Oral HPV Infection among Indigenous Australians; Incidence, Persistence, and Clearance at 12-Month Follow-up

Human papillomavirus (HPV) infection is one of the most prevalent sexually transmitted infections in the world, with serious consequences, potentially cancer, associated with its subclinical persistence in the human body. There is sufficient evidence to support the etiologic role of this virus in cervical, oropharyngeal, anal, and penile precancerous and cancerous lesions (1). There are more than 200 HPV types, with HPV-16 and -18 recognized as having the most severe carcinogenic potential for cervical and oropharyngeal cancer (2).

Although the prognosis for oropharyngeal squamous cell carcinoma (OPSCC) is not favorable, HPV-associated OPSCC has a much better prognosis with recommended “downstaging” in the current Tumor—node—metastasis (TNM) staging system (3). The TNM Classification is a system for classifying malignancies, primarily used in solid tumors and can be used to assist in prognostic cancer staging. The system has its basis on assessing the tumor, regional lymph nodes, and distant metastasis (4). Due to its unique disease progression and behavior patterns, HPV-associated OPSCC has now been identified as a distinct disease entity (5). OPSCC has now surpassed cervical cancer as the most common HPV-driven cancer; oral HPV-16 DNA is now also used as a screening tool to detect early OPSCC (6).

With an ever-increasing global burden of disease due to oral HPV infections (7, 8), it has become imperative to recognize the significant risk factors which effect incidence, persistence, and clearance of oral HPV infection. Although the virus may quickly clear, in some cases it may be retained, leading to increased risk of viral carcinogenesis in certain tissues (9). The natural history of cervical HPV infection and cancer has been extensively reviewed over the past 35 years. It serves as a useful standard and model for comparison with oral HPV infection (10). Improved understanding of oral HPV infection may lead to increased knowledge of the epidemiology and pathogenesis of HPV-associated oral/oropharyngeal cancers. This increased knowledge is crucial for evidence-based management (prevention, screening, and treatment) for oral cancer (11).

Although traditional oral cancers are associated with tobacco- and alcohol-related behaviors, HPV-associated oropharyngeal cancers are associated with sexual behaviors (12). The literature suggests a higher prevalence and incidence of HPV-associated oral cancers among Caucasians, men, and younger individuals of higher socioeconomic status (13–16). The recent surge in oral HPV infection-associated cancers can be attributed to the aging of the “sexual revolution” cohort of the 1960s, which characteristically show a higher number of sexual partners per individual on average and younger age of onset of sexual activities compared with other age cohorts (12, 17, 18). Demographic variations in sexual behaviors and oral health-related behaviors including smoking and tobacco use, explain at least in part the unique epidemiology of both HPV infection and its associated oral and oropharyngeal cancers (15, 19).

Apart from the high-risk HPV types (for example 16 and 18) which could lead to potential malignant lesions, there are some low-risk types which cause papillary or wart-like lesions. One of those lesions include, focal epithelial hyperplasia, or Heck disease, which is a comparatively rare benign condition, caused by oral HPV types 13 or 32 (20–23). It was first identified among a Navajo population in the United States (24) and has since been reported among other Indigenous population groups throughout the world (25, 26).

Indigenous Peoples, as defined by the United Nations (2004), includes all “people with a historical continuity with preinvasion and precolonial societies that developed on their territories, and who consider themselves distinct from other sectors of the societies now prevailing on those territories.” A history of colonial settlement in many countries has resulted in environmental dispossession of traditional lands, resources, languages, and lore (27–30). Indigenous peoples are overrepresented among the poor and underprivileged across the world, and carry a higher burden of preventable health inequalities in comparison with non-Indigenous populations (31). While screening tools and vaccination programs have been successful in reducing overall HPV infection rates, Indigenous communities continue to experience significantly higher infection rates compared with non-Indigenous populations (32–34). In an international study, incidence of cervical cancer was found to be higher among Indigenous women than non-Indigenous women in most countries (Australia, New Zealand, Canada, and the United States; ref. 33). In Australia, there are higher rates of cervical cancer and mortality among Indigenous compared with non-Indigenous women (35).

The main aim of this paper is to describe: (i) acquisition of new oral HPV infections; (ii) persistence of existing oral HPV infection; and (iii) clearance of oral HPV infection among a cohort of Indigenous Australians across 12 months, and to examine associations with sociodemographic, sexual, and health behavior risk factors among Indigenous Australians. The baseline findings for this study have been published elsewhere (36).

