Human papillomavirus (HPV) is responsible for a growing subset of oropharyngeal carcinomas (OPCs) and nearly all cervical malignancies.1,2 Human papillomavirus–positive oropharyngeal and cervical malignancies are strongly associated with sexual behaviors; oral and cervical HPV infection, respectively; and seropositivity to oncogenic types of HPV.1,3–6 Human papillomavirus–positive OPCs disproportionately affect whites when compared with HPV-negative OPCs.7 In contrast, cervical cancers are more common among black women.8,9
It is not understood why HPV-related malignancies of the oropharynx are more common among whites, yet HPV-related malignancies of the cervix are more common among blacks. Racial differences in sexual behavior may contribute to this heterogeneity. There are few reports on racial differences in oral and vaginal sexual behaviors within the context of HPV infection. Therefore, we examined differences in sexual behaviors in a cohort of women with available sexual history and HPV serology data.
MATERIALS AND METHODS
This was a secondary analysis of existing data and samples from a previously described cohort study.10,11 The “HPV in Perimenopause” (HIP) Study was a Johns Hopkins Bloomberg School of Public Health Institutional Review Board–approved prospective cohort study designed to assess the natural history of cervical HPV infection during the menopausal transition.10,11 Women between the ages of 35 and 60 years attending outpatient obstetrics and gynecology clinics at Johns Hopkins Medicine (Baltimore, MD) between March 2008 and March 2011 were approached. Eligibility criteria included English language fluency, presence of an intact uterus, and willingness to provide informed consent. Women were ineligible for participation in this study if they were pregnant or planning to become pregnant during the 2-year study period, had a history of organ transplantation, or had a known human immunodeficiency virus infection. Women were enrolled at a baseline study visit and were followed up at 6-month intervals for 2 years. Informed consent was obtained from all study participants.
This cross-sectional secondary analysis used data from the baseline visit only. A standardized telephone-administered questionnaire was used to ascertain demographic and current and lifetime sexual behavioral data including recent and lifetime number of vaginal sex partners, ever and recent performance of oral sex on a man, and number of male oral sex partners. Serum was collected to determine serostatus for HPV types 6, 11, 16, 18, 31, 35, 45, and 52. Serostatus was determined by virus-like-particle (VLP)–based enzyme-linked immunosorbent assay as previously described.12 Briefly, VLPs were produced in insect cells from recombinant baculoviruses expressing the L1 gene of the respective HPV types and purified by density gradient ultracentrifugation. Microtiter plates were coated with 50 ng of VLP protein. Serum samples were tested in duplicate at a 1:100 dilution, and bound immunoglobulin was detected using peroxidase-conjugated antihuman IgG, followed by addition of a 2,2′-azino-di-(3-ethylbenzthiazoline-6-sulfonate) hydrogen peroxide solution. Absorbance was measured at 405 nm in an automated microtiter plate reader. Positive and negative control sera were included with each enzyme-linked immunosorbent assay run. The optical density (OD) cut-points for seropositivity were defined as a mean absorbance value greater than the mean plus 3 standard deviations of the reactivity of 60 serum samples from children younger than 2 years, excluding outliers.
Descriptive variables were summarized with frequencies and proportions for categorical variables, and medians and interquartile ranges (IQRs) for continuous variables. The associations between race and demographic characteristics were evaluated using χ2 tests for categorical variables and a nonparametric rank sum test for age (considered as a continuous variable). The associations between characteristics of interest (e.g., race, age, and sexual behaviors) and binary outcomes (e.g., HPV serostatus and ever oral sex) were evaluated using log binomial regression in both univariate and multivariate models and reported as prevalence ratios (PR) with 95% confidence intervals (CIs). Age was considered a binary variable with a cutoff of 45 years for ease of discussion. Mean OD for each HPV serotype was also log-transformed and considered a continuous variable to evaluate for association with age using univariate and multivariate linear regression analysis. Variables were selected for inclusion in multivariate models if they were considered biologically relevant and/or were statistically significant in univariate analysis. P values less than 0.05 were considered statistically significant. Data analysis was performed using Stata 11.2 (College Station, TX).
Characteristics of the Study Population
A total of 885 women enrolled in the HPV in Perimenopause Study and completed both the baseline questionnaire and cervical sample collection. Most of the study participants were white (n = 620; 70%) or black (n = 161; 18%), and the remainder were self-designated as other (American-Indian, Asian, Pacific, or unknown). This analysis was restricted to white and black women given the small number of nonblack and nonwhite women in the study population, for a final analytic sample size of 781.
