Serum Immunoglobulin G Response to Human Papillomavirus Type 16 Virus-Like Particles in Human Immunodeficiency Virus (HIV)Positive and Risk-Matched HIV-Negati

¹Johns Hopkins University School of Medicine and ²School of Public Health, Baltimore, and ³Division of AIDS, National Institute of Allergy and Infectious Diseases, Bethesda, Maryland; ⁴Cook County Hospital and Rush Medical College, Chicago, Illinois; ⁵The Miriam Hospital, Brown University, Providence, Rhode Island; ⁶Montefiore Medical Center and ⁷Albert Einstein College of Medicine, Bronx, and ⁸Maimonides Medical Center, Brooklyn, New York; ⁹University of Southern California School of Medicine, Los Angeles, and ¹⁰University of California at San Francisco, San Francisco; ¹¹Centers for Disease Control and Prevention, Atlanta, Georgia; ¹²Wayne State University School of Medicine, Detroit, Michigan; ¹³Georgetown University Medical Center, Washington, DC

Received 6 May 2002; revised 9 September 2002; electronically published 6 January 2003.

     Presented in part: 19th International Papillomavirus Conference, Costao do Santinho, Florianopolis, Brazil, 17 September 2001 (abstract 139).
     Informed consent was obtained from all subjects, and the human subjects investigational committee at each participating institution approved the study.
     Financial support: National Institutes of Health (AI-42058 to R.P.V.; U01-AI-35004, U01-AI-31834, U01-AI-34994, AI-34989, U01-HD-32632, and U01-AI-34993; and U01-AI-42590, for the Women's Interagency HIV Study ); General Clinical Research Center at the University of California at San Francisco (M01-RR00079 and M01-RR00083, for WIHS); Centers for Disease Control and Prevention (U64/CCU106795, U64/CCU206798, U64/CCU306802, and U64/CCU506831, for the HIV Epidemiology Research Study).
     Reprints or correspondence: Dr. Raphael P. Viscidi, The Johns Hopkins Hospital, Blalock Rm. 1105, 600 N. Wolfe St., Baltimore, MD 21287 ().

     Human immunodeficiency virus (HIV) infection in women is associated with a high prevalence of anogenital human papillomavirus (HPV) infection. Investigations of HPV infection in HIV-positive women have focused on detection of viral DNA and virus-induced cellular abnormalities. HPV DNA is found 23 times as frequently in cervicovaginal lavage and anal swab specimens from HIV-positive women as it is in those from HIV-negative women [18]. HIV-positive women are 25 times as likely as HIV-negative women to have HPV-associated cervical squamous intraepithelial lesions and anal intraepithelial neoplasia [13, 5, 914].

     Additional insights into an infectious disease can come from serological studies. An antibody response is one indicator of past infection that can be used in the absence of detectable virus or viral cytologic changes. Numerous epidemiological studies of HIV-negative women have examined immunological responses to cervical HPV infection by measuring antibodies to virus-like particles (VLPs). HPV VLPs are empty capsids that are generated in cell culture after expression of the L1 or L1 and L2 structural genes. The VLPs are morphologically and immunologically similar to native virus particles and perform well as reagents in serological assays [15, 16]. The low reactivity of serum samples from children and virgins and the strong association of seroreactivity with sexual behavior suggest that HPV VLPbased assays are specific for the anogenital HPV types [1724]. Animal models demonstrate that VLP assays can identify serum samples that are reactive to different but closely related anogenital HPV types [2528]. Human studies have suggested that VLP assays have moderate-to-good type specificity. These studies demonstrate stronger associations of seroreactivity with detection of viral DNA in the cervix of the same type and with type-specific seroconversion after incident infection [20, 2932]. Numerous studies indicate that VLP assays performed on serum samples have a sensitivity of 50% for cervical HPV infection that is detectable by polymerase chain reaction (PCR) [16, 24, 29, 3337]. The low sensitivity may reflect a weak or delayed antibody response to infection, a transient HPV infection, an antibody response that is restricted to the mucosal compartment, the absence of a capsid antigenspecific antibody response, or insensitivity of VLP-based assays.

     Two previous studies have examined seroreactivity to HPV VLPs among HIV-positive women [38, 39]. Both studies, which included small numbers of subjects, reported a higher HPV seroprevalence among HIV-positive women than among HIV-negative women, but neither study examined other risk factors for HPV seropositivity. In this study, we examined serum specimens from subjects enrolled in the Women's Interagency HIV Study (WIHS) and the HIV Epidemiology Research Study (HERS), 2 large prospective cohort studies of HIV-positive and HIV-negative women recruited at multiple sites in the United States [40, 41]. The aim of our study was to determine the baseline prevalence of serum IgG antibody to HPV-16 capsids in relation to HIV status, genital tract HPV infection at baseline, and demographic and behavioral risk factors for HPV infection.

POPULATION AND METHODS

     Study population.     Serological testing was done on serum samples obtained at enrollment from 1978 HIV-positive and 546 HIV-negative women enrolled in the WIHS study in Bronx/Manhattan and Brooklyn, New York; Chicago; Los Angeles; San Francisco; and Washington, DC, and 837 HIV-positive and 417 HIV-negative women enrolled in the HERS study in Baltimore, Bronx, Detroit, and Providence, Rhode Island. All women for whom a serum sample obtained at the enrollment visit was available were included in the study. These women constituted 96% of the subjects enrolled in WIHS and HERS.