Participants included 1,011 Indigenous South Australians aged 18+ years taking part in a broader study involving oral HPV infection and OPSCC (11). This study was conducted in partnership with key Indigenous stakeholder groups, and governed by an Indigenous Reference Group. The Indigenous Reference Group was established to provide oversight and cultural guidance on recruitment strategies and data collection. This included Indigenous community members, councilors, and health workers, and was chaired by an Indigenous health manager. All components of data collection, including sensitive sexual behavior questions, were pilot tested and tailored for cultural sensitivity. Baseline data was collected from February 2018 to January 2019 with 12-month follow-up data collected from February 2019 to January 2020. Participants provided a saliva sample at both time points through spitting and dribbling that was collected in a commercially available kit (DNA Genotek, Inc.), from which microbial DNA was extracted for HPV testing and genotyping. Information on sociodemographic factors, sexual behaviors, and health-related behaviors were ascertained by self-report questionnaire, with assistance provided by the study's Senior Indigenous Research Officer (J. Hedges) where required.

Ethical approval was received from the University of Adelaide Human Research Ethics Committee (Adelaide, South Australia, Australia) and the Aboriginal Health Council of South Australia's Human Research Ethics Committee. All participants provided signed informed consent.

A questionnaire with items including demography, income, health behaviors, and sexual behaviors was used.

• (i) Sociodemographic characteristics: Age (dichotomized based on median split at 37 years; 37 and less, or 38 and more), sex (male or female), geographic location (metropolitan, i.e., residing in Adelaide, South Australia's capital city; and nonmetropolitan, i.e., residing elsewhere in the state) and ownership of a government-administered, means-tested health care card.

• (ii) Health and oral health-related behaviors: Measures included tobacco smoking (currently smoking/formerly smoked or never smoked), alcohol consumption (daily/weekly/monthly, or never) and nonprescription tobacco substitute (i.e., vaporizer or e-cigarette; former or current/former user or never used)

• (iii) Sexual behaviors: Included number of people passionately kissed (less than 3 or more than 3), having ever given and received oral sex (yes or no), age of first giving and receiving oral sex (above 17 years or younger than 16), use of protection when giving and/or receiving oral sex (yes or no), having experienced sexual intercourse (yes or no), age of onset of sexual activity (above 17 years or younger than 16), total number of sexual partners (more or less than 3), and sexual preferences (unisexual or bisexual).

Presence of human DNA in saliva samples was confirmed by β-globin PCR using the primers PCO3 and PCO4 on all samples in the absence of PCR-inhibiting agents (36). Participants with β-globin–positive saliva samples were included in the data analysis, as β-globin checks the DNA integrity; and any samples with a negative β-globin were considered unusable. A nested PCR system MY09/11 and GP5⁺/6⁺ was used for HPV analysis and detection of a large spectrum of mucosal HPV types (36). HPV-positive samples were HPV types by sequencing.

The status of oral HPV infection was categorized as no infection at baseline or 12 months, new incident oral HPV infections at 12 months, persistent oral HPV infection at baseline and 12months, and clearance of oral HPV infection at 12 months. Oral HPV types were also categorized on the basis of risk, according to the guidelines specified by the International Agency for Research in Cancer (IARC). HPV types classified as high-risk were 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, 68, and 82 (37). Heck disease was classified as HPV types 13 and 32 and the rest were classified as low-risk HPV types.

Basic descriptive analyses were conducted to ascertain frequencies of incident, persistent, or cleared oral HPV infections and associated sociodemographic data, health-related behaviors, and sexual history characteristics. The incidence of infection with a given HPV type was calculated exclusively among patients whose first available HPV test result was negative. Persistence was defined as the presence of positive oral HPV infection at both baseline and 12-month follow up. Clearance of a prevalent type-specific HPV infection at baseline was considered to have occurred at the first follow-up visit if that infection was no longer detected.

A cumulative oral HPV infection incidence was calculated by calculating the new infection acquired over the 12-month period divided by the number of infection-free cases at baseline.

Rate ratios (RR) of incident oral HPV infection, persistent oral HPV infection, and clearance of oral HPV infection were determined using negative log binomial regression modeling. Exposure variables were classified into sociodemographic, health behaviors, and sexual health behaviors; variables with associations at the P < 0.2 level in bivariate analysis were included in multivariate analysis.

