Prevalence and genotype distribution of human papillomavirus in women living with HIV/AIDS in an area of Northeast Brazil
Abstract
Women infected by human immunodeficiency virus (HIV) are more likely to manifest oncogenic viral infections including human papillomavirus (HPV). It was investigated the HPV prevalence, genotype distribution and HPV relationship with cervical lesions among women living with HIV in Sergipe state, Northeast Brazil. A prevalence survey was conducted including 270 HIV-infected women who attended the reference center for HIV in Sergipe from August 2014 to November 2017. Cervical samples were processed by the polymerase chain reaction for HPV-DNA detection. Among the 270 HIV-infected women, 190 (70.4%) were between 26 and 49 years old and 159 (55.6%) were coinfected with HPV. Among the coinfected women, 24 viral types were identified; 113 (72%) subjects had high-risk HPV types, and the most prevalent was HPV 16 (53/35.3%). Positive HPV status was statistically associated with having 0 to 8 years of schooling compared with ≥9 years of schooling; and have been diagnosed with HIV infection less than 5 years ago compared with more than 10 years. Cytological abnormalities were found in 13.4% (31/231) of women, most with high-grade squamous intraepithelial lesions (16/51.6%). However, of women who had no cytological lesions or malignancy (200/86.6%), almost half were HPV DNA-positive (99/49.5%). In conclusion, the prevalence of HPV among women living with HIV in Sergipe was high. There was a high frequency of high-risk HPV infection, and a wide diversity of genotypes were detected, with HPV 16 being the most frequent.
Highlights
-
Women infected by HIV are more likely to manifest oncogenic viral infection including human papillomavirus (HPV);
-
The prevalence of HPV and its genotypes vary with the location and settings;
-
A high prevalence of HPV infection and a wide diversity of genotypes was found in HIV-positive women in Sergipe state, Northeast Brazil;
-
Twenty-four HPV genotypes were detected, with a predominance of high-risk HPV types, including HPV 16, 31, 33, 35, 45, and 58;
-
Almost half of women without cytological abnormalities are coinfected with HPV, and most with high-risk genotypes.
1 INTRODUCTION
Women living with human immunodeficiency virus (HIV) are more likely to manifest oncogenic viral infections, such as human papillomavirus (HPV) infection, due suppression of the immune system. HPV is considered the main cause of invasive cervical cancer, an acquired immunodeficiency syndrome (AIDS)-defining disease.1-3
Because of the high prevalence of high-risk HPV among women living with HIV, as well as its association with cervical lesions and cancer, it is necessary to improve cervical cancer screening through HPV-DNA detection and genotyping testing.4, 5 This strategy will enable monitoring and evaluation of the efficacy of HPV vaccines and may help in widening cervical cancer screening in areas where cytology test (Papanicolaou smear) coverage is less effective, such as in North and Northeast Brazil.6, 7
Because of the importance of HPV screening with highly sensitive tests, and the high frequency of high-risk HPV infection in HIV-infected women, it was investigated the HPV prevalence, genotype distribution and HPV relationship with cervical lesions among women living with HIV in Sergipe state, Northeast Brazil.
2 MATERIALS AND METHODS
2.1 Study design and population
This is a prevalence survey with HIV-positive women who attended the reference center for sexually transmitted infections, HIV, and AIDS (CRIST/AIDS) from August 2014 to November 2017. CRIST/AIDS is located in the municipality of Aracaju, capital of Sergipe state, Northeast Brazil, and is the unique reference care center for people living with HIV. Between January 1987 and June 2018, a total of 5579 AIDS cases in adults were reported in Sergipe, of which 1821 (32.6%) were women.8 The CRIST/AIDS has a gynecologist who periodically consults HIV-positive women accompanied by the service. In Brazil, the Ministry of Health recommends to perform the cytopathological examination in all women living with HIV, independently their pregnancy status, after they have started their sexual activity.
The sample size was estimated based on the number of women registered at CRIST/AIDS in 2014 (800 women). Considering an expected prevalence of HPV-HIV coinfection of 47.5%,9 with a confidence interval of 95% and a maximum error of 5%, the calculated number of patients for this study was 260 women. However, at the end of the study, 270 HIV-positive women were included.
