Pickard, Robert K. L. MS, MPH, EdM*; Xiao, Weihong MD*; Broutian, Tatevik R. MS*; He, Xin PhD†; Gillison, Maura L. MD, PhD*
Research over the last 15 years has established oral human papillomavirus (HPV) infection to be a cause of oropharyngeal squamous cell carcinoma (OSCC).1 A case-control study by D'Souza and colleagues estimated oral HPV-16 infection to increase odds of OSCC by nearly 15-fold.2 HPV-positive OSCC cases are distinct from HPV-negative cases with regard to risk factors, treatment response, and survival outcomes.3,4 The incidence of OSCC has been increasing in several countries during the past 3 decades.5–9
Little is known about the epidemiology and natural history of oral HPV infection, especially in healthy populations. A recent literature review by Kreimer and colleagues summarized 18 published studies of oral HPV infection in cancer-free, HIV-negative subjects.10 The pooled prevalence of HPV infection was approximately 4.5%, and prevalence was similar for men and women. Sample sizes among the 18 studies were small, and the majority did not include analysis of behavioral risk-factor associations. Although 3 studies included some form of follow-up, incidence rates were not provided.11–13
The objective of this study was to estimate the prevalence and incidence of oral HPV infection in a large cohort of young adults. Additionally, we explored factors associated with prevalent and incident infections for hypothesis generation for a larger natural history study.
MATERIALS AND METHODS
Study Population and Data Collection
The study population was a convenience sample of 1000 students at The Ohio State University (OSU; Columbus, OH). Eligibility criteria included men and women aged 18 to 30 years, enrollment in an educational program at OSU, and ability to comply with a 3-month follow-up visit. The study protocol was approved by the OSU Institutional Review Board, and written informed consent was obtained from all participants.
At both the baseline and follow-up visits, participants provided an oral rinse sample (ORS) and completed a computer-assisted self-interview survey on a tablet computer. The computer-assisted self-interview instrument assessed demographic information; past and current use of tobacco, alcohol, marijuana and cocaine; history of sexually transmitted infection; HPV vaccination and abnormal pap smears (for women); and sexual behavior history, including lifetime and recent (past 3 months) number of opposite- and same-sex sexual partners for open-mouth kissing and performing oral sex, and opposite-sex sexual partners for vaginal sex. In the survey, open-mouth kissing was referred to as “French” kissing, and defined as “putting your tongue into a man's [or woman's] mouth.” Performing oral sex was defined as “putting your mouth or tongue on a man's penis [or woman's vagina] or genitals.” Numbers of opposite- and same-sex kissing and oral sex partners were combined to generate total number of lifetime and recent partners.
Specimen Collection, Processing, and DNA Purification
An ORS was collected by use of 10 mL of Scope (Proctor and Gamble, Cincinnati, OH) mouthwash. The participant alternated between swishing (total of 20 seconds) and gargling (total of 10 seconds) for a total of 30 seconds, and then expectorated the sample into a sterile specimen collection cup. On receipt in the laboratory, oral exfoliated cells in the ORS were separated by centrifugation, rinsed with phosphate-buffered saline, resuspended in phosphate-buffered saline, and split into 2 equal aliquots. One aliquot was pelleted by centrifugation at 4000 × g for 10 minutes. The supernatant was decanted and the pellet was resuspended in 1.2 ml of Qiagen cell lysis buffer (Qiagen, Inc., Hilden, Germany) and incubated at room temperature for 15 minutes. The sample was then digested with DNase-free RNase A (5 μg/mL; Roche Molecular Systems, Inc., Pleasanton, CA) for 30 minutes at 37°C followed by Proteinase K digestion (final concentration of 0.5 mg/mL; Sigma-Aldrich, St. Louis, MO) overnight at 55°C. After heat inactivation at 95°C for 10 minutes, DNA was purified by use of the Qiagen Virus/Bacteria Midi Kit (Qiagen, Inc., Hilden, Germany) on the Qiasymphony SP instrument as recommended by the manufacturer. Isolated nucleic acid was eluted in 60 μL of QIAamp viral elution buffer.