For incident infection and clearance of oral HPV infection at 12 months, four regression models were constructed; Model 1 included sociodemographic indicators, Model 2 included health behavior indicators, Model 3 included sexual behavior indicators, and Model 4 included sociodemographic, health indicators, and sexual behavior indicators. The final regression model for the acquiring new incident oral HPV infection and clearance of oral HPV infection at 12 months was constructed by adjusting covariates with P value < 0.2 in the bivariate analyses. For persistent oral HPV infection, three regression models were constructed; Model 1 included sociodemographic indicators, Model 2 included sexual behavior indicators, and Model 3 included sociodemographic and sexual behavior indicators. The final regression model for the persistent oral HPV infection at 12 months was constructed by adjusting covariates with P value < 0.2 in the bivariate analyses.

RRs were considered to be statistically significant when P values derived from the Wald statistic were = 0.05. Data were analyzed using STATA version 15.0 (StataCorp) and SPSS (IBM; Version 27).

Recruitment of 1,011 participants was completed at baseline, all of whom provided saliva samples and completed questionnaires. Ninety-four baseline saliva samples (9.2%) tested negative for β-globin, and were not included in further analyses. There were no statistically significant differences in sociodemographic characteristics between participants testing positive or negative for β -globin. The sociodemographic characteristics of this cohort at baseline have been described elsewhere (36, 38). A flow diagram of participants through key study stages is provided in Fig. 1.

Of the 911 baseline saliva samples, 321 (35.2%) tested positive for oral HPV infection. We identified 38 HPV types: 3, 6, 7, 10, 13, 16, 18, 30, 31, 32, 33, 34, 35, 39, 40, 42, 44, 45, 51, 52, 53, 54, 56, 58, 59, 62, 66, 67, 68, 69, 72, 73, 81, 82, 84, 87, 90, and 106. The most prevalent HPV types were those associated with Heck disease; HPV-13 and -32, with a prevalence of 22.7% (n = 207). The data from the baseline findings have been published elsewhere (36).

At 12-month follow up, saliva samples were obtained from 743 of the original 1,011 participants, giving a retention rate of 73.5%. After testing for β-globin, 94 (12.6%) tested negative and were not included in subsequent analyses. All samples negative for β-globin at baseline and 12 months were excluded, yielding a final sample of 584.

Among these 584, 249 (42.6%) participants had no oral HPV infection at baseline and 12-month follow-up, 130 (22.2%) had new oral HPV infections at 12 months, 130 (22.2%) had persistent oral HPV infection (i.e., present at both baseline and 12 months), and 75 (12.8%) had oral HPV clearance from baseline to 12 months. A cumulative oral HPV infection incidence of 52.2% was observed.

There was an increase in the prevalence of HPV-13 and -32 from baseline to 12-month follow-up. Eight new cases of HPV-16 and one new case of HPV-18 were observed from baseline to 12-month follow-up. Among participants with persistent oral infection, just less than half (48%) were positive for HPV-13 or -32. The highest prevalence of oral HPV infection clearance from baseline to 12-month follow-up was observed among participants with HPV-13 or -32 (76.6%). The lowest prevalence was observed among participants with HPV-16 or -18 (18.6%). This information is portrayed in Tables 1 and 2.

The prevalence of oral HPV-13/32 at baseline was 22.7% which increased to 38% at the 12-month follow up. In this study, 79.4% of the new infections at 12 months, 51.4% of the clearance infections, and 64.6% of the persistent infections were positive for HPV-13/32.

The sociodemographic characteristics and health-related behaviors of participants (n = 130) with no HPV infection at baseline and incident oral HPV infection at 12-month follow up are shown in Table 3. Among females, 20.0% belonged to the group of new incident oral HPV infection. Nearly 25% were currently smoking nontobacco substances. Among participants with a new oral HPV infection at 12-month follow-up, 24.1% reported to have engaged in a sexual relationship after 16 years of age.

The sociodemographic characteristics and health-related behaviors of participants who had a persistent oral HPV infection from baseline to the 12-month follow up are described in Table 3. Among the participants of the study residing in regional locations, almost 26% belonged to the group of persistent oral HPV infection. Among this group of participants with persistent oral HPV infection 28.0% reported to have received oral sex.

The sociodemographic characteristics and health-related behaviors of participants who had cleared an oral HPV infection from baseline to 12-month follow up are described in Table 3. Among the participants of the study residing in regional locations, almost 17% belonged to the group of clearance of oral HPV infection, and almost 15% had a valid healthcare card. Of the 548 participants at 12-month follow up, among the group of clearance of infection 15.3% had never received oral sex.

