Cervical and oral human papillomavirus infection in women living with human immunodeficiency virus (HIV) and matched HIV-negative controls in Brazil

Despite the demonstrated role of human Papillomavirus (HPV) in the etiology of cervical cancer and the strong evidence suggesting the importance of HPV in the development of oropharyngeal cancer, several aspects of the interrelationship between HPV infection in both body sites remain unknown, specifically in female human immunodeficiency virus (HIV)-positive (HIV+) patients. We aimed to assess the prevalence, distribution, and concordance of cervical and oral HPV in HIV+ women and matched HIV-negative (HIV-) controls in Brazil. Cervical and endocervical samples for cytological screening and HPV detection and oral samples were collected from 115 HIV+ women using highly active antiretroviral therapy (HAART) and 139 HIV-matched controls (HIV-) in Maringá City, Brazil. Risk factors were assessed using a standardized questionnaire, and the data regarding HIV infection were obtained from the patients’ medical records. HPV detection and typing were performed using the Kit Multiplex XGEN Multi HPV Chip HS12. HIV infection was well controlled in this cohort, but women who exhibited detectable HIV loads were significantly associated with HPV-positive status overall (P = 0.03) and in cervical mucosa (P = 0.01). HIV+ women had significantly more abnormal cytological findings (P = 0.04) than HIV- women. Of the 115 HIV+ women, 48.7% were positive for cervical and/or oral HPV DNA; of the 139 HIV- women, 41% were positive for cervical and/or oral HPV (P = 0.25). Both HIV+ and HIV- women had a statistically higher prevalence of cervical HPV infection than oral infection. The concurrent HPV infection in two anatomical sites was similar in HIV+ and HIV- women; however, HPV type concordance was not observed. HPV type distribution was different between the anatomical sites in both groups, and HIV+ women presented less common types, mainly in oral mucosa. Our data support the importance of testing HPV infection in HIV+ women, even when the HIV infection is well controlled. Prospective studies are required to better understand the natural history of HPV infection in both anatomical sites, specifically in HIV+ women.


Background
The association between persistent high-risk human Papillomavirus (hrHPV), squamous cervical cancer (CC), and some vaginal and anal cancers has been well-established [1][2][3]. Additionally, recent data demonstrated that HPV is also associated with a subset of head and neck cancers (HNCs), including a worldwide range~20-80% of oropharyngeal cancers (OPCs) [4,5]. Individuals living with human immunodeficiency virus (HIV+) are more susceptible to infection, less likely to clear the virus and have a higher risk of HPV-related cancers than HIV-negative (HIV-) individuals [6,7]. Although the incidence of overall cancer has decreased in HIV+ individuals with the advent of highly active antiretroviral therapy (HAART), HPVrelated CC risk is higher among HIV+ individuals [8,9]. Moreover, in HIV+ women, CC tends to respond poorly to recommended therapies, becomes more aggressive, and may have a worse prognosis [6]. With the increase in survival of HIV+ women due to HAART, progression of oncogenic viral infection into malignancy may result in an increased incidence of HPV-associated oropharyngeal, genital, and anal cancers [8][9][10].
Currently, the HIV/acquired immunodeficiency syndrome (AIDS) pandemic impacts the poorest and the youngest in low-resource settings [11]. HIV+ young women are key populations at high risk for developing HPV-related cancers, particularly in low-and middle-income countries such as Brazil [12,13]. With the recent approval of the 9valent HPV vaccine, the risks of persistent HPV infection and HPV-related cervical precancerous lesions and malignancies are expected to decrease significantly [14]. With the appropriate vaccination rate thresholds, this vaccine provides a logical rationale for increasing screening intervals considering the anticipated decrease in the burden of disease [15]. Therefore, understanding the specific prevalence, distribution and concordance of cervical and oral HPV in at-risk women in specific geographic locations will provide further insight into the effectiveness of this new vaccine against oral HPV infection, which has not been proven yet. Considering the high burden of HPV-related cancers among HIV+ women and the possible effectiveness of a 9-valent HPV vaccine [16], it is critical to understand the prevalence and types of HPV infections in oral and cervical mucosa in HIV+ women (and matched controls). However, while several studies on HIV+ women have reported cervical, oral, or anal HPV type distribution [17][18][19][20][21][22][23], only a few studies have addressed concurrent cervical and oral HPV prevalence [24][25][26]. Despite the demonstrated role of HPV in the etiology of CC and the strong evidence suggesting the importance of HPV in the development of OPC [27][28][29][30], several aspects of the interrelationship between oral and cervical infections remain unknown, specifically in HIV+ female patients.
The present study aimed to assess the prevalence, distribution, and concordance of cervical and oral HPV in HIV+ women and matched HIV-controls in the southern region of Brazil, a geographic area with a high incidence of HIV and CC.