     The cohort characteristics, recruitment methods, and protocols of the 2 studies are described elsewhere [40, 41]. Both studies collected similar categories of data and used comparable, but not identical, methods. Each woman underwent an extensive structured interview detailing her medical, behavioral, and psychosocial history, including queries about past and current sexual practices. In the 2 studies, HIV-negative women were enrolled to achieve comparability between seropositive and seronegative cohorts according to age, race/ethnicity, level of education, injection drug use, lifetime number of sex partners, and enrollment site.

     Each woman had a complete physical and gynecologic examination and provided blood, a Papanicolaou (Pap) smear, and a cervicovaginal lavage sample (using 10 mL of sterile saline) for HPV DNA detection. The cervicovaginal fluid was tested for the presence of HPV DNA by PCR analysis with MY09/MY11 L1 consensus primers and a control primer set (PC04/GH20), which simultaneously amplified a -globin DNA fragment. The PCR methods and the results of HPV PCR testing of women in WIHS and HERS have been the subject of articles published elsewhere [6, 8]. HPV genotyping was done by dot blot with type-specific probes for the following HPV types: 6, 11, 16, 18, 26, 3133, 35, 39, 40, 45, 5156, 58, 59, 61, 66, 6870, 73, Pap 155, Pap 291, and AE2 in WIHS and 6, 11, 16, 18, 26, 31, 33, 35, 39, 40, 42, 45, 5156, 58, 59, 66, 68, 73, and 8284 in HERS. In both studies, -globinnegative samples (8% in WIHS and 1.7% in HERS) were excluded from analysis.

     In WIHS and HERS, 18% and 11%, respectively, of HIV-positive and HIV-negative women harbored untypeable HPVs in the lower genital tract. The HPV DNA prevalences at the 6 WIHS sites were 62%, 64%, 59%, 50%, 58%, and 48%; the HPV DNA prevalences at the 4 HERS sites were 51%, 45%, 47%, and 48%. The HPV-16 DNA prevalences at the 6 WIHS sites were 6.2%, 4.0%, 4.7%, 4.8%, 2.2%, and 5.4%; the HPV-16 DNA prevalences at the 4 HERS sites were 4.9%, 5.8%, 5.1%, and 4.2%. Hybridizations were scored for intensity: 0, negative; 14, weak to strong positive (up to 5 in WIHS). For analytical purposes, scores of 1 or 2 were considered low signal strength and scores of 3 were considered strong signal strength. The method of scoring dot-blot hybridization assays for HPV-16 in WIHS was validated by real-time PCR. In an analysis of 261 genital specimens, scores for the HPV-16 DNAspecific probe of 15 were significantly correlated with the median HPV-16 copy number (P = .017; Spearman's rank test) (P. E. Gravitt, personal communication).

     In each study, 2 cytotechnologists who used the Bethesda system for cervicovaginal cytologic diagnosis read Pap smears [42]. All abnormal smear findings including atypical cells of undetermined significance (ASCUS), and 10% of normal Pap smear findings were reviewed by a cytopathologist. For analysis, diagnoses were grouped as "benign findings," "ASCUS," "low-grade squamous intraepithelial lesion" (LSIL), "high-grade squamous intraepithelial lesion" (HSIL), and "cancer."

     Preparative purification of HPV-16 VLPs.     For production of VLPs, Trichoplusia ni (High Five) cells (Invitrogen) were infected with HPV-16 L1/L2 recombinant baculovirus, clone 114/K (gift from John T. Schiller, National Cancer Institute, Bethesda, MD) and grown as adherent cultures in tissue culture plates (245 × 245 mm; Nunc) in a volume of 90 mL of Ex-Cell 400 (JRH Biosciences) medium per plate. After 96 h of incubation at 27°C, the cells were harvested by scrapping the plate, pelleted by low-speed centrifugation, and frozen. The procedure for purification of VLPs combined elements of protocols described elsewhere [4345]. The cell paste was thawed at 4°C, resuspended in PBS (pH 7.2) containing a cocktail of protease inhibitors (Complete Mini; Roche), and sonicated on ice twice for 60 s using a 550 Sonic Dismembrator with a microtip (Fisher Scientific) at a setting of 5. The lysate was loaded onto a 40% sucrose-PBS cushion and centrifuged in an SW-28 rotor at 112,700 g for 2.5 h. The pellet was resuspended in 27% (wt/wt) CsCl-PBS by short-pulse sonication and was centrifuged overnight in an SW-28 rotor at 141,371 g to remove the contaminating buoyant layer. The clarified lysate was then centrifuged for 48 h at 198,409 g in a Vti50 vertical rotor. Fractions (2 mL) were collected and analyzed by SDS-PAGE and refractometry.

     Fractions containing L1 protein (55 kDa) at a density of 1.28 g/cm³ were pooled. The Vti50 pool was diluted 1 : 6 in 50 mM 3-(N-morpholino)propanesulfonic acid (MOPS; pH 7.0) to reduce the salt concentration of the solution and then fractionated by column chromatography, using POROS 50HS strongcation exchange resin (PerSeptive Biosystems). The column was preequilibrated with 50 mM MOPS (pH 7) containing 0.33 M NaCl (buffer A) at room temperature. After loading the VLP-containing solution, the column was washed with 10 column volumes of buffer A, and proteins were eluted at room temperature with a 20column volume linear salt gradient to 1.5 M NaCl. Fractions (1.5 mL) were collected, and total protein was measured using the Bio-Rad protein assay kit (Bio-Rad Laboratories) and IgG as a standard.