Eligible criteria for inclusion in this study were sexually active HIV-positive women who sought service for routine gynecological examinations, including pregnant women with a gestational age greater than 8 weeks. The exclusion criteria eligible for this study were hysterectomized HIV-positive women and women who had a menstrual period or had any other contraindications for oncotic cytology at the time of collection.
2.2 Data collection and procedures
After obtaining written informed consent to participate, participants were interviewed using a structured questionnaire that included demographic, obstetric and gynecological history, sexual behavior, and clinical information. In addition, the clinical records of the patients were reviewed to complete information on clinical data and laboratory tests, including HIV blood viral load (VL) quantification values (copies/mL) and CD4+ T cell count (cells/mm3). In Brazil, both VL and CD4+ T cell count are requested every 6 months for all patients using immunologically stable ART.10 It was considered the last results of both VL and CD4+ T cell count performed up to 6 months before cervical sample collection. For data analysis, it was dichotomized CD4+ into <350 cells/mm3 and >350 cells/mm3 and VL into suppressed VL (below 40 copies/mL), ≤1000 copies/mL, 1001-10 000 copies/mL and >10 000 copies/mL.
All enrolled women underwent a gynecological examination, which included oncotic cytology, and subsequently, they continued to be followed up at the health service.
In addition, cervical samples from the endocervix and ectocervix were collected by a gynecologist for HPV-DNA testing. Samples were immersed in phosphate-buffered saline solution and stored at −20°C. Afterwards, the samples were sent for molecular analysis to investigate the HPV genome and, in the positive cases, molecular genotyping was performed.
The cytology results were classified according to the guideline of the Brazilian Ministry of Health into the following strata: (a) low-grade squamous intraepithelial lesion (LSIL) that includes cervical intraepithelial neoplasia (CIN) I; (b) high-grade squamous intraepithelial lesions (HSIL), that includes CIN II and CIN III; and (c) atypical squamous cells (ASC), that includes ASC of undetermined significance (ASCUS) and ASC could not be excluded a HSIL (ASC-H).11
2.3 HPV detection and typing
Genomic DNA of the samples was extracted using the Wizard Genomic DNA Purification Kit (Promega) according to the manufacturer's instructions. For HPV detection, a region of the L1 gene was submitted for amplification by nested polymerase chain reaction (PCR) strategy using MY09/1112 and GP5+/6+ primers.13 The final PCR reaction contained 10 μL DNA sample, 25 μL of Master Mix (Promega), 3 mM MgCl2, and 20 pmol of each primer, with a total volume of 50 μL.
The first reaction was performed with MY09/11 primers using the following strategy: initial denaturation at 94°C for 5 minutes, 40 cycles of 95°C for 1 minute, annealing at 50°C for 1 minute, and extension at 72°C for 1 minute and a final extension at 72°C for 10 minutes. The amplified product was used as a template for the second reaction using GP6+/5+ primers using the following protocol: initial denaturation at 95°C for 5 minutes, 44 cycles of denaturation at 95°C for 45 seconds, annealing at 47.7°C for 45 seconds, and extension at 72°C for 1 minute, followed by a final extension at 72°C for 7 minutes.
The amplified product for each primer pair was visualized via 1.5% agarose gel electrophoresis and was purified using Wizard SV Gel and PCR Clean-Up System kit (Promega). For HPV genotyping, positive samples were subjected to DNA sequencing in an ABI 3500 automated sequencer (Applied Biosystems) using a BigDye Terminator Cycle Sequencing v 3.1 Ready Reaction kit (Applied Biosystems). Sequencing quality and assembly of contigs were performed with Pregap4 and Gap4 programs, incorporated in Staden package.14 Finally, typing of the isolates was determined by sequence identity with reference genotypes using Blast program (http://blast.ncbi.nlm.nih.gov/Blast.cgi) with standard parameters.
