected with a single non-HPV16 in the α9 species and non-α9 species, and that HPV18 seroprevalence was comparable among women infected with a single non-HPV18 in the α7 and non-α7 species suggests no or little evidence of assay cross-reactivity. Alternatively, our findings may also be indicative of higher immune reactivity for HPV18 than HPV16 infections (i.e., HPV18 infections are more likely to result in seroconversion than HPV16 infections) and/or of slower immune decay among those previously infected with HPV18 than HPV16. To better understand assay cross-reactivity and immune seroconversion and decay over time, these findings should be corroborated in longitudinal studies of incident infection.
We found that a large proportion of seropositives were DNA-negative (82% of HPV16 and 93% of HPV18 seropositive), an indication of the transient nature of HPV DNA infection; while this observation suggests seropositivity as a measure of past exposure, it is unknown whether it reflects past exposure at the cervix or other sites (such as the vulva, oropharynx, skin), resulting in HPV antibody production.
Our findings indicate that although serologic-based measures of HPV are a better reflection of cumulative exposure, they are imperfect, as a substantial fraction of HPV DNA positive women are seronegative (37% of cervical HPV16 DNA-positives were HPV16-seronegative; 42% of cervical HPV18 DNA-positives were HPV18-seronegative). This finding highlights that seronegativity does not imply nonexposure currently or in the past. It could also be indicative of misplaced timing of serology measurement, of low DNA viral load, or of low antibody production. Randomized clinical trials of vaccine efficacy and studies evaluating the population level impact of the new prophylactic HPV vaccines should consider that an unknown fraction of individuals classified as naïve by currently available DNA-based and serology tests might in fact be previously exposed/infected.
In concordance with the literature,19–22 the main correlates of seropositivity were a greater amount of time and opportunity for exposure to HPV as evidenced by the association between seropositivity with time since first sex, lifetime sexual partners, and viral burden, highlighting the utility of anti-VLPs as markers of cumulative exposure. Interestingly, we found time since first sex was a better correlate of seropositivity than age, as regardless of age, increasing time since first sex was positively associated with seropositivity. Seroprevalence estimates increased with increasing time since sexual debut and lifetime number of sexual partners, whereas DNA prevalences remained steady. We found seroprevalence estimates of over 50% for women 8 or more years past sexual debut reporting ≥4 partners, similar to literature on high risk women for HPV infection.23
We observed a significant association between HPV16 seropositivity and injectable contraceptive use. There are several possible explanations as follows: (1) enhanced detection of HPV antibodies among injectable users, (2) a potential virus-hormone interaction increasing the likelihood an HPV infected woman will seroconvert, or (3) increased exposure to HPV among injectable users. To ensure that injectable contraceptive use was not a surrogate for unmeasured sexual behaviors and residual confounding, we compared injectable users to nonusers and found they were of similar age, had similar number of lifetime sexual partners, viral load, cytology results, and number of current or past STIs. While this finding strengthens the argument for a biologic mechanism rather than a difference in risk behaviors, we did not have information on partner characteristics and timing of infection. The association between seropositivity and hormonal pathways should be assessed in future studies.
HPV16 and HPV18 seroprevalence estimates from our study (approximately 30%) were higher than recently published studies, where in women of similar ages HPV16 seroprevalences of 3.4% to 24.7% and HPV18 seroprevalences of 1.6% to 12.8% were reported.19,20,24–27 We believe that these differences are in part because of differences across assays and cut points for defining seropositivity and at this time direct comparison across assays may not be possible. The direct ELISA assay measures both neutralizing and nonneutralizing antibodies, and does not provide information on the specific neutralizing epitopes. In contrast, the competitive luminex-based immunoassay measures the presence of antibodies against the HPV16 V5 epitope and HPV18 J4 epitope that are believed to be primary neutralizing epitopes. In addition to assay differences, some of the seroprevalence estimate differences in the literature could be due to differences in study populations.
We found similar anti-HPV16 titers in women with a single HPV16 infection and those infected with multiple types at the cervix. Although there was no statistically significant difference, women with a single HPV18 DNA infection had higher anti-HPV18 titers than women with multiple types at the cervix. The reason for this discrepancy is unclear. There were only 28 and 77 women in these categorizations, and thus we had limited statistical power. Whereas this could be a spurious finding, it warrants further investigation in other settings.