Ten microliters of purified ORS DNA were analyzed for 37 types of HPV by use of a multiplex polymerase chain reaction assay that targets the conserved L1 region of the viral genome using PGMY primers and primers for an internal control β-Globin gene (Roche Linear Array HPV Genotyping Test; Roche Molecular Systems, Inc., Pleasanton, CA). Polymerase chain reaction products were denatured in 0.13 mol/L NaOH and hybridized to an immobilized HPV probe array using reverse line blot hybridization. β-Globin positive samples were considered evaluable and classified as HPV-positive if any of the 37 HPV DNA types were detected, including high-risk (16, 18, 26, 31, 33, 35, 39, 45, 51, 52, 53, 56, 58, 59, 66, 68, 73, 82) or low-risk (6, 11, 40, 42, 54, 55, 61, 62, 64, 67, 69, 70, 71, 72, 81, 82 subtype [IS39], 83, 84, 89 [cp6108]) types.14
Current substance use was defined as having used a substance in the 3 months before the baseline survey. Differences between groups with respect to behavioral characteristics were tested by χ2 test for categorical variables and nonparametric Wilcoxon rank sum test for continuous variables. Individuals found positive for any of the HPV types evaluated were considered “infected,” as is the standard practice in the general literature. HPV point prevalence and 95% confidence interval (CI) were calculated. Odds ratios (ORs) and 95% CIs for factors associated with oral HPV infection were estimated using univariate and bivariate logistic regression analyses. In addition to factors associated with any HPV infection, associations with high- and low-risk HPV infections were explored using univariate multinomial logistic regression. Analysis of incidence was conducted by individual and by type-specific infection. Because an individual can be infected with multiple types of HPV, an incident infection was defined as a new type-specific infection detected at follow-up that was not detected at baseline. Incidence rate and 95% CI were estimated using the method outlined by Rosner,15 and univariate logistic regression was used to estimate ORs for factors associated with incident infection among individuals. For all statistical tests, a P value of ≤0.05 was considered a statistically significant association. Statistical analyses were conducted using Stata version 10.1 (StataCorp, College Station, TX).
One thousand men and women were enrolled in the protocol starting in September 2009, and follow-up visits were conducted through March 2010. The demographic and behavioral characteristics of the study population are shown in Table 1. More than half of the study population was female, the majority was white (69%), and the median age was 21 (interquartile range [IQR], 19–23) years. The majority of the cohort was sexually experienced, as 74% had performed oral sex and 69% had had vaginal sex. Approximately, a third had 3 or more lifetime vaginal sex partners and a third had 3 or more lifetime oral sex (performing) partners. Fifty-six percent had 5 or more lifetime open-mouth kissing partners. Only 8% reported no history of open-mouth kissing or other sexual contact. A correlation was observed between number of lifetime vaginal and oral sexual partners (φ correlation coefficient = 0.59), but correlations between lifetime kissing and vaginal (0.42) or oral sexual partners (0.38) were more moderate.
Substance use was common in the study population. In the past 3 months, 85% had consumed alcohol, 18% had smoked marijuana, and 19% had smoked cigarettes. However, the median number of cigarettes smoked per week among those having smoked in the previous month was one (IQR, 0–9).
Sexual and substance use behaviors were strongly associated with gender and race (Table 1). Women reported a greater number of lifetime and recent oral sex partners than men. Women were also more likely to report current or past alcohol use, but men were more likely to report current or past marijuana use. The number of lifetime and recent kissing and oral sex partners was greater among whites than among blacks and other races.
Oral HPV infections were detected at baseline among 2.4% (95% CI: 1.4–3.4) of the study population. The prevalence of low- and high-risk HPV types was similar (1.2%, 95% CI: 0.5–1.9). Only 2 participants had an oral HPV-16 infection (0.2%, 95% CI: 0–0.4).
Demographic and behavioral associations with prevalent oral HPV infection are shown in Table 2. Increasing age, lifetime number of kissing or vaginal or oral sex partners, and frequency of marijuana use were significantly associated with oral HPV infection in univariate analysis. Additionally, gender, black race, history of genital warts or a sexually transmitted infection, and never having had alcohol were nonsignificantly associated with oral HPV infection. By contrast, neither cigarette smoking nor receipt of an HPV vaccine (women only) appeared associated with oral HPV infection. Multinomial analysis indicated that the higher prevalence among blacks versus whites was restricted to low-risk infections, and the association with increasing age was stronger for high-risk infections (data not shown).