2.3.1 Statistical analysis
The data were analyzed using SPSS® software (Statistical Package for the Social Sciences) version 20.0 (IBM Corp., Armonk, NY). The prevalence of HPV infection in HIV-positive women was defined by sample positivity for the HPV-DNA PCR test.
Categorical variables were described using frequencies and percentages. Factors associated with the HPV status were established using the odds ratio (OR) with 95% confidence intervals (CIs). Logistic regression was used to identify factors independently associated with HPV status. P < .20 were used to select variables for inclusion in the logistic regression and covariates were identified that had significant bivariate tests. Backwards stepwise modeling was used, removing covariates if their statistical significance was lost (P > .05) and the variable was not a confounder through its effect on other parameters in the models.
2.4 Ethical aspects
This study was approved by the Research Ethics Committee of the Federal University of Sergipe (CAEE: 23374014.1.0000.5546). All investigation has been conducted according to the Declaration of Helsinki. Informed consent written was obtained from the participants. Parents or guardians provided written informed consent before enrolling their adolescents in the study.
3 RESULTS
A total of 270 women living with HIV/AIDS were enrolled. The median (interquartile range) age was 37 (29-45), with a predominance of women aged 26 to 49 years (190/70.4%). Most of the women were non-white (203/86.4%), had fewer than 8 years of schooling (178/65.9%), had income from one to two minimum wage (165/61.4%), and lived in urban areas (233/86.3%). Most women have been infected with HIV by sexual intercourse (219/81.1%). In total, 114 (47.3%) started their sexual activity ≤15 years-old, 185 (78.4%) reported having one to two sexual partners in the last 12 months, and 183 (67.8%) had a steady partner. Sex for money and drug use were reported by 26 (9.7%) women, while regular condom use was reported by 127 (51.8%) women (Table 1).
| Variables | N (%) |
|---|---|
| Age group, y (n = 270) | |
| 15-25 | 49 (18.1) |
| 26-49 | 190 (70.4) |
| ≥50 | 31 (11.5) |
| Years of education, y (n = 270) | |
| 0-8 | 178 (65.9) |
| ≥ 9 | 92 (34.1) |
| Ethnicity (n = 235) | |
| White | 32 (13.6) |
| Nonwhite | 203 (86.4) |
| Occupation (n = 270) | |
| Employed | 89 (33.0) |
| Unemployed | 136 (50.3) |
| Benefit salarya | 45 (16.7) |
| Family income (n = 269) | |
| No income | 73 (27.1) |
| 1-2 salaries | 165 (61.4) |
| >2 salaries | 31 (11.5) |
| Urban area (n = 270) | 233 (86.3) |
| HIV exposure category (n = 270) | |
| Sexual intercourse | 219 (81.1) |
| Vertical transmission | 5 (1.9) |
| Unknown | 46 (17.0) |
| Age at first sexual intercourse, y (n = 270) | |
| ≤15 | 114 (43.7) |
| >15 | 147 (56.3) |
| Number of sexual partners in the last year (n = 236) | |
| No partner | 38 (16.1) |
| 1 or 2 | 185 (78.4) |
| >2 | 13 (5.5) |
| Sexual partner (n = 270) | |
| Steady partner | 183 (67.8) |
| Casual partner | 18 (6.7) |
| Steady and casual partner | 2 (0.7) |
| No partner | 67 (24.8) |
| Drug use (n = 269) | 26 (9.7) |
| Sex for money (n = 269) | 26 (9.7) |
| Condon use (n = 245) | 127 (51.8) |
| Pregnant (n = 270) | 47 (17.5) |
| Number of pregnancies (n = 269) | |
| Nulligravid | 17 (6.3) |
| 1-3 | 181 (67.3) |
| ≥4 | 71 (26.4) |
| Number of deliveries (n = 269) | |
| Nulliparous | 40 (14.9) |
| 1-3 | 182 (67.7) |
| ≥4 | 47 (17.4) |
| Abortion (n = 269) | 98 (36.4) |
| HIV diagnosis (n = 268) | |
| Rapid HIV test or routine exam | 152 (56.7) |
| Previous prenatal or delivery | 64 (23.9) |
| Current prenatal | 25 (9.3) |
| Hospitalization | 27 (10.1) |
| Time of HIV diagnosis, y (n = 270) | |
| <5 | 133 (49.3) |
| 5-10 | 63 (23.3) |
| >10 | 74 (27.4) |
| CD4+ T-lymphocyte count >350 cells/μL (n = 231) | 174 (75.3) |
| HIV viral load (copies/mL) (n = 252) | |
| Suppressed viral load | 163 (64.7) |
| ≤1000 | 31 (12.3) |
| 1001-10 000 | 21 (8.3) |
| >10.000 | 37 (14.7) |
| Antiretroviral use (n = 270) | 255 (94.4) |
| Time of antiretroviral use, y (n = 255) | |
| <5 | 161 (63.1) |
| 5-10 | 48 (18.8) |
| >10 | 46 (18.1) |
- Abbreviations: HIV, human immunodeficiency virus; HPV, human papillomavirus.