     The major protein peak eluting between 0.70 and 0.95 M NaCl was collected and dialyzed extensively against MOPS containing 1 M NaCl. The dialysate was diluted 1 : 3 in 50 mM MOPS (pH 7) and applied to a heparin-sepharose (Amersham Pharmacia Biotech) column equilibrated with buffer A. After washing with 8 column volumes of buffer A, the column was eluted with a 25column volume linear salt gradient to 1.5 M NaCl. Fractions (1 mL) were collected, and total protein was measured as described above. The peak protein-containing fractions, which eluted between 0.65 and 0.80 M NaCl, were pooled and stored at 4°C. Analysis of a Coomassie bluestained SDS-PAGE gel with National Institutes of Health Image software showed that HPV-16 L1 protein made up 90% of total protein. The protocol described above yielded 23 mg of HPV-16 VLPs from 1012 g of wet cell paste.

     HPV-16 VLP ELISA.     For ELISA, HPV-16 VLPs were diluted to 0.4 g of total protein/mL in PBS, and the VLP solution was added to each of 96 wells of a polystyrene flat-bottom PolySorp plate (Nunc) in 100-L volumes. After overnight incubation at 4°C, the antigen solution was removed, and 300 L of blocking solution (10% Superblock in PBS containing 0.05% Tween 20 ) was added to each well. The plates were incubated at room temperature for 3 h, and then the blocking solution was removed and 300 L of PBS was added to each well. The plates were stored at -20°C. Before use and after each incubation step, the plates were washed 4 times with wash solution (PBS0.05% Tween 20) in an automatic plate washer (Skanwasher 300; Skatron). One microliter of serum in 100 L of sample dilution buffer (10% Superblock and 0.05% Tween 20 in PBS) was added to each well, using a MultiPROBE II robotic liquid handling system (Packard Instruments). Plates were incubated at 37°C for 1 h on a microplate shaker. Chainspecific goat antihuman IgG conjugated with horseradish peroxidase (Zymed) was diluted 1 : 4000 in conjugate buffer (10% Superblock; 2.5% polyethylene glycol, molecular weight 20,000 ; and 0.5% Igepal CA-630 in PBS), and 100 L was added to each well. Plates were incubated at 37°C for 30 min, and then freshly prepared 2,2-azinobis(3-ethylbenzthiazoline-6-sulfonate) and hydrogen peroxide solution (Kierkegaard and Perry) prewarmed to 50°C was added to each well in 100-L volumes. The plates were incubated at room temperature in the dark for 20 min. The enzyme reaction was stopped by the addition of 100 L of 1% dodecyl sulfate/well to all the plates. The plates were read at 405 nm, with a reference wavelength of 490 nm, in an automated microtiter plate reader (Molecular Devices).

     Three controls were included on each plate, a weak positive control, a moderate-to-strong positive control, and a negative control. The cutoff point for positive results was determined from the reactivity of plasma samples obtained during another study from 108 self-reported virgins [46]. The mean and SD of optical density (OD) values for the control subjects were calculated, and values greater than mean + 3 SD were excluded. The analysis was repeated for the remaining samples until no further OD values could be excluded by this criterion. The cutoff point established after 4 rounds of analysis, after which 6 samples had been excluded, was OD >0.136. Paired serum and plasma samples obtained at the same visit from 42 women were tested by ELISA. The plasma and serum OD values were highly correlated (Pearson's correlation coefficient, r = 0.95), and the distributions of OD values of serum and plasma samples were not statistically significantly different (P = .54; Mann-Whitney rank sum test). The variability observed when plasma samples were retested (coefficient of variation, 15%) was similar to the variability seen when serum and plasma samples were compared (coefficient of variation, 17%).

     Statistical analysis.     The distributions of absorbance values were compared by HIV status and CD4 cell count, using a nonparametric Kruskal-Wallis test. Prevalence rates of HPV-16 seropositivity were computed and compared using Pearson's ² test and the Mantel-Haenszel trend test for ordered categorical data. Odds ratio (OR) estimates with 95% confidence intervals (CIs) were obtained by logistic regression. Covariates with P values <.01 in univariate analyses were considered for the multivariate models for each cohort. Statistical analyses were performed using SAS version 8.2 (SAS Institute).

RESULTS

     HIV-positive women in WIHS and HERS were similar to each other and to HIV-negative women in terms of age, race/ethnicity, socioeconomic status, and sexual behavior () [40, 41]. The average age of women in WIHS and HERS was 35 years. More than one-half of the women in WIHS and HERS were of black non-Hispanic origin, and approximately one-fourth of the women in the WIHS cohort and nearly one-fifth of the women in the HERS cohort were of Hispanic/Latina descent. Approximately two-thirds of the subjects had annual incomes $12,000. A history of recent injection drug use (within 6 months of study entry) was more common among women in HERS (25%) than among women in WIHS (10%). Among HIV-positive women, a greater proportion of women in WIHS (30.4%) than of women in HERS (17.1%) had a baseline CD4 cell count <200 cells/mm³, which may be explained by the inclusion of women who had already received a diagnosis of AIDS at study entry in WIHS. The median lifetime number of sex partners was 10, and one-fourth of the women reported having had >25 sex partners. More than one-half of the women had had a sex partner in the past 6 months. HPV DNA in the genital tract was common among all subjects and was seen most frequently among those who were HIV positive (63.8% in both WIHS and HERS at baseline) than among HIV-negative women (30.2% in WIHS and 27.8% in HERS) [6, 8]. Similarly, cervical cytologic abnormalities were more prevalent among HIV-positive women (16.9% in WIHS and 18.2% in HERS at baseline) than among HIV-negative women (3.6% in WIHS and 4.8% in HERS) [9, 14].

fig.ommitted

Table 1.          Selected characteristics, by human immunodeficiency virus (HIV) serostatus, of subjects in the Women's Interagency HIV Study (WIHS) and HIV Epidemiology Research Study (HERS) cohorts.