Study population
Participants included 115 HIV+ women receiving HAART and 139 HIV-aged 19 to 66 years who attended the Specialized Assistance Service (SAE) for sexually transmitted diseases (STD/AIDS in Maringá, southern Brazil) from September 2017 to May 2018. Women with the following characteristics were included in this study: women with confirmed HIV/AIDS diagnosis using two different methods were included in the HIV+ group and women with two HIV/AIDS negative results using two different methods were included in the HIV-group. Exclusion criteria include: women with previous hysterectomy, pregnant, younger than 19 years, and women with no history of sexual intercourse.
Of the 778 HIV+ women enrolled in the SAE, 324 were eligible for the study. The sample size was calculated with a HPV cervical prevalence of 50% in HIV+ women [19], a 95% confidence interval (CI), and an error estimate of 5%. With an increase of 10% for possible participant losses, the total sample size was fixed at 138 randomly selected women. SAE also provides other services and HIV screening. Therefore, to obtain a comparable sample, 138 matched controls from the list of patients served by the SAE were identified. The controls were matched by age. Written informed consent was obtained from 254 women, and based on their HIV serology status, participants were assigned to the HIV+ (n = 115) or HIV-group (n = 139). Participants were interviewed using a standardized questionnaire to obtain demographic information (e.g., age, educational attainment, household income, race/ethnicity); tobacco and alcohol use/abuse; obstetric and gynecologic history (e.g.,

Sample collection
Nursing staff contacted all women, administered the questionnaire, and collected the cervical and oropharyngeal samples. Ecto−/endocervical samples were collected using an Ayre's spatula and cytobrush for cervical cytology and polymerase chain reaction (PCR) for HPV; the samples for HPV testing were stored in ThinPrep® Pap Test solution. The conventional cytological smears were sent to the Clinical Cytology Laboratory at State University of Maringá (UEM) and were graded according to the Bethesda System [31]. Full-mouth oral/oropharyngeal scraping, including the cheeks, tongue, palate, tonsils, and oropharynx, was performed using a sterile brush with soft bristles; samples were stored in ThinPrep® Pap Test solution. Moreover, a gargle sample was obtained by having the participant gargle with 10 mL of 0.9% sterile saline for a total of 30 s (10s, rinse; 5 s, gargle; 10s rinse; 5 s, gargle), collected in a sterile cup and stored at − 4°C [20,32]. All oral/oropharyngeal samples will be referred as "oral samples."

HPV DNA testing
Detection and typing of HPV were performed using the Kit Multiplex XGEN MULTI HPV CHIP HS12 (Mobius Life Science), following the manufacturer's instructions. ). An additional HPV universal probe was used to detect other, non-specified types of HPV. Samples with invalid outcomes were retested, and the second result was considered definitive. Women were considered positive for oropharyngeal HPV if one of the samples, oral and/ or scraping, was positive for HPV. Furthermore, women were considered negative for oropharyngeal HPV if their oral and scraping samples were negative for HPV.

Statistical analysis
Statistical analyses were performed using the Graph-Pad Prism 6.0 (San Diego, California, USA) software. All variables were expressed as absolute and relative frequencies. For univariate analysis (unadjusted odds ratios [ORs]), categorical variables were compared with HPV infection using the chi-squared and Fisher's exact test. Crude ORs and 95% confidence intervals (CIs) were calculated. A P-value < 0.05 was considered significant.

Results
A total of 254 (115 HIV+ and 139 HIV-) women were included in the study, with all 254 cervical and 508 oral samples (oral scraping and gargle) having sufficient DNA for HPV assessment. All 254 (100%) participants had conclusive HPV test results from both anatomical sites.