To explore which factors may be associated with oral HPV infection after accounting for sexual behavior, bivariate analysis was performed, adjusting for lifetime number of vaginal sex partners (dichotomized to 0–4 and ≥5). A history of 5 or more lifetime open-mouth kissing partners (OR, 4.0; 95% CI: 1.1–14.8) or oral sex partners (OR, 4.0; 95% CI: 1.3–11.9), and ever having consumed alcohol (OR, 0.2; 95% CI: 0.1–0.8) were associated with oral HPV infection after adjusting for number of vaginal sex partners (Table 2). Frequency of recent marijuana use, gender, and black race were also associated with infection, albeit not significantly. Similar associations were observed when the number of oral sex partners was substituted for vaginal sex partners for adjustment (data not shown). Lifetime number of vaginal sex partners was not significantly associated with infection after adjusting for either kissing or oral sex partners.
Retention at the follow-up visit was high (98.6%), with a median follow-up time of 3.7 months (IQR, 3.3–3.8). Sexual and substance use behaviors were similar during the 3 months before enrollment and the follow-up visit (Table 3). Approximately one-tenth reported new oral or vaginal sex partners within the previous 3 months, and 27% reported kissing a new partner.
The follow-up observation time among the 985 participants with evaluable samples was 3525.2 person-months (PM). Cross-sectional prevalence at follow-up was similar to baseline (2.7%, 95% CI: 1.7–3.8). All 24 individuals with prevalent oral HPV infection at baseline completed the follow-up visit. Thirty-nine percent of the 33 type-specific HPV infections detected at baseline among the 24 individuals persisted (Fig. 1).
A total of 20 type-specific incident infections were detected in 19 individuals at the follow-up visit. Therefore, we estimated the crude incidence rate to be 5.67 per 1000 PM (95% CI: 3.12–8.16/1000 PM). In univariate analysis, incident infection was significantly associated with black race (OR, 4.1; 95% CI: 1.4–11.9) and having open-mouth kissed a new partner in the previous 3 months (OR, 2.6; 95% CI: 1.0–6.4).
Natural history studies of cervical HPV infection performed in convenience samples of college students subsequently provided data critical for public health interventions related to cervical cancer. HPV has only recently been associated with oropharyngeal cancer, and natural history studies of oral HPV infection have yet to be performed. Although the low prevalence and incidence of oral HPV infection observed in this large pilot study limited our ability to identify factors independently associated with infection, our data indicate that age, sexual behaviors, alcohol use, marijuana smoking, gender, and race are important factors to investigate in future natural history studies. Additionally, we provide the first estimate for incidence rate of infection, necessary for the design of such studies.
The prevalence of oral HPV infection in our cohort was 2.4%, which was consistent with prior reports in healthy adolescent and college-aged populations from United States and Canadian studies. Our estimate is slightly lower than that reported in a recent comprehensive literature review of oral HPV prevalence by Kreimer and colleagues,10 likely because of the younger age range of our study population. Prevalence of oral HPV infection has been observed to increase with age in several studies.16,17
Oral sexual behaviors have been associated with oral HPV infection16–18 as well as transmission of other viral infections to the oral cavity, such as HSV.18 In this study, lifetime number of partners for oral sex (performing) and open-mouth kissing partners were associated with prevalent oral HPV infection after accounting for vaginal sex partners, indicating that kissing and oral sex behaviors may be important factors in the transmission of HPV to the oral cavity. Importantly, this is the first study to link open-mouth kissing with incident oral HPV infection, albeit in univariate analysis. Researching the link between open-mouth kissing and oral HPV infection will be important to determining age range recommendations for HPV vaccination, as the age of initiation of kissing is substantially earlier than oral sex. Prospective studies of younger populations will be helpful in further elucidating these associations, as distinct groups of individuals with and without kissing and oral and vaginal sexual experiences could be observed. Natural history studies in populations with higher oral HPV prevalence, such as HIV-infected individuals and men who have sex with men, are less likely to be informative in this regard.19 But smaller sample sizes may make them more feasible.
Unexpectedly, ever having consumed alcohol was associated with lower odds of oral HPV infection, even after adjustment for lifetime number of vaginal sex partners. This finding is counter-intuitive, given that alcohol consumption has been associated with risky sexual behaviors.20 In an animal model, high concentrations of ethanol were found to denature the viral capsid and prevent infection by HPV-16.21 Clearly, further analysis of the association between alcohol consumption and oral HPV infection is warranted.