     Distributions of OD values from the HPV-16 VLP ELISA for women in WIHS and HERS are shown in . The median (interquartile range) OD values for HIV-positive and HIV-negative women in WIHS were 0.149 (0.0710.282) and 0.111 (0.0480.272), respectively (P < .001); the corresponding values for women in HERS were 0.160 (0.0810.322) and 0.130 (0.0560.325) (P = .003). Serum samples were also categorized as positive or negative for HPV-16 antibodies, using the designated cutoff point. The prevalence of serum IgG antibodies to HPV-16 VLPs was higher among HIV-positive women than among HIV-negative women (53% vs. 44.7% in WIHS [P < .001] and 56.5% vs. 49.2% in HERS [P = .014]). The HPV-16 seroprevalences at the 6 WIHS sites were 52%, 43%, 48%, 58%, 49%, and 48%; the HPV-16 seroprevalences at the 4 HERS sites were 44%, 58%, 58%, and 50%. There were no statistically significant differences in seroprevalence among HIV-positive women in WIHS when data were stratified by CD4 cell count (). In HERS, HPV-16 seroprevalence was lower among women with CD4 cell counts >500 cells/mm³ (50%) than among women with CD4 cell counts 200500 cells/mm³ (60%) or <200 cells/mm³ (62%). In the WIHS cohort, data from plasma HIV load testing were available, and there was no significant difference in the distribution of OD values or seroprevalence when data from HIV-positive women were stratified by virus load (data not shown).

fig.ommitted

Figure 1.        Distribution of optical density values by human immunodeficiency virus (HIV) status in the Women's Interagency HIV Study (WIHS; A) and the HIV Epidemiology Research Study (HERS; B). The length of each box corresponds to the interquartile range, with the upper boundary of the box representing the 75th and the lower boundary the 25th percentile. The horizontal line in the box indicates the median value. The lines extending upward and downward from the box mark the 10th to 90th percentile range. Outlier values are shown as closed circles. The difference in optical density values between HIV-positive and HIV-negative women was statistically significant in WIHS (P < .001) and HERS (P = .003).

fig.ommitted

Table 2.          Human papillomavirus type 16 (HPV-16) seropositivity and CD4 cell counts among human immunodeficiency viruspositive women in the Women's Interagency HIV Study (WIHS) and HIV Epidemiology Research Study (HERS) Cohorts

     Because serum antibodies are a marker of past as well as present infection, we examined the relationship between HPV-16 VLP seroreactivity and known risk factors for exposure to HPV . There were no or weak associations with race/ethnicity, marital status, and injection drug use. In contrast, younger age was associated with HPV-16 seropositivity among HIV-positive women in WIHS, with a seroprevalence of 61% among women 30 years old and of 50.4% among women >30 years old (P < .001). A strong association was observed for lifetime number of sex partners among both HIV-positive women and HIV-negative women. Among HIV-positive women, HPV-16 seroreactivity was 60% for women who had had 11 sex partners, compared with 45% for women with 2 sex partners (P < .001, for WIHS, and P = .007, for HERS), and among HIV-negative women, the corresponding rates were 50% and 30% (P < .001, for WIHS, and P = .03, for HERS). Having had a sex partner in the past 6 months was associated with a higher HPV-16 seroprevalence among HIV-positive women in WIHS (57.7% among HIV-positive women vs. 43.5% among HIV-negative women; P < .001). For all other groups of women, HPV-16 seroprevalence was higher among women who reported having a new sex partner in the past 6 months than among those who did not, but the differences were not statistically significant. Self-report of prior gonorrhea, syphilis, or vaginal herpes was associated with HPV-16 seropositivity among HIV-positive subjects in both studies and among HIV-negative women in WIHS but not in HERS.

fig.ommitted

Table 3.          Human papillomavirus type 16 (HPV-16) seropositivity and risk factors for HPV infection, by human immunodeficiency virus (HIV) serostatus, among women in the Women's Interagency HIV Study (WIHS) and HIV Epidemiology Research Study (HERS) cohorts.

     We also examined the relationship of HPV-16 VLP seroreactivity with baseline HPV infection and HPV-associated disease. Among HIV-negative women for whom the results of PCR analysis of cervical vaginal lavage were positive for HPV-16 DNA, the HPV-16 seroprevalences were 90% in WIHS and 43% in HERS, respectively (62.5% for all HIV-negative women); among HIV-positive women, the seroprevalences were 53% and 61% . The HPV-16 seroprevalence was greater among women with high DNA PCR signal strength (64% in WIHS and 62% in HERS) than among women with low DNA PCR signal strength (39% in WIHS and 44% in HERS; P = .007, for WIHS and HERS combined) (data not shown). HPV-16 seroreactivity was independent of concurrent genital tract infection with other HPV types. The HPV-16 seroprevalence was 57% among women infected with other HPV types and 57% among women without other HPV types (data not shown). Women without HPV DNA had similar HPV-16 seropositivity rates, regardless of HIV serostatus (43% and 49% of HIV-negative women and 46% and 51% of HIV-positive women in WIHS and HERS, respectively). HIV-positive women with HPV types other than type 16 detected in the lower genital tract were more likely than HIV-positive women with no HPV DNA to be HPV-16 seropositive (56% vs. 46% in WIHS [P < .001] and 60% vs. 51% in HERS [P = .008]). The difference in HPV-16 seroprevalence cannot be explained by increased reactivity to HPV-16 VLPs in women with HPV-16related HPV types (types 31, 33, 35, 52, and 58), because the seroprevalence among women with HPV-16related types and those with HPV-16unrelated types was comparable (59% vs. 54%; P = .07). There was no statistically significant difference in HPV-16 seropositivity rates among HIV-negative women with no HPV DNA and HIV-negative women with nontype 16 HPV DNA in either study cohort.

fig.ommitted

Table 4.          Human papillomavirus type 16 (HPV-16) seropositivity and HPV cervical infection, by human immunodeficiency virus (HIV) serostatus, among subjects in the Women's Interagency HIV Study (WIHS) and HIV Epidemiology Research Study (HERS) cohorts.