The characteristics of the study population are summarized in Table 1. The median age of the study participants was 47 years (IQR, 42–52 years). White women had a slightly higher median age than did black women (47 vs. 46 years, P = 0.04) and were more likely to be married (65% vs. 46%, P = 0.001). Most of the study population was educated (58% had a college or graduate degree). White women were more likely than black women to have completed a postbaccalaureate educational program (34% vs. 17%, P < 0.001) and report an annual household income of $80,000 or greater (56% vs. 35%, P < 0.001).
Differences in Sexual Behavior by Race and Age
Differences in sexual behaviors of interest were considered by race (Table 2). In univariate analysis, white women were significantly less likely than black women to have had 5 or greater lifetime male vaginal sex partners (PR, 0.86; 95% CI, 0.77–0.97), but equally likely to report vaginal sex in the previous 6 months (PR, 1.01; 95% CI, 0.92–1.11). In contrast, white race was significantly associated with all measures of oral sex. White women were more likely to report 5 or greater lifetime oral sex partners (PR, 2.38; 95% CI, 1.62–3.49), ever having oral sex (PR, 1.12; 95% CI, 1.04–1.20), and performing oral sex in the previous 6 months (PR, 1.22; 95% CI, 1.01–1.47). Even after accounting for the effects of income and marital status, in multivariate analysis, the positive associations between white race and lifetime oral sexual behaviors remained robust, although the associations of race with vaginal sexual behaviors were no longer significant (adjusted PRs [aPRs] in Table 2). There were no significant differences by race with respect to anal sexual behaviors.
Differences in sexual behaviors by age were also examined (Table 3). Younger women (35–44 years) were significantly more likely than older women (45–60 years) to report 5 or more lifetime partners for both vaginal and oral sex (PRs, 1.24 [95% CI, 1.12–1.38] for vaginal and 1.42 [95% CI, 1.15-1.73] for oral sex). When stratified by race, the higher prevalence of 5 or more vaginal sex partners observed in younger women was similar for black compared with white women (PRs, 1.23 [95% CI, 1.09–1.40] for whites and 1.21 [95% CI, 1.00-1.48] for blacks); however, the higher prevalence of 5 or more oral sex partners for younger women compared with older was more pronounced among white women, and indeed was not significant for black women (PRs, 1.49 [95% CI, 1.21–1.83] for whites compared with 1.38 [95% CI, 0.66-2.88] for blacks). These findings were consistent after multivariate analysis (Table 3); however, there was not a significant interaction between race and age cohort in association with number of oral or vaginal sex partners (P value for interaction = 0.84 for oral and P = 0.88 for vaginal sex).
Differences in HPV Seroprevalence by Race
Serologic data were available for 664 women, most of whom were white (n = 536; 80.7%). The demographic characteristics and reported sexual behaviors of women with and without available serologic data were similar, with the exception that a lower proportion of white compared with black women had serologic data available (72% vs. 81%, P = 0.03; Supplementary Table 1, http://links.lww.com/OLQ/A130).
White women were significantly less likely than black women to be seropositive to HPV overall (59% vs. 76%; PR, 0.78; 95% CI, 0.69–0.88; Table 4). When considering type-specific HPV seroprevalence, white women were significantly less likely than black women to be seropositive to HPV18 (16% vs. 32%), HPV31 (19% vs. 38%), HPV35 (14% vs. 27%), HPV45 (26% vs. 41%), and HPV52 (31% vs. 48%; Table 4). Whites were also less likely than blacks to be seropositive to HPV16, albeit nonsignificantly (25% vs. 32%; PR, 0.77; 95% CI, 0.57–1.03). Using number of seroprevalent types as a measure of the breadth of exposure to HPV, whites were significantly less likely than blacks to be seroreactive to at least 3 types (24% vs. 39%; PR, 0.62; 95% CI, 0.47–0.80). These associations were essentially unchanged after adjustment for age and lifetime number of vaginal sex and oral sex partners (Table 4).
HPV Seroprevalence, Sexual Behaviors, and Age With Stratification by Race
Differences in seroprevalence of HPV16, HPV18, and multiple (≥3) HPV types by race, age, and sexual behaviors were further examined (Table 5). For the study population overall, a higher lifetime number of vaginal sex partners was associated with seropositivity to HPV16, HPV18, and at least 3 HPV types in both univariate and multivariate analyses, and a higher lifetime number of oral sex partners was associated with seropositivity to HPV16, HPV18, and at least 3 HPV types in univariate analysis but only to HPV18 in multivariate analysis. In race-stratified analysis, factors associated with HPV16 seroprevalence differed (Table 5). Among black women, younger age and higher number of lifetime oral sex partners were associated with HPV16 seroprevalence in univariate analysis, but in multivariate analysis, only age remained significantly associated with HPV16 seroprevalence (aPR for HPV16, 0.55; 95% CI, 0.32–0.89). Among white women, although a higher number of both vaginal and oral sex partners were associated with HPV16 seroprevalence in univariate analysis, only higher number of vaginal sex partners was significantly associated with seroprevalence in multivariate analysis (adjusted Ptrend < 0.001). Similar trends were observed with HPV18 and multiple HPV types (Table 5).