HPV16 and HPV18 seroprevalences in self-reported virginal women (∼10%) was higher than previous assessments in young women and children,19,28,29 reflecting possible assay misclassification or indicative of a cutoff point chosen for seropositivity that is too low. In this analysis, we were unable to assess sexual history of the virgins; thus, the high virginal seroprevalence could also reflect the inherent biases in self-report of virginity, or reflect possible transmission through other nonpenetrative sexual contact, or perinatal or horizontal transmission.19,28,29
A limitation of this study was the cross-sectional nature of the analysis; hence, we had no information on the duration of infection, and may have misplaced timing of DNA and antibody detection. In addition, we did not have information on other immune markers, nor on seropositivity to other HPV types at this time. The main strength of this study was the rich and comprehensive questionnaire and medical history obtained on 7463 young adult women at peak exposure to HPV infection.
In summary, the increasing seroprevalence with time since sexual debut and lifetime number of sexual partners, and our finding that the main correlates of HPV16 and 18 seropositivity were associated with sexual behavior highlights the fact that HPV serology is a cumulative marker of HPV infection regardless of current infected tissue. However, not all HPV16/18 infected women were seropositive in our study, thus HPV serology should be viewed as an imperfect measure of past exposure to HPV, at best. Furthermore, while we found no evidence of cross-reactivity with other HPV types, because of the cross-sectional nature of this analysis there remain unresolved questions about assay specificity and cross-reactivity.
1. Dunne EF, Unger ER, Sternberg M, et al. Prevalence of HPV infection among females in the United States. JAMA 2007; 297:813–819.
2. Cogliano V, Baan R, Straif K, et al. WHO International Agency for Research on Cancer Carcinogenicity of human papillomaviruses. Lancet Oncol 2005; 6:204.
3. Ho GY, Burk RD, Klein S, et al. Persistent genital human papillomavirus infection as a risk factor for persistent cervical dysplasia. J Natl Cancer Inst 1995; 87:1365–1371.
4. Schiffman M, Herrero R, Desalle R, et al. The carcinogenicity of human papillomavirus types reflects viral evolution. Virology 2005; 337:76–84.
5. Bosch FX, Manos MM, Munoz N, et al. Prevalence of human papillomavirus in cervical cancer: A worldwide perspective. J Natl Cancer Inst 1995; 87:796–802.
6. Schiffman M, Herrero R, Hildesheim A, et al. HPV DNA testing in cervical cancer screening: Results from women in a high-risk province in Costa Rica. JAMA 2000; 283:87–93.
7. Block SL, Nolan T, Sattler C, et al. Comparison of the immunogenicity and reactogenicity of a prophylactic quadrivalent human papillomavirus (types 6, 11, 16, and 18) L1 virus-like particle vaccine in male and female adolescents and young adult women. Pediatrics 2006; 118:2135–2145.
8. Pedersen C, Petaja T, Strauss G, et al. Immunization of early adolescent females with human papillomavirus type 16 and 18 L1 virus-like particle vaccine containing AS04 adjuvant. J Adolesc Health 2007; 40:564–571.
9. Stanley M. Immunobiology of HPV and HPV vaccines. Gynecol Oncol 2008; 109:S15–S21.
10. Schwarz TF, Leo O. Immune response to human papillomavirus after prophylactic vaccination with AS04-adjuvanted HPV-16/18 vaccine: Improving upon nature. Gynecol Oncol 2008; 110:S1–S10.
11. Studentsov YY, Schiffman M, Strickler HD, et al. Enhanced enzyme-linked immunosorbent assay for detection of antibodies to virus-like particles of human papillomavirus. J Clin Microbiol 2002; 40:1755–1760.
12. Hildesheim A, Herrero R, Wacholder S, et al. Effect of human papillomavirus 16/18 L1 virus-like particle vaccine among young women with preexisting infection: A randomized trial. JAMA 2007; 298:743–753.