     Among women for whom the results of PCR were positive for HPV-16 DNA, the HPV-16 seroprevalence was higher among those with LSIL or HSIL Pap smear abnormalities than among those with normal Pap smear findings or ASCUS (60% vs. 56% in WIHS and 68% vs. 52% in HERS), but the differences were not statistically significant . HIV-positive women and HIV-negative women were not analyzed separately, because very few HIV-negative women were HPV-16 DNA positive (10 women in WIHS and 14 in HERS). Among women who were HPV-16 PCR negative, the HPV-16 seroprevalence was significantly higher among those with LSIL or HSIL cytologic abnormalities than among those with normal smear findings or ASCUS (60% vs. 49% in WIHS [P = .001] and 66% vs. 53% in HERS [P = .006]). The association between HPV-16 seropositivity and squamous intraepithelial lesions was limited to HIV-positive women for whom the results of PCR were positive for nontype 16 HPV DNA, because, as expected, cytologic abnormalities were uncommon among women who were HPV DNA negative, and very few HIV-negative women had abnormal Pap smear findings (data not shown).

fig.ommitted

Table 5.          Human papillomavirus type 16 (HPV-16) seropositivity and Pap smear results among subjects in the Women's Interagency HIV Study (WIHS) and HIV Epidemiology Research Study (HERS) cohorts.

     To examine the association between HPV-16 seropositivity and HIV infection, 2 logistic regression models were constructed for each cohort. One model examined the association between HPV-16 seropositivity and HIV infection after adjustment for demographic and sexual behavior risk factors for prior exposure to HPV that were significantly associated with HPV seropositivity in univariate analyses (). The second model adjusted for baseline HPV infection and HPV-associated cytologic abnormalities (). HPV-16 seropositivity was significantly associated with HIV infection in the WIHS cohort (OR, 1.33; 95% CI, 1.081.54) and in the HERS cohort (OR, 1.51; 95% CI, 1.142.01) after adjustment for demographic and sexual behavior variables. However, there was no association between HPV-16 seropositivity and HIV infection after adjustment for baseline HPV infection and Pap smear abnormalities.

fig.ommitted

Table 6.          Logistic regression model for human papillomavirus type 16 (HPV-16) seropositivity in relation to human immunodeficiency virus (HIV) infection, demographic characteristics, and past sexual behavior variables in the Women's Interagency HIV Study (WIHS) and the HIV Epidemiology Research Study (HERS).

fig.ommitted

Table 7.          Logistic regression model for human papillomavirus type 16 (HPV-16) seropositivity in relation to human immunodeficiency virus (HIV) infection, HPV infection, and HPV-associated cytologic abnormalities in the Women's Interagency HIV Study (WIHS) and the HIV Epidemiology Research Study (HERS).

DISCUSSION

     Our study is, to our knowledge, the largest reported to date of serum IgG antibody responses to HPV-16 capsids in HIV-positive women. The inclusion of women from 2 cohorts made it possible to replicate findings. The HIV-positive women included in the study are a geographically and ethnically diverse population representative of HIV-infected women in the United States [40, 41]. The HIV-negative women were frequency-matched by age and risk factors for HIV infection. The women in both populations were primarily non-white and were drawn from groups with low socioeconomic status. The HIV-positive and HIV-negative women in the 2 studies had similar sexual behavior risk factors.

     The results of our analysis of demographic and behavioral risk factors for HPV infection indicated that the determinants of HPV-16 seroreactivity for HIV-positive women were similar to those for risk-matched HIV-negative women. The risk factor that was demonstrated to have the strongest association with HPV-16 seropositivity in both HIV-positive women and HIV-negative women and for which the result was replicated between studies was the lifetime number of sex partners. An association between seroreactivity to HPV capsid proteins and larger lifetime number of sex partners has been observed in multiple studies of HIV-negative women and is the expected pattern of seroreactivity for a sexually transmitted infection [20, 22, 29, 37]. Among HIV-positive men, there is an association between HPV-16 seropositivity and having >50 sex partners in a lifetime [47]. The only other factor that was significantly associated with HPV-16 seroreactivity across studies was a self-report of gonorrhea, syphilis, or vaginal herpes, which is also a risk factor related to sexual activity. Some risk factors were associated with HPV-16 seropositivity in only 1 of the 2 studies or in a particular HIV serostatus group. For example, younger age and having had a sex partner recently were associated with HPV-16 seropositivity among HIV-positive women in WIHS; genital warts were associated with HPV-16 seropositivity among HIV-positive women in HERS; and age at first intercourse of <14 years was associated with HPV-16 seropositivity among HIV-negative women in HERS. For each risk factor, a higher HPV-16 seroprevalence was also observed in the other cohort, but the differences did not reach statistical significance. Because multiple comparisons were performed, these findings should be confirmed in an independent study.