Differences in Mean OD by Age in Whites Compared With Blacks
The relationships between serologic response to HPV types, age, and race were further explored using type-specific mean OD value, a surrogate for strength of serologic response. Median of the mean OD was significantly higher among black than white women for all HPV types examined (Table 6), consistent with the higher seroprevalence that was observed among blacks compared with whites.
There was a significant interaction of age and race in association with mean OD for most HPV types. Among black women, older age was negatively associated with mean OD for all HPV types (Table 6 and Supplementary Figure 1, http://links.lww.com/OLQ/A130). In contrast, age was not associated with mean OD for most HPV types among white women, with the exception of HPV6 and HPV52 (Table 6). This pattern persisted after adjustment for lifetime number of vaginal and oral sex partners, except that mean OD for HPV6 was no longer significantly associated with age among whites.
The different racial distributions in HPV-related cervical and oropharyngeal cancers are not well understood. Cervical cancer is more common among blacks,8,9 whereas OPC is more common among whites.7 This study reports racial differences in specific sexual behaviors associated with cervical and oral HPV infection, HPV seroprevalence, and determinants of HPV serostatus in a cohort of older black and white women. These findings provide insight to the unique racial differences in the epidemiology of HPV-related cancers.
White women in this cohort reported higher levels of oral sexual exposure on univariate and multivariate analysis, whereas black women reported higher levels of vaginal sexual exposure, although this was only significant in univariate analysis. This is consistent with previous population-based studies that reported that whites are more likely to have had oral sex and with more partners than blacks, whereas blacks tend to have more vaginal sexual partners.13–15 Our findings extend these observations to an older population of women with relatively high socioeconomic status and examine the relationship of oral and vaginal sexual behaviors with HPV seropositivity.
Consistent with nationally representative US survey data,16,17 seroprevalence of any HPV type, most individual HPV types, and multiple HPV types was higher among black women as compared with white women. Strength of serologic response to most HPV types as measured by mean OD was also higher among blacks. These observations in the context of black women reporting higher genital sexual exposure and white women reporting higher oral sexual exposure may reflect differential immune responses elicited by genital compared with oral exposure to HPV. Indeed, immune microenvironments of the genital tract and oral cavity are distinct with little overlap in types and concentrations of immune markers.18 Moreover, the determinants of clearance and persistence for oral and cervical HPV likely differ, as supported by distinct type-specific HPV infection prevalence and persistence rates, and age distributions.19,20 Therefore, seroconversion may be more likely after genital than oral exposure to HPV.
However, it is likely that determinants of seroprevalence in blacks compared with whites are more nuanced than simply a differential degree of cumulative genital compared with oral sexual exposures by race. The relationships between race, sexual behaviors, and serostatus were complex in our study, with a significant interaction between race and age in association with serologic response to HPV. Among blacks, older age was independently associated with lower HPV seroprevalence and strength of serologic response. In contrast, among whites, sexual behaviors but not age were independently associated with seropositivity. Furthermore, seropositivity to most HPV types was higher among black women even after adjustment for lifetime number of vaginal and oral sex partners. Indeed, a recent study of cervical HPV infection in college-aged women reported that black women cleared the virus significantly later than white women, leading to the conclusion that undefined biologic and genetic factors influencing susceptibility to HPV infection—in addition to behavioral factors—likely contribute to racial differences in HPV epidemiology.21
An important variable to consider may be the order of sexual exposure, which has not been well explored. Other studies have reported that compared with blacks, whites generally initiate oral sex at a younger age, whereas blacks tend to engage in vaginal sex at a younger age than whites.13,22,23 If prior oral infection confers partial protection against subsequent genital infection, then the latter infections would be more rapidly cleared or result in a lower viral load, two variables strongly associated with a lower seroprevalence in women with genital HPV infection.24 It follows, then, that the higher seroprevalence to HPV among blacks in our study may be explained by a lower likelihood of oral before vaginal sexual debut. However, data regarding sequence of sexual debut were not collected in this study, so further investigation will be required to explore this hypothesis.