The association between frequent recent marijuana use and oral HPV infection was marginally significant in our cohort. Frequency and intensity of marijuana smoking has been associated with HPV-16–positive OSCC,3 suggesting that marijuana use behaviors may be associated with oral HPV infection. Research has also found associations with frequent marijuana use and periodontal disease22 and cytokine suppression,23 supporting a possible local immunosuppressant role for marijuana use.
Tobacco smoking was not associated with oral HPV infection in this cohort, unlike in prior studies.17,24,25 However, very few individuals in this study were daily tobacco users. The median number of cigarettes smoked per week among recent smokers was one. These data indicate that most current smokers were likely experimenting with tobacco or casual users, making associations with daily use difficult to assess.
Male gender has been associated with oral HPV infection16 and higher incidence rates for HPV-positive OSCC.6 Our study found a marginally significant association with male gender and infection after adjusting for lifetime number of vaginal sex partners. Behavioral factors about sexual behavior or substance use may explain men's greater risk of oral HPV infection, and biologic factors such as varying HPV viral loads in male and female genital sites or immune response differences may also play a role.
We observed an association between black race and prevalent and incident oral HPV infection. In exploratory analysis, the higher prevalence was restricted to low-risk infections. Our data therefore do not explain the observation of lower incidence of HPV-positive OSCC among blacks in the United States.26 Associations between race and low- and high-risk oral HPV infection thus require further study.
To our knowledge, this study is the first to estimate the incidence rate of oral HPV infection in a healthy population as well as factors associated with incident infection. Our data suggest that approximately 4% to 10% of individuals from a similar cohort would be expected to acquire an oral HPV infection each year. As with HPV prevalence, estimates of HPV incidence in anogenital sites in women and men have been substantially higher.27,28 Hypotheses explaining the lower prevalence and incidence of HPV infection in the oral cavity compared with anogenital sites would include the following: (1) HPV infection has a lower viral load in the oral cavity compared with anogenital sites, resulting in reduced detection of infections; (2) transmission of HPV infection to the oral cavity on exposure is more difficult than for genital infection because of differences in local mucosal immunity, permissiveness of cells to infection, epithelial resistance to microtrauma, or of the unique properties of the oral cavity, such as high salivary flow; and (3) HPV infection is cleared at a greater rate in the oral cavity compared with anogenital sites. Further longitudinal research is warranted to investigate these hypotheses.
Our study has several important limitations. The conclusions derived from our convenience sample of students at OSU cannot be generalized to a larger population of young adults in the United States. Selection bias is possible because of the socio-demographic and behavioral differences between college students and the larger population of young adults, as well as differences that may have been associated with an individual's decision to self-select for participation in the study. Additionally, the number of outcome events for both prevalent infections observed at baseline and incident infections observed at follow-up limited our ability to generate robust risk factor models. Our analysis was constrained to univariate and bivariate models, and therefore our ability to simultaneously adjust for multiple factors was limited. Two significant limitations in this study about the estimate of the incidence of oral HPV infection are the short 3-month period of follow-up and that only one follow-up test was conducted.
Findings from this study that show an association between the lifetime number of open-mouth kissing and oral sex partners reflect the results of prior research, and provide further support that oral sexual contact in the form of both oral-oral and oral-genital contact may play a role in the transmission of oral HPV infection. Natural history studies of cervical HPV infection provided essential information for the development of cervical cancer interventions, and the sustained decrease in cervical cancer incidence is a public health success story. Similar studies regarding oral HPV infection are warranted to inform the increasing burden of HPV-positive OSCC in the United States and other countries, and this study provides important preliminary data necessary for the design of such studies.
1. Gillison ML, Koch WM, Capone RB, et al.. Evidence for a causal association between human papillomavirus and a subset of head and neck cancers. J Natl Cancer Inst 2000; 92:709–720.
2. D'Souza G, Kreimer A, Viscidi R, et al.. Case-control study of human papillomavirus and oropharyngeal cancer. N Engl J Med 2007; 356:1944–1956.
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. Fakhry C, Westra WH, Li S, et al.. Improved survival of patients with human papillomavirus-positive head and neck squamous cell carcinoma in a prospective clinical trial. J Natl Cancer Inst 2008; 100:261–269.
5. Blomberg M, Nielsen A, Munk C, et al.. Trends in head and neck cancer incidence in Denmark, 1978–2007: Focus on human papillomavirus associated sites. Int J Cancer 2011; 129:733–741.