HAART compliance
Yes 93 (80.9) --- HAART, highly active antiretroviral therapy; OR = Odds ratio; CI = confidence interval Demographic and clinical characteristics of the two groups of patients are presented in Table 1. The median age was 42.17 ± 10.18 years old for HIV+ women and 41.4 ± 12.31 years old for HIV-women. Compared to HIV-women, HIV+ women were significantly more likely to have less than 8 years of schooling (P = 0.0003), non-white skin color (P = 0.009), their first sexual intercourse at < 18 years old (P = 0.04), more than two sexual partners (P = 0.004 for 2-7 partners and P = 0.007 for > 7 partners), and a higher number of parities (P = 0.0006 for 1-2 parities and P 0.0001 for ≥3 parities), and were less likely to report screening for CC within the past three years (P = 0.01). HIV-women were, however, more likely to be a widowed (P = 0.004).
Cytology showed no signs of malignancy in most women from both groups. Overall, 13.0% of HIV+ women and 5.0% of HIV-women presented abnormal cytological findings (P = 0.04). Atypical squamous cells of undetermined significance (ASC-US) were observed in 2.6% of HIV+ women and in 1.4% of HIV-women (P = 0.66); lowgrade squamous intraepithelial lesions (LSIL) were observed in 7.8% of HIV+ women and in 2.8% of HIVwomen (P = 0.15); high-grade squamous intraepithelial lesions (HSIL) were observed in 2.6% of HIV+ women and 0.8% of HIV-women (P = 0.33).
Overall, 56 (48.7%) of the 115 HIV+ women were positive for cervical and/or oral HPV DNA, while 57 (41%) of the 139 HIV-women were positive for cervical and/or oral HPV (P = 0.25). Both HIV+ and HIV-women had a statistically higher prevalence of cervical HPV infection than oral infection, including higher rates of hrHPV, lrHPV, universal HPV, and infection by multiple HPV types in cervical samples compared to oral ones, as shown in Table 2.
HPV DNA was detected in oral samples from 17 (14.8%) HIV+ women and 13 (9.4%) HIV-women (P = 0.24) ( Table 3). Multiple HPV infections were detected in five samples (4.4%) from HIV+ women and in four samples (2.9%) from HIV-women (Table 2). Statistical analysis did not reveal an association between HIV+ status and the presence of HPV DNA in the oral mucosa (P = 0.83).
Fifty-one (44.3%) HIV+ women and 52 (37.4%) HIVwomen had HPV-positive cervical samples (P = 0.48) ( Table 3). Multiple HPV types were detected in 28 (24.3%) of HIV+ and in 29 (20.9%) of HIV-patients (Table 2). There was no significant difference between the HIV+ and HIV-women with regard to HPV types. To better understand the association between cervical and oral HPV infection, the prevalence of concurrent HPV infection in this population was investigated. Eight (7.0%) HIV+ women and 7 (5.0%) HIV-women had concurrent HPV infection in their cervical and oral samples (P = 0.6) ( Table 3). Concordance between HPV types in the cervical and oral samples was not observed in either group.
When the characteristics of the HIV+ women were analyzed in relation to HPV status, only current smoking was associated with overall (OR = 3.80; CI = 1.22-10.42; P = 0.01) and cervical HPV-positive status (OR = 3.94; CI = 1.31-11.76; P = 0.01 for both), as shown in Table 5.

