Seroprevalence of Human Papillomavirus Type 16 in Children

    Division of STD Prevention, National Center for HIV, STD, and TB Prevention
    Division of Viral and Rickettsial Diseases, National Center for Infectious Diseases, Centers for Disease Control and Prevention, Atlanta, Georgia

    Seroprevalence of Human Papillomavirus Type 16 in Children

    Eileen F. Dunne,1 Kevin L. Karem,2 Maya R. Sternberg,1 Katherine M. Stone,1 Elizabeth R. Unger,2 William C. Reeves,2 and Lauri E. Markowitz1

    1Division of STD Prevention, National Center for HIV, STD, and TB Prevention, and 2Division of Viral and Rickettsial Diseases, National Center for Infectious Diseases, Centers for Disease Control and Prevention, Atlanta, Georgia

    Genital human papillomavirus (HPV) infection is the most common sexually transmitted infection in the United States [1]. More than 30 HPV types infect the genital area, and persistent infection with certain genital types is the most important risk factor for cervical cancer [2]. More than 80% of the US adult population has been infected with HPV by the age of 50 years [3], and most infections are asymptomatic [4].

    Genital HPV infection occurs primarily via sexual contact, and detection of genital types in children is relatively uncommon. Infection with these types is usually asymptomatic in children; rarely, infection with HPV-6 or -11 is associated with anogenital warts or recurrent respiratory papillomatosis. Possible modes of transmission of genital types in children include vertical transmission, auto- and heteroinoculation, sexual abuse, and fomites [5]. In a prospective cohort study that followed mothers and infants, vertical transmission was found to be rare [6].

    Of the genital types, HPV-16 is the most common and is the cause of >50% of cervical cancers in the United States [7]. Detection of HPV-16 DNA is often transient; therefore, it is not a useful marker of long-term exposure or of immune response to infection. Detection of antibodies to HPV-16 virus-like particles (VLPs) provides a measure of both cumulative exposure and immune response to HPV-16. Although many persons infected with HPV-16 do not develop antibodies [8], seroprevalence studies provide some measure of cumulative exposure.

    A recent study found that HPV-16 seroprevalence was 13% in a representative sample of persons 1259 years of age in the United States, with seroprevalence higher in women than in men [9]. The seroprevalence of HPV-16 has not been characterized in US children. Here, we describe the seroprevalence of HPV-16 in a representative sample of children 611 years of age in the United States.

    Participants, materials, and methods.

    The National Health and Nutrition Examination Survey (NHANES), conducted by the National Center for Health Statistics, used representative samples from noninstitutionalized US civilian populations. The third NHANES (NHANES III) was conducted from 1988 to 1994, in two 3-year phases. Prevalence estimates were weighted to represent the total US population and to account for oversampling and nonresponse. The poverty index was calculated by dividing total family income by the poverty threshold, as defined by the US Census, with adjustment for family size at the time of the interview. Urban residence was defined as residence in a county in a metropolitan area; residence in all other counties was defined as rural. Eighty-six percent of the children interviewed provided serum samples for serologic testing.

    We tested for HPV-16 IgG antibodies in serum samples obtained from participants during the second phase (19911994) of NHANES III. VLPs were produced by expression of a recombinant HPV-16 L1 baculovirus in insect cell tissue culture. ELISAs using VLPs were standardized and were performed as described elsewhere [9, 10]. For quality control and evaluation, we used known positive and negative human serum samples (provided by M. Hagensee, Louisiana State University, Baton Rouge, and H. Strickler, National Cancer Institute, Bethesda, MD [present affiliation: Albert Einstein College of Medicine, the Bronx, NY]). Additional negative serum samples were obtained from Egleston Children's Hospital (Atlanta, GA). Control serum samples were also used to prepare pools of high-positive, low-positive, and negative controls. Quality-control assessment included monitor of day-to-day and plate-to-plate variation. Results for individual control serum samples that were run throughout the course of the NHANES III analysis were used in receiver operating characteristic (ROC) analysis, to determine the cutoff value for discrimination between positive and negative serum samples. Evaluation of ROC plots indicated that an OD cutoff value of 0.64 resulted in 93.0% sensitivity and 98.5% specificity. Data were analyzed by use of SUDAAN software (version 7; Research Triangle Institute). We evaluated bivariate associations between HPV seropositivity and selected characteristics. Confidence intervals (CIs) for the prevalence estimates were calculated on the basis of a log transformation, with SEs calculated by the delta method [11].