Another potential influence not accounted for in our study is the prevalence of HPV infection among sexual partners. Black men have higher reported measures of sexual behaviors associated with genital HPV infection than do white men, whereas white men are more likely to report behaviors associated with oral HPV infection.13 Therefore, the risk of exposure to HPV per sexual encounter may also differ according to partner race group.
The lower seroprevalence and strength of serologic response with increasing age among black women, and similar but statistically nonsignificant findings among older white women, were consistent with national cross-sectional survey data demonstrating lower seroprevalence in women after approximately the fourth decade.16,25 This decline may be secondary to either lower cumulative sexual exposures in older age cohorts, such as we observed, that may not have been fully adjusted for in multivariate models, or waning immunity with age, or elements of both.16,26
The clinical and public health impact of naturally acquired HPV antibody level, beyond its utility as a marker of exposure, is unclear. Higher antibody level from natural infection has been associated with a lower risk of subsequent cervical infection for HPV16 and HPV18,27,28 albeit inconsistently,12 whereas higher antibody level was not found to protect from subsequent HPV16 infection in the only study of oral HPV infection thus far.29 Whether there are racial differences in any immune protection afforded by naturally acquired HPV antibodies is unknown.
The racial differences in oral compared with vaginal sexual behaviors observed in our study are consistent with the epidemiology of HPV-related cervical and oropharyngeal cancers, which are caused most commonly by HPV16 (both oropharyngeal and cervical cancers) and HPV18 (cervical cancer). Blacks had higher measures of vaginal sex in this and other studies13 and have higher incidence of cervical cancer,8,9 although other important factors including socioeconomic disparities30 also impact cervical cancer incidence rates by race. Consistent with our finding that white race was strongly associated with measures of oral sex, the higher incidence of HPV-related oropharyngeal cancer among whites has been attributed, in large part, to higher rates of oral sexual behaviors.13 However, as our study highlights, the interactions between HPV, sexual exposures and the immune system are complex, and measures of sexual behavior alone likely do not fully account for these epidemiologic differences.
Importantly, a limitation of our study is that it included only women, and HPV-positive oropharyngeal cancers currently occur primarily in men. However, based on previous studies, correlates of oral sexual behaviors are similar by sex.14 We also did not have oral HPV infection data available in this study, which is of much lower prevalence than genital HPV infection, so that the relative prevalence and associated risk profiles in our study population were unknown. In addition, this is a cross-sectional study that cannot fully account for sexual norms that change over time, and sexual behavior was measured using self-report, which is subject to recall bias. Finally, the generalizability of our findings is limited because our cohort was geographically restricted to a single US metropolitan area.
In summary, in this study of midadult and older women with relatively high socioeconomic status, sexual behaviors were significantly associated with race. Measures of oral sex were higher among white compared with black women. The relationships among race, age, sexual behaviors, and serologic response to HPV are complex, suggesting that additional undescribed factors influence HPV serostatus. Further research is required to fully characterize differences in immune response to HPV by race, age, and sexual exposures, and to better understand how these differences influence HPV natural history and the unique epidemiology of HPV-related malignancies.
1. Walboomers JM, Jacobs MV, Manos MM, et al. Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J Pathol 1999; 189: 12–19.
2. Chaturvedi AK, Engels EA, Pfeiffer RM, et al. Human papillomavirus and rising oropharyngeal cancer incidence in the United States. J Clin Oncol 2011; 29: 4294–4301.
3. Gillison ML, D'Souza G, Westra W, et al. Distinct risk factor profiles for human papillomavirus type 16–positive and human papillomavirus type 16–negative head and neck cancers. J Natl Cancer Inst 2008; 100: 407–420.
4. D'Souza G, Kreimer AR, Viscidi R, et al. Case-control study of human papillomavirus and oropharyngeal cancer. N Engl J Med 2007; 356: 1944–1956.
5. Agrawal Y, Koch WM, Xiao W, et al. Oral human papillomavirus infection before and after treatment for human papillomavirus 16–positive and human papillomavirus 16–negative head and neck squamous cell carcinoma. Clin Cancer Res 2008; 14: 7143–7150.
6. Herrero R, Brinton LA, Reeves WC, et al. Sexual behavior, venereal diseases, hygiene practices, and invasive cervical cancer in a high-risk population. Cancer 1990; 65: 380–386.
7. Settle K, Posner MR, Schumaker LM, et al. Racial survival disparity in head and neck cancer results from low prevalence of human papillomavirus infection in black oropharyngeal cancer patients. Cancer Prev Res (Phila) 2009; 2: 776–781.
8. Rositch AF, Nowak RG, Gravitt PE. Increased age and race-specific incidence of cervical cancer after correction for hysterectomy prevalence in the United States from 2000 to 2009. Cancer 2014; 120: 2032–2038.