     The serum IgG response to baseline type-specific HPV cervical infection and HPV-associated cytologic abnormalities was similar in HIV-positive and HIV-negative women. When results of the 2 studies were combined, the prevalence of HPV-16 seroreactivity was 50%60% among HIV-positive and HIV-negative women for whom an HPV-16 genital tract infection was detected by PCR. The apparent discrepancy in HPV-16 seroprevalence among HIV-negative women in WIHS and HERS may be due to the small number of HPV-16 cervical infections that were detected in these groups (11 in WIHS and 14 in HERS). An HPV-16 seroprevalence of 50%60% is similar to or slightly higher than that reported in most other studies of diverse populations of HIV-negative women [16, 29, 36, 37, 48]. The absence of detectable antibody responses in some women with HPV infections has been attributed, in part, to a lag period between the time that HPV DNA can be detected by PCR and the development of an antibody response [31]. The level of viral replication may also influence the antibody response by increasing the amount of antigen available to stimulate an immune response. We found that the HPV-16 seroprevalence increased from 40% to 60% as the PCR signal strength went from low to high. In a study of HPV-16 seroreactivity among college women, we also observed a correlation between seropositivity and HPV load [37]. The capsid antibody response to HPV-16 infection in HIV-positive women was similar to that in HIV-negative women, even though HPV infections are more persistent in HIV-positive women [49] and HPV persistence is a determinant of HPV seropositivity [29, 33, 36]. A possible explanation for this finding may be the observation by Carter et al. [31] that, after reaching a peak, the rate of seroconversion following incident infection decreases over time, even in the presence of HPV DNA. The HIV-positive women in our study might have already reached this peak. The relationship between capsid antibody response and Pap smear findings was also similar for HIV-positive and HIV-negative women. Among women for whom PCR results were positive for HPV-16 DNA and among those with negative results, slightly more than one-half of the HIV-positive and HIV-negative women who had normal Pap smear findings or ASCUS were HPV-16 seropositive, whereas nearly two-thirds of those with LSIL or HSIL were HPV-16 seropositive.

     A major focus of our study was to explore the relationship between HIV infection and HPV-16 seroreactivity. In the unadjusted analysis, HPV-16 seropositivity was significantly associated with HIV infection in WIHS (OR, 1.39; 95% CI, 1.062.21) and HERS (OR, 1.34; 95% CI, 1.061.70). The strength of the association was unchanged after adjustment for demographic and sexual behavior factors that are likely to reflect past exposure to HPV, which indicates that the difference between the HPV-16 seroprevalence among HIV-positive women and the seroprevalence among HIV-negative women cannot be explained by differences in past exposure to HPV-16. This finding is not surprising, because the HIV-negative women in WIHS and HERS were reasonably well matched in behavioral and demographic risk factors for sexually transmitted infections. After adjustment for baseline HPV cervical infection and HPV-associated cytologic abnormalities, there was no statistically significant association between HPV-16 seropositivity and HIV infection in either WIHS (OR, 1.21; 95% CI, 0.981.5) or HERS (OR, 1.18; 95% CI, 0.901.55). This suggests that the finding of an HPV-16 seroprevalence among HIV-positive women that is higher than that among high-risk HIV-negative women is explained by the higher prevalence of HPV infection among HIV-positive women. One factor that has been shown to contribute to the high prevalence of HPV infection among HIV-positive women is the longer duration of HPV infection in this group [49], and a longer duration of HPV infection is one of the major determinants of HPV seropositivity [36]. In the WIHS cohort, we found, as others have reported elsewhere [39, 47], that HPV seropositivity was not associated with CD4 cell count. In addition, in WIHS, plasma HIV load was not associated with HPV-16 seropositivity. However, in the HERS cohort, HPV-16 seropositivity was significantly associated with a CD4 cell count <500 cells/mm³, even after adjustment for other factors shown to influence seropositivity (OR, 1.65; 95% CI, 1.072.53). That no association was found between HPV seropositivity and severity of HIV infection is not surprising, because HIV infection primarily perturbs cellular immune responses and not humoral immune responses, except in end-stage disease. On the other hand, HPV prevalence has been shown to increase as the CD4 cell count decreases [6], and active HPV infection might be expected to stimulate an antibody response. Further studies of the relationship between HIV-induced immunosuppression, HPV infection, and HPV seropositivity are warranted.

     The high prevalence of serum IgG antibodies to HPV-16 VLPs among HIV-positive women (50%60%), regardless of baseline HPV-16 status, indicates that these women have a high lifetime cumulative exposure to HPV-16. It has been assumed, on the basis of self-reported sexual behavior information, that HIV-positive women have high rates of past exposure to HPV. Serological assays provide a confirmatory biomarker for past exposure and a measure of the extent and type specificity of the exposure. Our study was cross-sectional in design and thus did not address the question of whether seropositivity is a marker of immune protection against reinfection. However, the findings may have implications for HPV vaccine efforts. If HPV vaccines do not protect seropositive individuals from reinfection, then many HIV-positive individuals may not benefit from vaccination. This would be unfortunate, because HPV infection and cytologic abnormalities are more common among HIV-positive women and may increase the risk for cervical cancer in this group. As the AIDS epidemic spreads globally, the greatest burden of HPV-associated disease may fall on HIV-positive women, making them an important target group for HPV vaccines. The difference between the results of serological testing for HPV-16 infection and the results of PCR analysis of cervicovaginal lavage was dramatic. Although more than one-half of the HIV-positive women had serological evidence of past infection with HPV-16, only 5%6% of the women had detectable HPV-16 DNA in the genital tract. Thus, most HIV-positive women appear to be able to control HPV replication at the cervix, and reactivation, if it occurs at all, is uncommon.

References

1. 