    Results.

    Overall, 2.4% of 1316 children 611 years of age had antibodies to HPV-16 VLPs. The median OD of the samples positive by ELISA was 0.84 (range, 0.652.14). Prevalence was higher in boys than in girls (3.5% vs 1.2%), but this difference was not statistically significant (P = .08).

    None of the demographic variables tested for, including race/ethnicity, socioeconomic status, urban or rural residence, region of the country (data not shown), and whether the child was born outside of the United States (data not shown), were significantly associated with HPV-16 seropositivity (table 1). Age was significantly associated with higher prevalences of antibodies to HPV-16 VLPs. A higher prevalence of antibodies to HPV-16 was found in children >7 years of age than in children 7 years of age (3.3% vs. 0.4%; P < .05). There was a statistically significant linear trend of increasing seroprevalence in girls, but not in boys, as they became older (data not shown).

    Conclusions.

    The present study is the largest population-based sample to provide an estimate of HPV-16 seroprevalence in children. Overall seropositivity in children 611 years of age was low, indicating little exposure to HPV-16 before adolescence. Age >7 years was associated with a significantly higher prevalence of antibodies to HPV-16, suggesting some increasing exposure to HPV-16 with increasing age. Among NHANES III participants, seropositivity in those 1259 years of age who had never had sex was 5.0%, and seropositivity in children 1213 years of age was 2.9%.

    In 7 previous studies conducted in other countries, estimates of HPV-16 seroprevalence in children ranged from 0% to 7.6%; in most studies, it was 3.0% (table 2) [1218]. All but 2 of these studies included serum samples from a convenience sample of children. Many of the studies found low levels of antibodies. In a study of >1000 children <13 years of age, most of the optical-density values were at or near the cutoff for positivity (OD, 0.100) [16], indicating a low level of seropositivity, similar to our results. Cubie et al. reported that 7.6% of children were seropositive and that 97.0% of these children had optical-density values close to the cutoff [15].

    In only 1 study could we identify differences in seroprevalence with age; an increased seroprevalence was noted with younger age, unlike in our study [19]. In 1 study, higher seroprevalence was found in boys than in girls [19]. In a prospective study of antibody to HPV-16 in children, 3% developed antibodies between 1 and 2 years of age [17]. However, infants born to women who were positive for antibody to HPV-16 were not at significantly higher risk than were those born to women who were negative, suggesting that the rate of vertical transmission is low.

    The present data are subject to limitations. Because the relative SEs for many of the comparisons in our analysis were high, point estimates should be interpreted with caution, and more emphasis should be placed on the CIs. Comparisons of the results of serologic studies of HPV-16 have limitations because of the differences in methods used. Despite these differences, seroprevalence estimates in children have been similar across studies. Although methods used for HPV-16 VLP serologic testing are type specific [10, 20], it is not possible to totally exclude nonspecific reactivity. To optimize specificity without compromising sensitivity, our serologic method used ROC analysis of results for interlaboratory control serum samples to determine the cutoff value for positivity [9, 10].

    Prophylactic HPV vaccines, including vaccines for HPV-16, are currently in clinical trials. Data on HPV-16 seroprevalence in the US population is useful before the introduction of prophylactic HPV vaccines. Our data suggest that little exposure to HPV-16 occurs before adolescence. The available literature on HPV-16 seroprevalence does not have information on behaviors that would help to characterize potential exposures to HPV in this age group. To explain HPV-16 seropositivity in this population, further study is required.

    Acknowledgments

    We thank Alicia Poon (now Alicia Cordell) and Geraldine M. McQuillan, for their work on this project.

