Prevalence of Human Papillomavirus (HPV) infection in women living in remote riverine communities in the Amazonas, Brazil

Human papillomavirus (HPV) is the most common sexually transmitted infection worldwide. Since some HPV types are associated with the development of cervical cancer, routine screening for HPV plays a crucial part in cancer prevention programs at global level. In Brazil, the cervical cancer rate is among the highest in the world, in particular in indigenous women living in remote areas with limited access to public health services. The objective of this study was to determine the prevalence of HPV infections and their co-occurrence with cervical lesions in Brazilian women living in isolated riverine communities. Moreover, the association between social environment, sexual activity and prevalence of HPV infection was assessed.


Background
Cervical cancer is one of the most common cancers affecting women worldwide 1 with >85% of the cancer-related deaths occurring in low-and middle-income countries. The high mortality in these regions is often attributed to limited access to preventive measures such as regular cervical cancer In the Brazilian Amazon, numerous indigenous communities live in remote areas with no or limited access to health care services, including cancer screening and treatment. Furthermore, low levels of education and low social status contribute further to the vulnerability of these remote populations.
The situation is particularly difficult for women as services relating to family planning and reproductive and maternal health are often insufficient or not existent. Despite better accessibility to general health care services in urban areas, the Brazilian health system is precarious and does not prioritise disease prevention and general health promotions.
Pap smears (cytology) are often recommended for cervical cancer screening 9 . Until recently, however, these tests were only available at the referral centre in the state capital Manaus, thus limiting the accessibility for women living in the Amazon region. Additional factors related to cultural practices, fear of taking tests and lack of knowledge about the importance of cancer prevention further contribute to the low adherence to regular gynaecological examinations 7,8 .
The "Doutores das Águas" Assistance Program, a non-governmental organisation created in 2011, is operating in the Amazon region, providing medical and dental care for these remote indigenous communities. In order to address the above mentioned issues the Women's Health Centre was introduced into the assistance program in 2017; providing women with comprehensive health assistance including cancer screening and detection of sexually transmitted infections (STIs).
In addition to cytology, and were accessible, methods for direct HPV detection are becoming increasingly important in the early diagnosis of cervical carcinomas. Through molecular HPV typing, women infected with high-risk types can be identified even before any pathological changes have developed. Subsequently, by offering these women the possibility to participate in more frequent gynaecological screenings, the risk, and hence the rate of cervical cancer could be reduced substantially. However, knowledge regarding the epidemiology of intracervical epithelial abnormalities and HPV is scarce, in particular among indigenous women living in remote areas of Brazil. Therefore, the aim of this study was to determine the prevalence of cervical abnormalities and high/low risk HPV types in women living in riverine communities of the Amazon Basin. Furthermore, social and behavioural determinants, associated with an increased risk of HPV infections and cervical cancer development were investigated. The knowledge generated here will play a crucial part in the development and implementation of an improved women's health program.

Ethical approval and informed consent
Ethical approval was obtained from the ethics committee of Santo Amaro University São Paulo (Brazil 4 Platform -CAAE: 61414216.4.0000.0081). Written informed consent was obtained from all study participants. For participants < 18 years, the written consent was duly signed by the parents or guardians. The scheduling and recruitment of women, follow-up of collection procedures and organization of the site during the activities was supervised by the principal investigator while the sample collections were performed and supervised by the health team of the Women's Health Program. Women with cytological alterations received referral for further treatment by the responsible health agent from the local program.

Study area and population
The riverine populations investigated here inhabit the vast and complex waterways that are characteristic of the Amazon Biome, a set of ecosystems interconnected by the Amazon rainforest and the Amazon Basin. These communities share common characteristics such as economic and social isolation and exclusion from public services, basic health care and education 10,11 . Community leadership is predominantly female, although the family model is still patriarchal in communities further away from large centers. They live off traditional fishing, hunting and agricultural activities.
Their houses are built on stilts near the riverbanks and lack electricity, piped water and basic sanitation. Rafts or boats are the only means of transport in this region. In a recent decree, the existence of these traditional peoples and communities was recognized, and a national policy, focused on improving the living conditions of these peoples was instituted 13 . This cross-sectional study included women living in riverine communities along the two main tributaries of the Amazon River; Rio Madeira and Rio Negro and was performed in April 2017. All nonpregnant, non-menstruating, sexually active women attending the gynecological clinic were invited to participate in this study. None of the women were vaccinated against HPV.

Socio-demographic and clinical data
Study participants were in-depth interviewed by a team of psychologists specialized in conducting gynecological anamnesis. The women were asked to answer questions regarding socio-demographic variables as well as behavioral/psychological determinants using a detailed questionnaire

Sample collection
The cervico-vaginal material was collected from the ectocervical (ectocervix and vagina) and endocervical regions using a standard cytobrush, then conserved in preservative medium (CellPreserv ™, Kolplast™ ci Ltda, Braziland stored at XX°C until further cytological and molecular analysis.

Cytological analysis
Cervical cell suspensions were fixed and processed using a KLP 2000 slide processor (Kolplast™ ci Ltda, Brazil) according to manufacturer's instructions. This was followed by Papanicolaou (Pap) staining and analysis by optical microscopy. The slides were analyzed by two experienced, blinded cytologists and classified according to the Bethesda system.

HPV genotyping
DNA was extracted from cervical cell suspensions using PureLink™ Microbiome DNA Purification kit (Invitrogen, USA) according to the manufacturer's instructions. DNA concentration and purity was determined using nanodrop ND-ONE-W (Invitrogen, USA). HPV genotyping was performed using EUROArray HPV (Euroimmun, Germany). Briefly, the HPV array is based on the detection of the viral oncogenes E6/E7 via multiplex PCR amplification and hybridization of the fluorescently labelled products with immobilized DNA probes. The use of type-specific primers and probes allows for simultaneous detection and typing of 30 anogenital high-risk (n=18) and low-risk (n=12) HPV types in one test run. Data analysis and interpretation was performed fully automated using the EUROArrayScan software according to the manufacturer's instructions.