Sun XW, Ellerbrock TV, Lungu O, Chiasson MA, Bush TJ, Wright TCJ. Human papillomavirus infection in human immunodeficiency virusseropositive women. Obstet Gynecol 1995; 85:6806.

2. 

Miotti PG, Dallabetta GA, Daniel RW, et al. Cervical abnormalities, human papillomavirus, and human immunodeficiency virus infections in women in Malawi. J Infect Dis 1996; 173:7147.

3. 

Hillemanns P, Ellerbrock TV, McPhillips S, et al. Prevalence of anal human papillomavirus infection and anal cytologic abnormalities in HIV-seropositive women. AIDS 1996; 10:16417.

4. 

Minkoff H, Feldman J, DeHovitz J, Landesman S, Burk R. A longitudinal study of human papillomavirus carriage in human immunodeficiency virusinfected and human immunodeficiency virusuninfected women. Am J Obstet Gynecol 1998; 178:9826.

5. 

Williams AB, Darragh TM, Vranizan K, Ochia C, Moss AR, Palefsky JM. Anal and cervical human papillomavirus infection and risk of anal and cervical epithelial abnormalities in human immunodeficiency virusinfected women. Obstet Gynecol 1994; 83:20511.

6. 

Palefsky JM, Minkoff H, Kalish LA, et al. Cervicovaginal human papillomavirus infection in human immunodeficiency virus-1 (HIV)positive and high-risk HIV-negative women. J Natl Cancer Inst 1999; 91:22636.

7. 

Sun XW, Kuhn L, Ellerbrock TV, Chiasson MA, Bush TJ, Wright TCJ. Human papillomavirus infection in women infected with the human immunodeficiency virus. N Engl J Med 1997; 337:13439.

8. 

Jamieson DJ, Duerr A, Burk R, et al. Characterization of genital human papillomavirus infection in women who have or who are at risk of having HIV infection. Am J Obstet Gynecol 2002; 186:217.

9. 

Massad LS, Riester KA, Anastos KM, et al. Prevalence and predictors of squamous cell abnormalities in Papanicolaou smears from women infected with HIV-1. Women's Interagency HIV Study Group. J Acquir Immune Defic Syndr 1999; 21:3341.

10. 

Six C, Heard I, Bergeron C, et al. Comparative prevalence, incidence and short-term prognosis of cervical squamous intraepithelial lesions amongst HIV-positive and HIV-negative women. AIDS 1998; 12:104756.

11. 

Minkoff HL, Eisenberger-Matityahu D, Feldman J, Burk R, Clarke L. Prevalence and incidence of gynecologic disorders among women infected with human immunodeficiency virus. Am J Obstet Gynecol 1999; 180:82436.

12. 

Maiman M, Fruchter RG, Sedlis A, et al. Prevalence, risk factors, and accuracy of cytologic screening for cervical intraepithelial neoplasia in women with the human immunodeficiency virus. Gynecol Oncol 1998; 68:2339.

13. 

Wright TCJ, Ellerbrock TV, Chiasson MA, Van Devanter N, Sun XW. Cervical intraepithelial neoplasia in women infected with human immunodeficiency virus: prevalence, risk factors, and validity of Papanicolaou smears. New York Cervical Disease Study. Obstet Gynecol 1994; 84:5917.

14. 

Duerr A, Kieke B, Warren D, et al. Human papillomavirusassociated cervical cytologic abnormalities among women with or at risk of infection with human immunodeficiency virus. Am J Obstet Gynecol 2001; 184:58490.

15. 

Kirnbauer R, Booy F, Cheng N, Lowy DR, Schiller JT. Papillomavirus L1 major capsid protein self-assembles into virus-like particles that are highly immunogenic. Proc Natl Acad Sci USA 1992; 89:121804.

16. 

Kirnbauer R, Hubbert NL, Wheeler CM, Becker TM, Lowy DR, Schiller JT. A virus-like particle enzyme-linked immunosorbent assay detects serum antibodies in a majority of women infected with human papillomavirus type 16. J Natl Cancer Inst 1994; 86:4949.

17. 

Andersson-Ellstrom A, Dillner J, Hagmar B, et al. Comparison of development of serum antibodies to HPV16 and HPV33 and acquisition of cervical HPV DNA among sexually experienced and virginal young girls: a longitudinal cohort study. Sex Transm Dis 1996; 23:2348.

18. 

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.

19. 

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.

20. 

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; 173:13948.

21. 

Olsen AO, Dillner J, Gjoen K, Magnus P. Seropositivity against HPV-16 capsids: a better marker of past sexual behaviour than presence of HPV DNA. Genitourin Med 1997; 73:1315.

22. 

Wideroff L, Schiffman MH, Hoover R, et al. Epidemiologic determinants of seroreactivity to human papillomavirus (HPV) type 16 virus-like particles in cervical HPV-16 DNA-positive and-negative women. J Infect Dis 1996; 174:93743.

23. 

Silins I, Kallings I, Dillner J. Correlates of the spread of human papillomavirus infection. Cancer Epidemiol Biomarkers Prev 2000; 9:9539.

24. 

Kjellberg L, Wang Z, Wiklund F, et al. Sexual behaviour and papillomavirus exposure in cervical intraepithelial neoplasia: a population-based case-control study. J Gen Virol 1999; 80:3918.

25. 

Rose RC, Bonnez W, Da Rin C, McCance DJ, Reichman RC. Serological differentiation of human papillomavirus types 11, 16 and 18 using recombinant virus-like particles. J Gen Virol 1994; 75:24459.

26. 

Hines JF, Ghim SJ, Christensen ND, et al. The expressed L1 proteins of HPV-1, HPV-6, and HPV-11 display type-specific epitopes with native conformation and reactivity with neutralizing and nonneutralizing antibodies. Pathobiology 1994; 62:16571.