- a The number of women in each category may not add up to 270 due to missing information.
Regarding obstetric and gynecologic history, 47 (17.5%) were pregnant at the time of cervical sample collection, 181 (67.3%) had one to three pregnancies, 182 (67.7%) had one to three deliveries, and 98 (36.4%) reported having had at least one abortion during their lifetime. Most HIV infected women were diagnosed less than 5 years before the data collection period (133/49.3%). There was a predominance of women with CD4+ T-cell count higher than 350 cells/mm3 (174/75.3%)and with suppressed HIV VL (163/64.7%). Two hundred and fifty-five (94.4%) reported the use of antiretrovirals, and 161 (63.1%) had used them less than 5 years ago (Table 1).
The HPV prevalence was 55.6% (150/270). Of these women with HPV, 150 (55.6%) 113 (72.0%) had high-risk HPV (HR-HPV), 29 (19.3%) had low-risk HPV (LR-HPV), and 13 (8.7%) had probably HR-HPV. A total of 24 HPV genotypes were identified among the HPV positive women and their distribution is shown in Table 2. Out of the 32 pregnant HIV-HPV coinfected women, 28 (87.5%) had HR-HPV. Of these, the most prevalent HPV genotypes were HPV 16 (17/53.1%), HPV 45 (3/9.4%), and HPV 31/35/58 (2/6.3% each).
| HPV genotypes | N (%) |
|---|---|
| High-risk | 113 (72.0) |
| 16 | 53 (35.3) |
| 18 | 4 (2.7) |
| 31 | 9 (6.0) |
| 33 | 9 (6.0) |
| 35 | 9 (6.0) |
| 45 | 8 (5.3) |
| 51 | 1 (0.7) |
| 52 | 3 (2.0) |
| 56 | 4 (2.7) |
| 58 | 7 (4.7) |
| 59 | 1 (0.7) |
| Probable high-risk | 13 (8.7) |
| 30 | 2 (1.3) |
| 53 | 6 (4.0) |
| 66 | 5 (3.3) |
| Low-risk | 29 (19.3) |
| 6 | 9 (6.0) |
| 11 | 4 (2.7) |
| 44 | 1 (0.7) |
| 54 | 1 (0.7) |
| 61 | 1 (0.7) |
| 62 | 4 (2.7) |
| 81 | 4 (2.7) |
| 86 | 1 (0.7) |
| 87 | 1 (0.7) |
| 114 | 3 (2.0) |
- Abbreviations: HIV, human immunodeficiency virus; HPV, human papillomavirus.