In multivariate modelling, sociodemographic indicators significantly associated with incident oral HPV infection at 12 months included healthcare card ownership (Table 4, Model 1). Participants with a health care card were 1.4 times more susceptible of getting a new infection at 12 months. Significant health behavior indicators in Model 2 included less often to no alcohol consumption and current or former nontobacco substance use. Nontobacco consumption increased the risk of incidence by 1.6 times. Significant sexual behavior indicators in Model 3 included lower age of first giving oral sex, never used protection on receiving oral sex, and lower age of onset of sexual activity. In Model 4, sexual behavior indicators that remained significantly associated with new incident infections after adjusting for confounding included lower age of first giving oral sex and never using protection when receiving oral sex.

Multivariate modelling for persistent oral HPV infection shows the metropolitan location of residence as a significant sociodemographic indicator (Table 5, Model 1). Participants residing in metropolitan locations were 1.7 times more likely to show persistent oral HPV infection. Significant sexual behavior indicators in Model 2 included a higher age of first giving oral sex and a lower age of first onset of sexual activity. Participants who stated the age of onset of giving oral sex older than 16 years were 4.3 times more likely to present a persistent oral HPV infection. In Model 3, sexual behavior indicators that remained significantly associated with persistent oral HPV infections after adjusting for confounding included older age of first giving oral sex (4.3 times more likely) and younger age of first onset of sexual activity.

In multivariate modelling, sociodemographic indicators significantly associated with clearance of oral HPV infection at 12 months included regional location of residence and no healthcare card status (Table 6, Model 1). Significant health behavior indicators in Model 2 included alcohol consumption and nontobacco substance use. No significant sexual behavior indicators were observed in Model 3 and Model 4.

This study provides epidemiologic data on oral HPV infection incidence, persistence, and clearance rates in a large cohort of Indigenous South Australians across a follow-up period of 12 months. It also identifies risk factors for incidence, persistence, and clearance of oral HPV infection.

The baseline prevalence of oral HPV infection was 15.3 times than reported in a study of young Australians (non-Indigenous; ref. 39). At 12-month follow up, the prevalence increased from 35.2% to 42.1%, which is 8.7 times the estimate reported in a systematic review by Wood and colleagues (40). A longitudinal study (0, 6, 12, and 24 months) of 704 people from Brisbane (18–70 years old) reported an oral HPV prevalence of 10.7% at baseline among the 636 participants who tested positive for β-globin (41).

A higher prevalence of incident oral HPV infection was found among current smokers (64.2%). This is consistent with other reports (13, 16, 42). Investigators have reported that smoking may have an effect on the Langerhans cells, which are involved in instigating an immune response against invading pathogens within the epithelium, leading to a suppressed immune response (13). This smoking-induced immunosuppression leads to a high viral load, which has been previously correlated with HPV-associated cervical cancer (43–45) and oral diseases including periodontitis (46). An in vitro study by Alam and colleagues (47) demonstrated the effects of Benzo[a]pyrene (a major carcinogen in cigarette smoke) on cervical cells. They observed increased levels of virion synthesis in HPV-infected cell lines. Results with oral mucosal cell lines are not yet available but similar findings and effects can be anticipated.

Previous research has shown that younger age in women is associated with a higher rate of cervical HPV acquisition and clearance, with longer durations of infection in the older groups (48). There is extensive research on HPV-associated cervical neoplasias in women, but due to limited research on oral HPV in women, there is lack of comparative data to demonstrate the differences in age groups showing persistence and clearance in men and women. Most studies have highlighted the burden of oral HPV infection among men (16). In this study the oral HPV infection rates were higher (at both baseline and 12-month follow up) among women, across both younger and older demographic characteristics. These findings highlight a large gap in the HPV vaccine uptake by Indigenous Australian adolescents, and speaks to an urgent need to ensure high vaccination coverage among all Australians, both Indigenous and non-Indigenous. Although there is little evidence to show the efficacy of the vaccine against specific low-risk HPV types (like 13 or 32), there is sufficient indication to demonstrate an effect on decreasing the prevalence of oral HPV infections (49, 50).

The oral carriage of HPV types associated with Heck disease (HPV-13 or -32) in our study was higher than what has been presented in the literature to date (20, 39, 51–55). In this study, the age of participants carrying oral HPV-13/32 was 1.7 times higher (39.8 years) compared with a systematic review (23.1 years) involving 95 studies from other countries (20, 56). Studies have shown that the prevalence of HPV-13 or -32 differs according to residence location (rural or urban; ref. 56) and is common in populations with low annual incomes (26) and overcrowded households (25, 57). The high levels of HPV-13 or -32 may be associated with the Indigenous origin of the population sample in this study, as previous research has established a genetic and ethnic link with Indigeneity and Heck disease (51, 52, 58–62).