13. Herrero R, Hildesheim A, Rodriquez AC, et al. Rationale and design of a community-based double-blind randomized trial of an HPV 16 and 18 vaccine in Guanacaste, Costa Rica. Vaccine 2008; 26:4795–4808.
14. van Doorn LJ, Molijn A, Kleter B, et al. Highly effective detection of human papillomavirus 16 and 18 DNA by a testing algorithm combining broad-spectrum and type-specific PCR. J Clin Microbiol 2006; 44:3292–3298.
15. Safaeian M, Herrero R, Hildesheim A, et al. Comparison of the SPF10-LiPA system to the hybrid capture 2 assay for detection of carcinogenic human papillomavirus genotypes among 5683 young women in Guanacaste, Costa Rica. J Clin Microbiol 2007; 45:1447–1454.
16. Baay MF, Quint WG, Koudstall J, et al. Comprehensive study of several general and type-specific primer pairs for detection of human papillomavirus DNA by PCR in paraffin-embedded cervical carcinomas. J Clin Microbiol 1996; 34:745–747.
17. Dessy FJ, Giannini SL, Bougelet CA, et al. Correlation between direct ELISA, single epitope-based inhibition ELISA and pseudovirion-based neutralization assay for measuring anti-HPV-16 and anti-HPV-18 antibody response after vaccination with the AS04-adjuvanated HPV16/18 cervical cancer vaccine. Hum Vaccin 2008; 4:425–434.
18. Harper DM, Franco EL, Wheeler C, et al. Efficacy of a bivalent L1 virus-like particle vaccine in prevention of infection with human papillomavirus types 16 and 18 in young women: A randomised controlled trial. Lancet 2004; 364:1757–1765.
19. Clifford GM, Shin HR, Oh JK, et al. Serologic response to oncogenic human papillomavirus types in male and female university students in Busan, South Korea. Cancer Epidemiol Biomarkers Prev 2007; 16:1874–1879.
20. Laukkanen P, Koskela P, Pukkala E, et al. Time trends in incidence and prevalence of human papillomavirus type 6, 11, and 16 infections in Finland. J Gen Virol 2003; 84:2105–2109.
21. Wang SS, Schiffman M, Shields TS, et al. Seroprevalence of human papillomavirus-16, -18, -31, and -45 in a population-based cohort of 10,000 women in Costa Rica. Br J Cancer 2003; 89:1248–1254.
22. Dondog B, Clifford GM, Vaccarella S, et al. Human papillomavirus infection in Ulaanbaatar, Mongolia: A population-based study. Cancer Epidemiol Biomarkers Prev 2008; 17:1731–1738.
23. Touze A, de SanJose S, Coursaget P, et al. Prevalence of anti-human papillomavirus type 16, 18, 31, and 58 virus-like particles in women in the general population and in prostitutes. J Clin Microbiol 2001; 39:4344–4348.
24. Stone KM, Karem KL, Sternberg MR, et al. Seroprevalence of human papillomavirus type 16 infection in the United States. J Infect Dis 2002; 186:1396–1402.
25. Wang IJ, Viscidi R, Hwang KC, et al. Seroprevalence and risk factors of human papillomavirus in Taiwan. J Trop Pediatr 2007; 54:14–18.
26. Paavonen J; the Future II Study Group. Baseline demographic characteristics of subjects enrolled in international quadrivalent HPV (types 6/11/16/18) vaccine clinical trials. Curr Med Res Opin 2008; 24:1623–1634.
27. Naud P, Matos J, Hammes L, et al. Factors predicting intermediate endpoints of cervical cancer and exposure to human papillomavirus (HPV) infections in young women screened as potential targets for prophylactic HPV vaccination in south of Brazil. Eur J Obstet Gynecol Reprod Biol 2006; 124:110–118.
28. af Geijersstam V, Eklund C, Wang Z, et al. A survey of seroprevalence of human papillomavirus types 16, 18, and 33 among children. Int J Cancer 1999; 80:489–493.
29. Marais DJ, Sampson CC, Urban MI, et al. The seroprevalence of IgG antibodies to human papillomavirus (HPV) types HPV-16, HPV-18, and HPV-11 capsid-antigens in mothers and their children. J Med Virol 2007; 79:1370–1374.