27. 

Christensen ND, Kirnbauer R, Schiller JT, et al. Human papillomavirus types 6 and 11 have antigenically distinct strongly immunogenic conformationally dependent neutralizing epitopes. Virology 1994; 205:32935.

28. 

Roden RB, Hubbert NL, Kirnbauer R, Christensen ND, Lowy DR, Schiller JT. Assessment of the serological relatedness of genital human papillomaviruses by hemagglutination inhibition. J Virol 1996; 70:3298301.

29. 

Carter JJ, Koutsky LA, Wipf GC, et al. The natural history of human papillomavirus type 16 capsid antibodies among a cohort of university women. J Infect Dis 1996; 174:92736.

30. 

Wideroff L, Schiffman M, Haderer P, et al. Seroreactivity to human papillomavirus types 16, 18, 31, and 45 virus-like particles in a case-control study of cervical squamous intraepithelial lesions. J Infect Dis 1999; 180:14248.

31. 

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.

32. 

Tachezy R, Hamsikova E, Hajek T, et al. Human papillomavirus genotype spectrum in Czech women: correlation of HPV DNA presence with antibodies against HPV-16, 18, and 33 virus-like particles. J Med Virol 1999; 58:37886.

33. 

de Gruijl TD, Bontkes HJ, Walboomers JM, et al. Immunoglobulin G responses against human papillomavirus type 16 virus-like particles in a prospective nonintervention cohort study of women with cervical intraepithelial neoplasia. J Natl Cancer Inst 1997; 89:6308.

34. 

Nonnenmacher B, Hubbert NL, Kirnbauer R, et al. Serologic response to human papillomavirus type 16 (HPV-16) virus-like particles in HPV-16 DNApositive invasive cervical cancer and cervical intraepithelial neoplasia grade III patients and controls from Colombia and Spain. J Infect Dis 1995; 172:1924.

35. 

Olsen AO, Dillner J, Gjoen K, Sauer T, Orstavik I, Magnus P. A population-based case-control study of human papillomavirus-type-16 seropositivity and incident high-grade dysplasia of the uterine cervix. Int J Cancer 1996; 68:4159.

36. 

Wideroff L, Schiffman MH, Nonnenmacher B, et al. Evaluation of seroreactivity to human papillomavirus type 16 virus-like particles in an incident case-control study of cervical neoplasia. J Infect Dis 1995; 172:142530.

37. 

Viscidi RP, Kotloff KL, Clayman B, Russ K, Shapiro S, Shah KV. Prevalence of antibodies to human papillomavirus (HPV) type 16 virus-like particles in relation to cervical HPV infection among college women. Clin Diagn Lab Immunol 1997; 4:1226.

38. 

Marais DJ, Vardas E, Ramjee G, et al. The impact of human immunodeficiency virus type 1 status on human papillomavirus (HPV) prevalence and HPV antibodies in serum and cervical secretions. J Infect Dis 2000; 182:123942.

39. 

Petter A, Heim K, Guger M, et al. Specific serum IgG, IgM and IgA antibodies to human papillomavirus types 6, 11, 16, 18 and 31 virus-like particles in human immunodeficiency virusseropositive women. J Gen Virol 2000; 81:7018.

40. 

Barkan SE, Melnick SL, Preston-Martin S, et al. The Women's Interagency HIV Study. WIHS Collaborative Study Group. Epidemiology 1998; 9:11725.

41. 

Smith DK, Warren DL, Vlahov D, et al. Design and baseline participant characteristics of the Human Immunodeficiency Virus Epidemiology Research (HER) Study: a prospective cohort study of human immunodeficiency virus infection in US women. Am J Epidemiol 1997; 146:45969.

42. 

Kurman RJ, Solomon D. The Bethesda system for reporting cervical/vaginal cytologic diagnoses. New York: Springer Verlag, 1994.

43. 

Kirnbauer R, Taub J, Greenstone H, et al. Efficient self-assembly of human papillomavirus type 16 L1 and L1-L2 into virus-like particles. J Virol 1993; 67:692936.

44. 

Cook JC, Joyce JG, George HA, et al. Purification of virus-like particles of recombinant human papillomavirus type 11 major capsid protein L1 from Saccharomyces cerevisiae. Protein Expr Purif 1999; 17:47784.

45. 

Joyce JG, Tung JS, Przysiecki CT, et al. The L1 major capsid protein of human papillomavirus type 11 recombinant virus-like particles interacts with heparin and cell-surface glycosaminoglycans on human keratinocytes. J Biol Chem 1999; 274:581022.

46. 

Herrero R, Schiffman MH, Bratti C, et al. Design and methods of a population-based natural history study of cervical neoplasia in a rural province of Costa Rica: the Guanacaste Project. Rev Panam Salud Publica 1997; 1:36275.

47. 

Hagensee ME, Kiviat N, Critchlow CW, et al. Seroprevalence of human papillomavirus types 6 and 16 capsid antibodies in homosexual men. J Infect Dis 1997; 176:62531.

48. 

Chua KL, Wiklund F, Lenner P, et al. A prospective study on the risk of cervical intra-epithelial neoplasia among healthy subjects with serum antibodies to HPV compared with HPV DNA in cervical smears. Int J Cancer 1996; 68:549.

49. 

Ahdieh L, Klein RS, Burk R, et al. Prevalence, incidence, and type-specific persistence of human papillomavirus in human immunodeficiency virus (HIV)positive and HIV- negative women. J Infect Dis 2001; 184:68290.