Women having 0 to 8 years of schooling (OR = 1.83; 95% CI, 1.10-3.06; P = .02), having been diagnosed less than 5 years ago (OR = 1.90; 95% CI, 1.06-3.40; P = .027) and being women pregnant (OR = 1.91; 95% CI, 0.98-3.81; P = 0.048) were factors associated with a more likely to have HPV infection. Of these, only having 0 to 8 years of schooling was independently associated with HPV-positivity (adjusted OR = 3.00; 95% CI, 1.40-6.42; .005) (Table 3).
| Variables | DNA-HPV PCR test | OR (95%CI) | P value | Adjusted OR (95% CI) | P value | |
|---|---|---|---|---|---|---|
| Positive(n = 150) | Negative(n = 120) | |||||
| N (%) | N (%) | |||||
| Age group, y (n = 270) | ||||||
| 15-25 | 29 (59.2) | 20 (40.8) | 1.99 (0.79-5,07) | .13 | 1.29 (0.48-3.45) | .61 |
| 26-49 | 108 (56.8) | 82 (43.2) | 1.81 (0.84-4.01) | .12 | 1.67 (0.75-3.65) | .19 |
| ≥50 | 13 (41.9) | 18 (58.1) | 1 | 1 | ||
| Years of education,y (n = 270) | ||||||
| 0-8 | 108 (60.7) | 70 (39.3) | 1.83 (1.10-3.06) | .02 | 3.00 (1.40-6.42) | .005 |
| ≥9 | 42 (45.7) | 50 (54.3) | 1 | 1 | ||
| Ethnical (n = 235)b | ||||||
| White | 14 (43.8) | 18 (56.2) | 1 | ⋯ | ⋯ | |
| Non-white | 110 (54.2) | 93 (45.8) | 1.52 (0.71-3.28) | .27 | ⋯ | ⋯ |
| Family income (n = 269)b | ||||||
| No income | 38 (52.1) | 35 (47.9) | 0.89 (0.38-2.07) | .79 | ⋯ | ⋯ |
| 1-2 salaries | 94 (57.0) | 71 (43.0) | 1.08 (0.49-2.37) | .82 | ⋯ | ⋯ |
| >2 salaries | 17 (54.8) | 14 (45.2) | 1 | |||
| Number of sexual partners in the last year (n = 236)b | ||||||
| No partner | 17 (44.7) | 21 (55.3) | 1 | ⋯ | ||
| 1 or 2 | 101 (54.6) | 84 (45.4) | 1.48 (0.73-3.03) | .26 | ⋯ | |
| >2 | 6 (46.2) | 7 (53.8) | 1.05 (0.51-2.04) | .93 | ⋯ | |
| HIV exposure category (n = 269)b | ||||||
| Sexual intercourse | 121 (55.3) | 98 (44.7) | 1.85 (0.26-15.7) | .49 | ⋯ | |
| Unknown | 28 (60.9) | 18 (39.1) | 2.29 (0.31-20.8) | .37 | ⋯ | |
| Vertical transmission | 2 (40.0) | 3 (60.0) | 1 | ⋯ | ||
| Drug use (n = 269)b | ||||||
| Yes | 13 (50.0) | 13 (50.0) | 0.77 (0.33-1.76) | .53 | ⋯ | |
| No | 137 (56.4) | 106 (43.6) | 1 | ⋯ | ||
| Sex for money (n = 269)b | ||||||
| Yes | 15 (57.7) | 11 (42.3) | 1.09 (0.47-2.54) | .83 | ⋯ | |
| No | 135 (55.6) | 108 (44.4) | 1 | ⋯ | ||
| Condon use (n = 245)b | ||||||
| Yes | 63 (49.6) | 64 (50.4) | 1 | ⋯ | ||
| No | 67 (56.8) | 51 (43.2) | 1.33 (0.80-2.21) | .26 | ⋯ | |
| Age at first sexual intercourse (years) (n = 270) | ||||||
| ≤15 | 66 (57.9) | 48 (42.1) | 1.25 (0.76-2.05) | .37 | ⋯ | |