Discussion
In the present study, we aimed to determine the HPV prevalence, distribution, and type concordance between cervical and oral samples of HIV+ women and HIVmatched controls in the southern region of Brazil, a geographic area with high incidences of HIV and CC. Our data demonstrated that HIV+ and HIV-women had a similar and high HPV prevalence in cervical and in oral sites. However, HIV+ women had higher prevalence of abnormal cytological findings than HIV-women. HPV type distribution was different between the anatomical sites within both groups, and HIV+ women commonly presented with narrower HPV type distributions, mainly in the oral mucosa. Finally, the frequency of concurrent HPV infection in both sites was low, and HPV type concordance was not observed.
Our results showed that the prevalence of cervical HPV infection was 44.3% in HIV+ women and 37.4% in HIV-women. Studies across different populations present varying rates of HPV infection: .1% in HIV+ women and 44.9-86.5% in HIV-women in the cervical site [20,24,25,33]. The high prevalence of HPV in HIV-negative women in this study can be explained, at least in part, by the fact they were recruited from a Specialized Assistance Service (SAE) for sexually transmitted diseases, and were therefore at higher risk for sexually transmitted infections, including HPV. Additionally, the rates of oral HPV infection in HIV+ women and HIV-control were 14.8 and 9.4%, respectively. Other studies have shown that the prevalence of HPV infection in the oral site can vary significantly (12-68.5% in HIV+ women and 2-31.4% in HIV-women) [20,24,33].
Although HIV infections were well controlled and cytology results showed no malignancies, abnormal cytological findings were significantly higher in HIV+ women than in HIV-women. These data are consistent with the results of the previous studies showing that HIV+ women are less likely to eliminate the virus, have subsequent persistent hrHPV infections and have higher risk of developing precancerous lesions and malignancies, even with the appropriate use of antiretroviral therapy, than HIV-women [8,9,19,34,35].
Our results demonstrated no significant difference in HPV detection in the uterine cervix or oral mucosa between HIV+ women and HIV-women; however, HPV infection was significantly higher in the cervical mucosa than in the oral site in both groups. These findings are consistent with previous studies suggesting that the natural history of HPV infection varies by anatomical site, and a higher prevalence of HPV infection is observed in the cervical mucosa than in the oral mucosa in HIV+ women [36][37][38]. This can be explained, at least in part,  by the evidence that the oral cavity is a hostile environment for the establishment of infectious agents due to the presence of both mechanical and molecular mechanisms related to digestion [39]. More specifically, Fakhry et al. [39] conducted a study that directly compared the local immunologic profiles of the oral cavities and cervices of healthy women using paired secretion specimens. This study showed that the oral cavity contained significantly higher concentrations of immunoregulatory factors that were related to the adaptive and cellmediated immune response than the cervix, which may in part explain the significantly lower burden of sexually transmitted infections such as Chlamydia trachomatis, HPV, and HIV-1 in the oral cavity than in the cervix. According to the authors, these findings provide additional information to better understand the differences in the etiology and natural history of pathogenic agents that are capable of colonizing both the oral cavity and female reproductive tract. The most prevalent HPV types in the cervical samples from both HIV+ and HIV-women were hrHPV18 and hrHPV58; however, hrHPV45 and hrHPV16 were frequently detected in HIV+ women and HIV-women, respectively. Prevalence studies around the world have shown that hrHPV types 16,18,31,33,35,52, and 58 are the most commonly detected hrHPVs in CC, with hrHPV16 being the most common in all populations, with the exception of HIV+ people [1]. Data have consistently shown that HIV+ women are frequently more infected with other hrHPV types than hrHPV16 and hrHPV18, such as hrHPV52 and hrHPV58 [40]. The high prevalence of non-vaccine hrHPV types of 2-valent and 4-valent vaccines in the cervical and oral mucosa found in our study suggests that the 9-valent HPV vaccine is significantly required, which is considered important to reduce the risk of developing HPV-related cancers, specifically in the HIV+ population.
Unlike the distribution of HPV types in cervical samples, HPV type concordance between the HIV+ and HIV-groups was not observed in oral samples. hrHPV39 was the most common hrHPV detected in oral samples of HIV+ women, followed by hrHPV18, hrHPV45, hrHPV52, and hrHPV68, with these types being also frequently found in cervical samples in the same group. However, in HIV-women, the oral hrHPV types observed were totally different, with hrHPV51 and hrHPV66 being the most frequently detected hrHPV. Current evidence has shown that hrHPV16 and hrHPV18 contribute to the majority (approximately 85%) of HNC cases worldwide, while the remaining cancers are caused by hrHPV33, hrHPV35, hrHPV52, hrHPV45, hrHPV39, and hrHPV58 [41,42].
Examination of concurrent HPV infections in different anatomical sites has been limited. Hence, we concurrently investigated the prevalence of HPV infection and HPV type distribution in cervical and oral sites to better understand their clinical significance. Simultaneous HPV cervical and oral infection was low in both HIV+ and HIV-women. Moreover, HPV type concordance was not observed, a finding consistent with previous studies [20,24]. Taken together, these observations suggest that various HPV types are more likely to occur in a different anatomical sites, and/or that the two anatomical sites can clear certain HPV types or distinct exposures, while allowing other HPV types to cause persistent oral and cervical HPV infections.
Prior investigations have demonstrated that engagement in some high-risk behaviors may facilitate HPV infection and act as a cofactor for viral persistence, contributing to cancer progression in the oral and cervical mucosa. In the present study, we found that current smoking was associated with overall and cervical HPV infection in HIV+ women. Several studies have already demonstrated that smoking status acts as a predictor and cofactor in cervical HPV infection, which is possibly due to the alteration of the mucosa cells, making them more susceptible to infection, changing the immune mediators, causing DNA damage, and promoting the integration of HPV DNA into the host genome [42][43][44][45].

Conclusion and future perspectives
Our study demonstrated that HIV+ and HIV-women in southern Brazil had high HPV prevalence in cervical and oral sites. In this population, HIV+ women had more abnormal cytological findings than HIV-women. This study provides important epidemiological data about the potential risk of developing CC. Furthermore, the high prevalence of non-vaccine hrHPV types of the 2-valent and 4-valent vaccines in the cervical and oral mucosa of   HIV+ and mainly in HIV-women found in our study highlights the importance of the 9-valent vaccine. Concurrent HPV infection in both sites was uncommon, and HPV type concordance was not observed, likely reflecting the differences in the risk factors and the natural history of HPV infection at the two anatomical sites. These findings confirm the prevalence of HPV infection in HIV+ women. Prospective studies are required to better understand the natural history of HPV infection in both anatomical sites, specifically in HIV+ women, and the impact of vaccination programs in at-risk groups.