6. Chaturvedi AK, Engels EA, Pfeiffer RM, et al.. Human papillomavirus and rising oropharyngeal cancer indcidence in the United States. J Clin Oncol 2011; 29:10.1200/JCO.2011.36.4596.
7. Auluck A, Hislop G, Bajdik C, et al.. Trends in oropharyngeal and oral cavity cancer incidence of human papillomavirus (HPV)-related and HPV-unrelated sites in a multicultural population: The British Columbia experience. Cancer 2010; 116:2635–2644.
8. Doobaree IU, Landis SH, Linklater KM, et al.. Head and neck cancer in South East England between 1995–1999 and 2000–2004: An estimation of incidence and distribution by site, state and histological type. Oral Oncol 2009; 45:809–814.
9. Braakhuis BJM, Visser O, Leemans R. Oral and oropharyngeal cancer in The Netherlands between 1989 and 2006: Increasing incidence, but not in young adults. Oral Oncology 2009; 45:e85–e89.
10. Kreimer AR, Bhatia RK, Messeguer AL, et al.. Oral human papillomavirus in healthy individuals: A systematic review of the literature. Sex Transm Dis 2010; 37:386–391.
11. Kurose K, Terai M, Soedarsono N, et al.. Low prevalence of HPV infection and its natural history in normal oral mucosa among volunteers on Miyako Island, Japan. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2004; 98:91–96.
12. Winer RL, Lee SK, Hughes JP, et al.. Genital human papillomavirus infection: Incidence and risk factors in a cohort of female university students. Am J Epidemiol 2003; 157:218–226.
13. Rintala M, Grénman S, Puranen M, et al.. Natural history of oral papillomavirus infections in spouses: A prospective Finnish HPV Family Study. J Clin Virol 2005; 35:89–94.
14. Muñoz N, Castellsagué X, Berrington de González A, et al.. HPV in the etiology of human cancer. Vaccine 2006; 24S3:S3/1–10.
15. Rosner BA. Fundamentals of Biostatistics. 5th ed. Pacific Grove, CA: Duxbury, 2000.
16. Kreimer AR, Alberg AJ, Daniel R, et al.. Oral human papillomavirus infection in adults is associated with sexual behavior and HIV serostatus. J Infect Dis 2004; 189:686–698.
17. D'Souza G, Agrawal Y, Halpern J, et al.. Oral sexual behaviors associated with prevalent oral human papillomavirus infection. J Infect Dis 2009; 199:1263–1269.
18. Edwards S, Carne C. Oral sex and the transmission of viral STIs. Sex Transm Infect 1998; 74:6–10.
19. Beachler DC, Weber KM, Margolick JB, et al.. Risk factors for oral HPV infection among a high prevalence population of HIV-positive and at-risk HIV-negative adults. Cancer Epidemiol Biomarkers Prev 2012; 21:122–133.
20. Cooper ML. Alcohol use and risky sexual behavior among college students and youth: evaluating the evidence. J Stud Alcohol 2002; S14:101–117.
21. Roden RB, Lowy DR, Schiller JT. Papillomavirus is resistant to desiccation. J Infect Dis 1997; 176:1076–1079.
22. Thomson WM, Poulton R, Broadbent JM, et al.. Cannabis smoking and periodontal disease among young adults. JAMA 2008; 299:525–531.
23. Shay AH, Choi R, Whittaker K, et al.. Impairment of antimicrobial activity and nitric oxide production in alveolar macrophages from smokers of marijuana and cocaine. J Infect Dis 2003; 187:700–704.
24. Kreimer AR, Villa A, Nyitray AG, et al.. The epidemiology of oral HPV infection among a multinational sample of healthy men. Cancer Epidemiol Biomarkers Prev 2011; 20:172–182.
25. Smith EM, Swarnavel S, Ritchie JM, et al.. Prevalence of human papillomavirus in the oral cavity/oropharynx in a large population of children and adolescents. Pediatr Infect Dis J 2007; 26:836–840.
26. 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 2009; 2:776–781.
27. World Health Organization International Agency for Research on Cancer. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Vol 90: Human Papillomaviruses. Lyon, France): WHO Press, 2007:689.
28. Giuliano AR, Lee JH, Fulp W, et al.. Incidence and clearance of genital human papillomavirus infection in men (HIM): A cohort study. Lancet 2011; 377:932–940.