| >15 | 77 (52.4) | 70 (47.6) | 1 | ⋯ | ||
| Pregnant (n = 270) | ||||||
| Yes | 32 (68.1) | 15 (31.9) | 1.91 (0.98-3.81) | .048 | 1.48 (0.70-3.13) | .29 |
| No | 117 (52.7) | 105 (47.3) | 1 | 1 | ||
| Number of pregnancies (n = 269)b | ||||||
| Nulligravid | 8 (47.1) | 9 (52.9) | 1 | ⋯ | ||
| 1-3 | 104 (57.5) | 77 (42.5) | 1.51 (0.54-4.26) | .40 | ⋯ | |
| ≥4 | 37 (52.1) | 34 (47.9) | 1.22 (0.41-3.65) | .70 | ⋯ | |
| Number of deliveries (n = 269)b | ||||||
| Nulliparous | 25 (62.5) | 15 (37.5) | 1 | ⋯ | ||
| 1-3 | 99 (54.4) | 83 (45.6) | 0.71 (0.34-1.44) | .35 | ⋯ | |
| ≥4 | 25 (53.2) | 22 (46.8) | 0.68 (0.85-1.62) | .38 | ⋯ | |
| Abortion (n = 269)b | ||||||
| Yes | 52 (53.1) | 46 (46.9) | 0.86 (0.52-1.42) | .56 | ⋯ | |
| No | 97 (56.7) | 74 (43.3) | 1 | ⋯ | ||
| Time of HIV diagnosis (years) (n = 270) | ||||||
| <5 | 84 (63.2) | 49 (36.8) | 1.90 (1.06-3.40) | .027 | 1.83 (0.98-3.40) | .06 |
| 5-10 | 31 (49.2) | 32 (50.8) | 1.07 (0.54-2.12) | .84 | 1.09 (0.54-2.20) | .81 |
| >10 | 35 (47.3) | 39 (52.7) | 1 | 1 | ||
| CD4 cell count (cells/μl) (n = 231)b | ||||||
| ≤350 | 38 (66.7) | 19 (33.3) | 1.34 (0.72-2.56) | .35 | ⋯ | |
| >350 | 104 (59.8) | 70 (40.2) | 1 | ⋯ | ||
| Suppressed HIV RNA viral loada (n = 252)b | ||||||
| Yes | 93 (57.1) | 70 (42.9) | 0.94 (0.56-1.6) | .83 | ⋯ | |
| No | 52 (58.4) | 37 (41.6) | 1 | |||
| Antiretroviral use (n = 270) | ||||||
| Yes | 143 (56.1) | 112 (43.9) | 1 | ⋯ | ||
| No | 7 (46.7) | 8 (53.3) | 0.68 (0.23-2.00) | .47 | ⋯ | |
| Time of antiretroviral use, y (n = 255)b | ||||||
| <5 | 95 (59.0) | 66 (41.0) | 1.4 (0.83-2.30) | .2 | ⋯ | |
| 5-10 | 24 (50.0) | 24 (50.0) | 0.74 (0.39-1.39) | .35 | ⋯ | |
| >10 | 24 (52.2) | 22 (47.8) | 0.83 (0.43-1.57) | .56 | ⋯ | |
- Abbreviations: aOR, adjusted odds ratio; CI, confidence interval; HIV, human immunodeficiency virus; HPV, human papillomavirus; OR, odds ratio; PCR, polymerase chain reactions.
- a It was considered as suppression of HIV RNA viral load, results below 40 copies/mL.
- b The number of women in each category may not add up to 270 due to missing information.
Regarding the 231 cytological results found, 200 (86.6%) were negative for intraepithelial lesions or malignancy (NILM) in cytology, and 31 (13.4%) had cervical abnormalities. Of the 200 NILM, 99 (49.5%) were positive for HPV and 77 (38.5%) were HR-HPV. Beyond that, the proportion of cytological abnormalities was higher in HIV-HPV coinfected women (22/71.0%) than in women with a negative HPV result (9/29.0%) (Table 4).