    References

    1.  Weinstock H, Berman S, Cates W Jr. Sexually transmitted diseases among American youth: incidence and prevalence estimates, 2000. Perspect Sex Reprod Health 2004; 36:610. First citation in article

    2.  Ho GY, Bierman R, Beardsley L, Chang CJ, Burk RD. Natural history of cervicovaginal papillomavirus infection in young women. N Engl J Med 1998; 338:4238. First citation in article

    3.  Myers ER, McCrory DC, Nanda K, Bastian L, Matchar DB. Mathematical model for the natural history of human papillomavirus infection and cervical carcinogenesis. Am J Epidemiol 2000; 151:115871. First citation in article

    4.  Koutsky L. Epidemiology of genital human papillomavirus infection. Amer J Med 1997; 102:38. First citation in article

    5.  Gutman LT, Herman-Giddens ME, Phelps WC. Transmission of human genital papillomavirus disease: comparison of data from adults and children. Pediatrics 1993; 91:318. First citation in article

    6.  Watts DH, Koutsky LA, Holmes KK, et al. Low risk of perinatal transmission of human papillomavirus: results from a prospective cohort study. Am J Obstet Gynecol 1998; 178:36573. First citation in article

    7.  Bosch FX, Manos MM, Munoz N, et al. Prevalence of human papillomavirus in cervical cancer: a worldwide perspective. International Biological Study on Cervical Cancer (IBSCC) Study Group. J Natl Cancer Inst 1995; 87:796802. First citation in article

    8.  Carter JJ, Koutsky LA, Hughes JP, et al. Comparison of human papillomavirus types 16, 18, and 6 capsid antibody responses following incident infection. J Infect Dis 2000; 181:19119. First citation in article

    9.  Stone KM, Karem KL, Sternberg MR, et al. Seroprevalence of human papillomavirus type 16 infection in the United States. J Infect Dis 2002; 186:1396402. First citation in article

    10.  Karem KL, Poon AC, Bierl C, Nisenbaum R, Unger E. Optimization of a human papillomavirus-specific enzyme-linked immunosorbent assay. Clin Diagn Lab Immunol 2002; 9:57782. First citation in article

    11.  Casella G, Berger BL. Statistical inference. Belmont, CA: Brooks/Cole, 1990:32830. First citation in article

    12.  Luxton JC, Rose RC, Coletart T, Wilson P, Shepherd PS. Serological and T-helper cell responses to human papillomavirus type 16 L1 in women with cervical dysplasia or cervical carcinoma and in healthy controls. J Gen Virol 1997; 78:91723. First citation in article

    13.  Marais D, Rose RC, Williamson AL. Age distribution of antibodies to human papillomavirus in children, women with cervical intraepithelial neoplasia and blood donors from South Africa. J Med Virol 1997; 51:12631. First citation in article

    14.  Mund K, Han C, Daum R, et al. Detection of human papillomavirus type 16 DNA and of antibodies to human papillomavirus type 16 proteins in children. Intervirology 1997; 40:2327. First citation in article

    15.  Cubie HA, Plumstead M, Zhang W, de Jesus O, Duncan LA, Stanley MA. Presence of antibodies to human papillomavirus virus-like particles (VLPs) in 11-13-year-old schoolgirls. J Med Virol 1998; 56:2106. First citation in article

    16.  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; 80:48993. First citation in article

    17.  Manns A, Strickler HD, Wikktor SZ, Pate EJ, Gray R, Waters D. Low incidence of human papillomavirus type 16 antibody seroconversion in young children. Pediatr Infect Dis J 1999; 18:8335. First citation in article

    18.  Marais DJ, Rose RC, Lane C, et al. Seroresponses to human papillomavirus types 16, 18, 31, 33, and 45 virus-like particles in South African women with cervical cancer and cervical intraepithelial neoplasia. J Med Virol 2000; 60:40310. First citation in article

    19.  Marais D, Rose RC, Lane C, Aspinall S, Bos P, Williamson AL. Seroresponses to virus-like particles of human papillomavirus types 16, 18, 31, 33, and 45 in San people of Southern Africa. J Med Virol 2000; 60:3316. First citation in article

    20.  Giroglou T, Sapp M, Lane C, et al. Immunological analyses of human papillomavirus capsids. Vaccine 2001; 19:178393. First citation in article