9. Saraiya M, Ahmed F, Krishnan S, et al. Cervical cancer incidence in a prevaccine era in the United States, 1998–2002. Obstet Gynecol 2007; 109(2 pt 1): 360–370.
10. Rositch AF, Burke AE, Viscidi RP, et al. Contributions of recent and past sexual partnerships on incident human papillomavirus detection: acquisition and reactivation in older women. Cancer Res 2012; 72: 6183–6190.
11. Gravitt PE, Rositch AF, Silver MI, et al. A cohort effect of the sexual revolution may be masking an increase in human papillomavirus detection at menopause in the United States. J Infect Dis 2013; 207: 272–280.
12. Viscidi RP, Snyder B, Cu-Uvin S, et al. Human papillomavirus capsid antibody response to natural infection and risk of subsequent HPV infection in HIV-positive and HIV-negative women. Cancer Epidemiol Biomarkers Prev 2005; 14: 283–288.
13. D'Souza G, Cullen K, Bowie J, et al. Differences in oral sexual behaviors by gender, age, and race explain observed differences in prevalence of oral human papillomavirus infection. PLoS One 2014; 9: e86023.
14. Leichliter JS, Chandra A, Liddon N, et al. Prevalence and correlates of heterosexual anal and oral sex in adolescents and adults in the United States. J Infect Dis 2007; 196: 1852–1859.
15. Mosher WD, Chandra A, Jones J. Sexual behavior and selected health measures: men and women 15–44 years of age, United States. Adv Data 2002; 2005: 1–55.
16. Markowitz LE, Sternberg M, Dunne EF, et al. Seroprevalence of human papillomavirus types 6, 11, 16, and 18 in the United States: National Health and Nutrition Examination Survey 2003–2004. J Infect Dis 2009; 200: 1059–1067.
17. Thompson DL, Douglas JM Jr, Foster M, et al. Seroepidemiology of infection with human papillomavirus 16, in men and women attending sexually transmitted disease clinics in the United States. J Infect Dis 2004; 190: 1563–1574.
18. Fakhry C, Marks MA, Gilman RH, et al. Comparison of the immune microenvironment of the oral cavity and cervix in healthy women. Cytokine 2013; 64: 597–604.
19. Steinau M, Hariri S, Gillison ML, et al. Prevalence of cervical and oral human papillomavirus infections among US women. J Infect Dis 2014; 209: 1739–1743.
20. D'Souza G, Fakhry C, Sugar EA, et al. Six-month natural history of oral versus cervical human papillomavirus infection. Int J Cancer 2007; 121: 143–150.
21. Banister CE, Messersmith AR, Cai B, et al. Disparity in the persistence of high-risk human papillomavirus genotypes between African American and European American women of college age. J Infect Dis 2015; 211: 100–108.
22. Auslander BA, Biro FM, Succop PA, et al. Racial/ethnic differences in patterns of sexual behavior and STI risk among sexually experienced adolescent girls. J Pediatr Adolesc Gynecol 2009; 22: 33–39.
23. Ompad DC, Strathdee SA, Celentano DD, et al. Predictors of early initiation of vaginal and oral sex among urban young adults in Baltimore, Maryland. Arch Sex Behav 2006; 35: 53–65.
24. Ho GY, Studentsov YY, Bierman R, et al. Natural history of human papillomavirus type 16 virus–like particle antibodies in young women. Cancer Epidemiol Biomarkers Prev 2004; 13: 110–116.
25. Liu G, Markowitz LE, Hariri S, et al. Seroprevalence of 9 human papillomavirus types in the United States, 2005–2006. J Infect Dis 2015; 213: 191–198.
26. Gravitt PE. The known unknowns of HPV natural history. J Clin Invest 2011; 121: 4593–4599.
27. Safaeian M, Porras C, Schiffman M, et al. Epidemiological study of anti-HPV16/18 seropositivity and subsequent risk of HPV16 and -18 infections. J Natl Cancer Inst 2010; 102: 1653–1662.
28. Ho GY, Studentsov Y, Hall CB, et al. Risk factors for subsequent cervicovaginal human papillomavirus (HPV) infection and the protective role of antibodies to HPV-16 virus-like particles. J Infect Dis 2002; 186: 737–742.
29. Beachler DC, Viscidi R, Sugar EA, et al. A longitudinal study of human papillomavirus 16L1, e6, and e7 seropositivity and oral human papillomavirus 16 infection. Sex Transm Dis 2015; 42: 93–97.
30. Garner EI. Cervical cancer: Disparities in screening, treatment, and survival. Cancer Epidemiol Biomarkers Prev 2003; 12: 242s–247s.