| Variables | DNA-HPV PCR test | |||
|---|---|---|---|---|
| Negative | Positive | |||
| LR-HPV | pHR-HPV | HR-HPV | ||
| N (%) | N (%) | N (%) | N (%) | |
| Cytological outcomesa (n = 231) | ||||
| NILM (n = 200) | 101 (50.5) | 15 (7.5) | 7 (3.5) | 77 (38.5) |
| Atypical squamous cellsb (n = 7) | 4 (57.1) | 1 (14.3) | ⋯ | 2 (28.6) |
| LSIL (CIN I) (n = 8) | 2 (25.0) | 1 (12.5) | 1 (12.5) | 4 (50.0) |
| HSIL (CIN II/III and CIN III) (n = 16) | 3 (18.8) | 5 (31.2) | 2 (12.5) | 6 (37.5) |
- Abbreviations: CIN, cervical intraepithelial neoplasia; DNA, desoxyribonucleic acid; HPV, human papillomavirus; HR-HPV, high-risk human papillomavirus; HSIL, high-grade squamous intraepithelial cervical lesions; LR-HPV, low-risk human papillomavirus; LSIL, low-grade squamous intraepithelial cervical lesions; NILM, negative for intraepithelial lesions or malignancy; PCR, polymerase chain reactions; pHR-HPV, probably high-risk HPV.
- a The number of women in cytological outcomes category may not add up to 270 due to missing information.
- b Atypical squamous cervical cells of undetermined significance (ASC-US) and atypical squamous cervical cells could not be excluded a high-grade squamous intraepithelial cervical lesion (ASC-H).
Among the 31 cytological abnormalities, eight (25.8%) were described as LSIL, 16 (51.6%) were HSIL, and seven (22.6%) were ASC. Regarding the 22 HIV-HPV coinfected women with cytological abnormalities, seven (31.8%) were identified with low-risk HPV, three (13.6%) were probably high-risk, and 12 (54.6%) were high-risk HPV. Of the women with LSIL, HR-HPV appeared in the highest proportion (4/50.0%), and the proportion of women with HSIL was high among those with LR-HPV (5/31.2%) and HR-HPV (6/37.5%). However, of seven ASC positive samples, four (57.1%) were negative for HPV (Table 4). In addition, among women with cytological abnormalities, five (22.7%) were genotyped as HPV 16, of which two had HSIL, two had LSIL and one had ASC. Four of them (18.2%) had HPV 35, of which three had HSIL and two had LSIL.
4 DISCUSSION
This study describes the prevalence of HPV, the circulating HPV genotypes, and their distribution across a spectrum of cervical conditions in women living with HIV in Sergipe state, an area of Northeast Brazil. It was found a high prevalence of HPV among HIV-infected women demonstrating a wide variety of HPV genotypes, although HPV 16 was predominant.
A systematic review and meta-analysis of cross-sectional studies,5 using similar HPV-DNA detection methods, showed that the global prevalence of HPV in women living with HIV was 57%, but with a wide variation across continents. The highest proportion of HPV-positivity in HIV-infected women was found in Africa (74%) and the lowest proportion in Asia (34%). The meta-analysis of studies conducted in Latin America revealed a prevalence of 69%, which is higher than the prevalence found in this study.5
The prevalence of HPV among HIV-positive women in Sergipe was similar to that estimated in other Brazilian states. Studies carried out in South15, 16 and Southeast17-19 regions have shown that the prevalence of HPV in HIV-infected women ranged from 46.6% to 89% and in the Northeast region9, 20 where prevalence ranged from 47.5% to 96%.
Only the variable having 0 to 8 years of schooling was independently associated with a higher likelihood of having HPV infection, and this finding is in agreement with other Brazilian study.9 Low education has been an important barrier to preventive measures for HPV infection, just as low education has been associated with difficulties in accessing screening tests for cervical cancer.21, 22
Although pregnant women are more vulnerable to HPV infection,19, 23 this condition was not associated with an increased risk of HPV infection in this study. A type II error can explain the absence of association between pregnancy and HPV infection in this study.
