Prevalence of vaginal HPV infection among adolescent and early adult girls in Jos, North-Central Nigeria

We randomly selected 15 out of 300 registered high schools in Jos, North Central Nigeria for this study and received permission to conduct the study in 10 (66.7%) of the schools. Between May and November 2016, we contacted 700 female students and their parents, of whom 232 girls and their parents agreed to participate in the study giving a 33.1% response rate. We obtained consent from the parents of girls younger than 18 years of age and assent from the girls. Older girls gave consent in their self-cognizance.

We conducted face-to-face interview with the girls in company of female nurses or researchers using structured questionnaires that covered socioeconomic factors, family characteristics, gynecologic history, sexual hygiene and practices. Questionnaire designs were guided by the research questions and literature review. The questionnaires were pre-tested on 10 individuals to evaluate their performance and corrected as required.

Prior to sample collection, we trained study participants on how to self-obtain vaginal samples using sterile swab sticks—a small cotton swab on a wooden handle packaged in an individual reclosable plastic pack. After the training, each girl was given 2 swab sticks and instructed to:

The first collected vulvo-vaginal swab was immediately stored on ice pack in a cooler until transported to the Plateau State Human Virology Research Centre (PLASVIREC) sample repository for storage at −80 °C, usually within three hours of collection. The second swab was immediately applied on Hydrion pH test paper (pH 2.8–4.6, 4.5–6.1 & 5.5–8.0) (Micro Essential Lab. Inc. New York) the colour change was compared with the provided standards.

Collected samples were transported to the African Collaborative Centre for Microbiome and Genomics Research (ACCME), Institute of Human Virology, Abuja, Nigeria for HPV detection and genotyping. To prepare the vulvo-vaginal swab for DNA extraction, 1 millilitre of PBS was added to the dry swabs inside the cryovials and vortexed to dislodge the cells from the cotton wool.

DNA extraction was carried out using Cador® Pathogen 96 QIAcube® HT Kit (Qiagen, Germany) on QIAcube HT (QIAGEN, Germany). The DNA was amplified and HPV detected using SPF₁₀ (DDL Diagnostic Laboratories, The Netherlands) which simultaneously detects HPV types 2, 3, 4, 5, 6, 7, 8, 11, 13, 14, 16, 18, 20, 26, 27, 28, 30, 31, 32, 33, 34, 35, 37, 39, 40, 42, 43, 44, 45, 51, 52, 53, 54, 55 (re-classified as a subtype of HPV44), 56, 57, 58, 59, 61, 62, 64 (re-classified as a subtype of HPV34), 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 81, 82, 83, 84, 85, 86, 87, 89, 90, 91, 95, 97, 102, 106, 114 and 115. Positive samples were tested with reverse hybridization line probe assay (LiPA₂₅) which can simultaneously identify 25 HPV types (6, 11, 16, 18, 31, 33, 34, 35, 39, 40, 42, 43, 44, 45, 51, 52, 53, 54, 56, 58, 59, 66, 68–73, 70, and 74) according to the manufacturer’s (DDL Laboratories, Netherlands) instructions. This is a highly sensitive broad-spectrum PCR-based assay for qualitative detection of HPV and suitable for epidemiological studies.

The study protocol was approved by the institutional review board of the Jos University Teaching Hospital (JUTH). Written informed consent was obtained from the parent/guardian of participants who were less than 18 year of age and assent obtained from the participants themselves. For participants who were greater than or equal 18 years of age, they provided written informed consent prior to enrolment.

We entered data into REDCap electronic database and analyzed using Stata version 14 (StataCorp, College Station, Texas USA). High-risk HPV types were defined as types 16, 18, 26, 31, 33, 35, 39, 45, 51, 52, 53, 56, 58, 59, 66, 88, 73 and 82, including the probable hrHPV types and low-risk types were defined as types 6, 11, 40, 42, 43, 44, 54, 61, 68/73, 70, 72, 74, 81 and CP6108 [20]. Univariate analysis was done using frequency distribution and proportion for categorical variables and descriptive statistics for continuous variables. Bivariate analysis to test for association was done using chi square or fisher’s exact test for categorical variables and t-test for continuous variables; means and standard deviations of continuous variables were computed. A p-value < 0.20 was used as a criterion for the inclusion of variables in the multivariable analysis. We set a p-value < 0.05 as statistically significant.

We excluded 27 participants (11.6%) because they were unable to collect vulvo-vaginal swab leaving 205 participants. The mean age (SD) of the excluded girls was 14.7 (1.97) years while that of the remaining study participants was 14.9 (2.34) years (p-value 0.67). Most of the excluded participants (17, 63.0%) were from public schools while the remaining ten participants (37.0%) were from private schools. About half of the remaining participants, (104/205, 50.7%), were recruited from public schools while 49.3% (101/205) were from private schools (p-value 0.24). There were no significant differences between girls who were unable to collect their samples and were excluded from the study and those who remained in the study.