Data analysis
Standard descriptive analyses were performed, including frequency distributions for categorical variables and calculation of means and standard deviations (SD). Contingency tables were analyzed by either chi-square tests (χ 2 ) or Fisher exact tests using SPSS, version 20. The level of significance 6 was 0.05.
The women were on average 14.9 (±2.8) years at sexarche (based on data from 122 women) and 17.5 (± 3.8) years old at first pregnancy (based on data from 90 women). In total 105 (86.1%) of the women had children and nearly half of them (n=52, 49.5%) had 4 or more (range 1 to 14, based on data from 122 women). Furthermore, most of the women (n=99, 80.5 %) were married or living with a partner. Sixteen women (13.1%) reported using condoms.
The women living in the Rio Madeira communities had fewer lifetime sexual partners than the women living in the communities in the Rio Negro area (Table 1., p <0.001). There were no other significant differences in socio-demographical or behavioural variables between women from the two regions (Table 1.); hence subsequent analyses regarding risk factors and HPV prevalence were performed for all women combined.

Cytological evaluation
Among the 121 cervical smear samples considered suitable for cytopathological evaluation, the 7 overall prevalence of epithelial cell abnormalities amounted to 9.9% (Table 2

HPV infection prevalence
The overall HPV prevalence amounted to 25.4%. There was no significant difference in HPV infection prevalence between the two regions (p=0.4). Infections with multiple HPV types were more common in women from Rio Madeira compared to women from Rio Negro (57.1% vs. 17.6%, p=0.03).

Determinants of risk factors for HPV infections
Among all women, the prevalence of HPV infections was significantly higher in women younger than 25 years (41.0%) and women older than 50 years (31.3%) compared to women between 26 and 49 years (14.3%) (p=0.006, based on data from 118 women). HPV infections were more common in single (42.1% infected) than in married women (21.4% infected) but the difference was however not significant. Moreover, the sexarche was not associated with HPV prevalence (p=0.6). In total 27.7% of the women that had first sexual intercourse at the age of 15 years or less were infected with HPV, while 21.1% were HPV positive in the group of women older than 16 years at sexarche (based on data from 121 women). No significant associations between income (≤1 salary vs. > 1 salary), number of lifetime sexual partners (1, 2-3, or ≥ 3 ), condom use, smoking or parity (0, 1-3 or >3) and prevalence of HPV infection were found (Table 5.).

Discussion
The Amazonas state in northern Brazil is an area with a high incidence rate of cervical cancer 14  The association between social and behavioural factors and HPV prevalence was investigated. The women's age was associated with the prevalence of HPV infection (p = 0.006), with infections being more common in women aged 25 years or less and 50 years or more when compared to women aged 26 to 49 years. Similar age-dependent patterns have previously been described 29 ,. The higher prevalence in older women could possibly be due to reactivation of latent infections due to the gradual loss of type-specific immunity or changes in the pattern of sexual behaviour over the age 30 .
In total 8,1% of the women were smokers. There are divergences in the literature regarding smoking and increased susceptibility to HPV infections; however, it has been implicated as a cofactor associated for the development of high-risk lesions (HSIL) and cervical carcinoma (REF). Furthermore, smoking appears to increase cervical and systemic susceptibility to general infections and HPV 6,24 .
Here, however, the relationship between smoking and the prevalence of HPV infections was not statistically significant (Table 5.). Similarly, condom use was not significantly associated with HPV prevalence. Even though the overall condom use was rather low here (13.1%), the results corroborates the literature which states that sexual contamination can only be avoided through complete sexual abstinence, since condoms do not guarantee full protection and the virus can still be transmitted via non-penetrative sex. In addition, different variables influence condom use such as age, number of sexual partners, knowledge of correct use, accessibility, partner age and the ability to negotiate condom use with partner. 25,26 Furthermore, the sexarche was not significantly associated with HPV prevalence,, thus contradicting literature which considers early sexual initiation as a factor related to a larger number of sexual partners, favouring the acquisition of STDs and early pregnancy 32 . In addition, as shown previously, early onset of sexual activity and teenage pregnancy increase the risk of developing cervical cancer 33 . Furthermore, the literature states that having multiple sexual partners is a risk factor for contracting HPV 35,36 . Here, the Rio Negro women had significantly more sex partners than the Rio Madeira women (average 6 vs. 1, p<0.001), there was however no significant association between the number of lifetime sexual partners and the prevalence of HPV (Table 5.). Furthermore, more single women were infected with HPV compared to married women in this study, but the difference was not significant. Previous studies have also associated high parity as a factor for cervical carcinogenesis 37,38 . Here, HPV was more prevalent in women that never have had children compared to multiparous women; the difference was however not significant and likely associated with that these women were younger.

Conclusion
Here we have shown that indigenous women living in remote communities near the Rio Madeira and the Rio Negro tributaries of the Amazon River, are frequently infected with high-risk HPV and are is such at a risk of developing intraepithelial lesions which could result in neoplastic transformation and cervical cancer. Notably, almost 10% of the women investigated here displayed cervical epithelial abnormalities. Furthermore, although not significantly associated with HPV infection prevalence in this study, risk factors and behaviours such as low level of education, early sexarche and early pregnancy, low condom use, high parity and high number of sexual partners (Rio Negro area only) characterized these populations. In addition, the high prevalence of cellular alterations associated with inflammation; suggestively due to other infectious agents further emphasizes the importance of an

Supplementary Files
This is a list of supplementary files associated with this preprint. Click to download. Tables.pdf