The most important risk indicators observed in this study included location of residence (regional or metropolitan), with an increased persistence observed in participants residing in regional locations and an increased clearance of infections in participants residing in Metropolitan locations. Early onset and unsafe (unprotected) oral sexual behaviors (giving or receiving sex) with were also significant indicators of persistent and new incident oral HPV infections.

This study had several important strengths. To the best of our knowledge, it was the first prospective, longitudinal cohort study that comprehensively examined and compared the risk factors of persistence and clearance of oral HPV infection in a large Indigenous Australian population. The main strength of the study is the engagement of South Australian Indigenous communities which was idealized through partnerships and involvement of the study's Indigenous Reference Group. Over 700 participants were followed up for 12 months. Previous research has focused on only one or a few HPV types, while our study covered a larger spectrum of high-risk and low-risk HPV types. A deeper understanding of these epidemiologic determinants helps in facilitating public education on early detection of oral HPV infection, and may contribute to the prevention of OPSCC. A comprehensive set of risk factors including sociodemographic variables, lifestyle factors, sexual history, and oral hygiene habits were examined. The findings could inform preventive initiatives and future research for oral HPV infections among Indigenous populations at a global level.

The main limitation is the lack of clinical examinations, anthropometrics, and blood samples that would yield important biomarker estimates. The study was not representative, with almost two thirds of participants being women. Higher oral HPV prevalence was associated with living in rural areas, but people living in these areas were overrepresented in our sample, which may have led to an overestimation. Some participants were lost to follow-up, meaning the impact on the persistence and clearance rates of oral HPV infection for the whole baseline sample remains unknown. Lastly, lifestyle habits and sexual behaviors were self-reported by study participants, and social desirability bias might have impacted these findings. The 12-month follow-up was delayed and ultimately ceased prematurely due to COVID-19 restrictions.

In this study, the overall prevalence of HPV detected in saliva samples, in a large convenience sample of Indigenous Australians was high, with a significant increase observed after a 12-month follow-up. The most prevalent HPV types were those associated with Heck disease (HPV-13 and HPV-32). Lower age of onset of oral sexual behaviors and decreased use of protection emerged as significant risk predictors for incident oral HPV infection at months. Higher ages of first giving oral sex and lower ages of first sexual experiences were significant risk predictors for persistent oral HPV infection at 12 months. Further funding will be pursued to continue follow-up of this cohort, and to include (after a full medical history) a thorough clinical examination of the external head and neck, a complete oral examination, and examination of the oropharynx.

X. Ju reports that this study was funded by Australia's National Health and Medical Research Council (NHMRC). K. Canfell is co-personal investigator (PI) of an investigator-initiated trial of cervical screening, Compass, run by the VCS Foundation, which is a government-funded not-for-profit charity; the VCS Foundation has received equipment and a funding contribution from Roche Molecular Diagnostics; in addition, K. Canfell is also co-PI on a major investigator-initiated implementation program Elimination of Cervical Cancer in the Western Pacific (ECCWP) which will receive support from the Minderoo Foundation, the Frazer Family Foundation, and equipment donations from Cepheid Inc. Neither K. Canfell nor her institution, on her behalf, receives direct funding from industry for any project. M.A. Smith reports grants from NHMRC and grants from Cancer Institute NSW during the conduct of the study. No disclosures were reported by the other authors.

S. Sethi: Conceptualization, resources, data curation, software, formal analysis, investigation, visualization, methodology, writing–original draft, writing–review and editing. X. Ju: Software, formal analysis, supervision, validation, methodology, project administration. A. Antonsson: Resources, software, funding acquisition, validation, visualization, writing–original draft, project administration, writing–review and editing. K. Canfell: Resources, software, formal analysis, supervision, visualization, methodology, project administration, writing–review and editing. M.A. Smith: Resources, data curation, supervision, validation, methodology, writing–original draft, project administration. G. Garvey: Resources, software, formal analysis, investigation, visualization, methodology, writing–review and editing. J. Hedges: Resources, data curation, formal analysis, supervision, validation, methodology, writing–original draft, project administration, writing–review and editing. L. Jamieson: Conceptualization, resources, software, formal analysis, supervision, funding acquisition, validation, investigation, methodology, project administration.

The authors gratefully acknowledge the support of the Indigenous Human Papillomavirus and Oropharyngeal Squamous Cell Carcinoma study participants, Indigenous Reference Group, staff who collected data, and key participating Aboriginal Community Controlled Health Organisations. G. Garvey was funded by a NHMRC Investigator Grant (no. 1176651)

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.