Some 13.2% (27/205) of the girls had any HPV infection. There was no difference in type of school attended and socio-economic status of HPV positive compared with HPV negative girls. Most (145/205, 70.7%) of the girls in this study had attained menarche and there was significant difference in menarcheal status comparing HPV positive (88.9%) and HPV negative (68.0%) girls (p = 0.03). Only 9.3% (19/186) of the girls reported positive history of sexual intercourse and this was significantly different (p = 0.01) comparing HPV positive (22.2%) and HPV negative (7.3%) girls. The mean (SD) age at first sexual intercourse for girls who reported positive sexual history was 13.3 (3.58) years, and this was of marginal statistical significance comparing HPV positive (12.0(3.46)) and HPV negative (16.2(1.72) girls (p = 0.07). Most of the girls (57.9%, 11/19) who reported sexual intercourse first experienced it at 15 years of age or less and 15.8% (3/19) experienced it before the age of 10 years. Among HPV positive girls, most (83.3%, 5/6) of those who have ever had sex had it after 15 years of age compared to only 23.1% (3/13) of HPV negative girls (p = 0.04). Most of the girls with sexual experience (17/19, 89.5%) reported one or two lifetime sexual partners while two girls (10.5%) reported more than two sexual partners. All the sexually experienced HPV negative girls had one or two sexual partners while 66.7% (4/6) of the HPV positive girls had one or two sexual partners and the rest (33.3%) have had three or four sexual partners (p = 0.09). Of the girls who had positive sexual history, 11 (57.9%) reported that the sex was forced and without their consent but the prevalence of this not different comparing HPV positive and HPV negative girls (p = 1.00). Among those girls with positive sexual history who reported the age of their sexual partners, most (6/8, 75.0%) were older than 21 years of age, as were the sexual partners of all the HPV positive girls compared to only 33.3% (1/3) of the HPV negative girls (p = 0.11).

We did not observe any significant differences or trends in the last time the girls had sexual contact, their partner’s marital status (where known), history of masturbation, use of toys for masturbation, use of condoms, douching, and pH of the vagina comparing HPV positive and HPV negative girls. The girls in this study discussed sex more with others than with their mothers or fathers. Among the others they discussed sex with, the most common were friends, aunts, and sisters. They had low level of knowledge about HPV infection or HPV vaccination (Table 1).

The prevalence of any HPV infection in this study was 13.2% (27/205), while that of hrHPV was 59.3% (16/27), lrHPV was 55.6% (15/27), mixed high and low risk HPV was 14.8% (4/27) and multiple hrHPV was 22.2% (6/27). The age-specific and age group prevalence of any HPV among study participants are shown in Fig. 1. The first instance of any HPV infection among the girls in this study was at 10 years of age and it was a low-risk HPV infection. The first instance of high-risk HPV infection was observed at 12 years of age. The peak age prevalence of anyHPV, hrHPV, lrHPV and multiple hrHPV was 16 years of age. At this age, the prevalence of anyHPV was 6.8%, while that of hrHPV was 4.9%, lrHPV was 2.6% and multiple hrHPV was 2.4%. The prevalence of anyHPV, hrHPV, lrHPV and multiple hrHPV was highest in the 13–16 age group. The prevalence of anyHPV infection was 4.9% among 9–12 years age group, 15.1% among 13–16 years and 9.3% among 17–20 years old while that of hrHPV infection was 3.4% among 9–12 years age group, 7.8% among 13–16 years and 5.4% among 17–20 years old. The prevalence of lrHPV was 2.0% in the 9–12 years age group, 8.8% in 13–16 years age group and 5.4% in the 17 – 20 years age group while the prevalence of multiple HPV infections is 1.5% in the 9–12 years age group, 3.4% in 13–16 years age group and 2.4% in the 17–20 years age group.

The prevalence of unknown HPV types found in this study population was 2.4%, suggesting substantial prevalence of HPV types beyond those detected using SPF₁₀LiPA₂₅ in this study population. We detected nine hrHPV types and the commonest hrHPV types were 52 (3.9%), 18 (1.5%) and 51 (2.4%. The other hrHPV types detected in this study were 58 (1.0%), 66 (1.0%), 45 (0.5%), and 31 (0.5%). We did not detect any HPV type 16 in this study sample. The most common lrHPV type we identified was type 6 (2.9%) followed by type 44 (1.0%). Other lrHPV types detected were types 68/73 (0.5%), and 74 (0.5). We did not detect lrHPV type 11 among this cohort of adolescent girls. The distribution of these HPV types in relation to the 27 positive participants is shown in Fig. 2.

The age at first detection of HPV types 18, 45, 66, 6, and 74 was 16 years, while HPV types 51, 52, and 58 was detected at 12 years. HPV types 31, 44 and 68/74 were first detected at 19, 15 and 17 years respectively.

In multivariable logistic regression model that included age, ever had sex and socio-economic status, only history of ever having sex (OR (95%CI): 3.11 (1.04–9.34), p-value = 0.04) was significantly associated with risk of anyHPV. There was no statistically significant associations between low or high risk HPV and any other variable in multivariable analyses (Table 2).