In this study observed 24 HPV genotypes, with a predominance of high-risk HPV types (11 genotypes) including HPV 16, 31, 33, 35, 45, and 58. This high proportion of women living with HIV and coinfected with high-risk HPV is worrisome, as immunodeficiency accelerates the progression of HPV disease. This in turn increases the chances of developing cervical lesions or cancer and makes the treatment more difficult and decreases women's survival.24
Because of Brazil's wide territorial extension and population plurality, the prevalence and distribution of HPV genotypes may vary according to the region. In addition, the techniques used to detect HPV can also influence this variation. Although most Brazilian studies have detected HPV genotypes similar to ours, in some studies a higher prevalence of non-HPV 16 has been identified, such as: 52, 53, 56, 68 e 72.9, 16, 25-27
However, in this study was documented a predominance of HPV 16 among women living with HIV/AIDS, which also is consistent with other studies.18, 20, 28 HPV 16 type has also been most often found in the general female population.29 A previous study conducted in Sergipe with HIV-negative women with cervical lesions showed that HPV 16 was the most common genotype.30 This means that HPV 16 type is the most frequently associated with cervical lesions, regardless of women's HIV status in Sergipe.
Despite the evidence that HPV 16 is slightly less affected by HIV-related immunodeficiency than other HPV types31 and that high-grade lesions in women living with HIV/AIDS are relatively more frequently caused by non HPV 16 types than in HIV-uninfected women,32 in this study it was revealed that HPV 16 was more frequent among cervical lesions.
This is in line with the findings of the systematic review and meta-analysis conducted by5 which demonstrated that HPV 16 remains the clear priority for prevention of invasive cervical cancers in HIV-infected women worldwide. HPV 16 is considered the most common and most carcinogenic type and its prevalence tends to increase with the severity of intraepithelial cervical lesions.5
Despite this evidence, a low number of cervical intraepithelial lesions was identified, including high-grade lesions. A systematic review and meta-analysis showed that early use and adherence to antiretroviral therapies appear to reduce the incidence and progression of cervical dysplasias.33 Thus, it is believed that the control of HIV infection, demonstrated by most women using antiretrovirals and presenting good levels of CD4+ T lymphocytes and suppressed VL, seems to provide some protection for the development of cervical lesions in women living with HIV/AIDS in Sergipe. However, despite the low prevalence of cervical intraepithelial injuries in this population, it is necessary that these be monitored periodically.
Therefore, the fact of half of women without cytological abnormalities are coinfected with HPV, and most with high-risk genotypes, demonstrates the importance of screening in this population for HPV using HPV-DNA detection techniques. Although less specific than cytological tests, the HPV-DNA molecular tests provide high diagnostic sensitivity and better accuracy about the risk of developing CIN III within 5 to 15 years, especially in women aged 30 years and older.34, 35 In addition, a negative HPV result by molecular test ensures greater safety against CIN III over a longer tracking time than cytology.36, 37
HPV-DNA screening by molecular testing has been more advantageous when it is incorporated into the cytology routine (cotest), as their association facilitates the identification of cervical cancer and precursor lesions among women regardless of HIV status.37, 38 In addition, because of the higher accuracy in detecting HPV infection, HPV-DNA screening allows larger spacing between tests when negative results are found.35, 37, 38
Both the Pap smear and the cotest for the screening of cervical cancer in women living with HIV is important.4, 39 The centers for disease control and prevention (CDC) recommends the Pap test to screen cervical cancer in HIV-positive women under 30 years-old and this test should be repeated in 12 months. If the results of three tests in a row are negative, Pap smears can be performed in 3 years.40
However, the CDC also recommends carrying out the cotest on HIV-infected women aged 30 or older. When the result from the cotest is negative, there is a possibility of larger spacing between cervical cancer screening tests, which can be performed every 3 years.40
In our study, most HIV-positive women had normal cytology and a positive HPV test, which may reflect a high risk of developing cervical lesions. Corroborating this, a prospective cohort study found that HIV-infected women with normal cytology testing and positive HPV DNA testing, as occurred in most women in this study, were at high-risk for the development of precancerous lesions (CIN II and CIN III), especially those with HPV 16.39
Despite the best efforts, this study has some limitations. We used a molecular test composed of MY09/11 and GP5+/6+ primers, which are used for general detection of HPV. Moreover, the molecular test used in this study does not allow the identification of multiple HPV types in the same sample, which has reduced the ability to evaluate multiple infections and their relationship to high-grade lesions.
In conclusion, there is a high prevalence of HPV among women living with HIV/AIDS in Sergipe. Also highlighted was the high frequency of high-risk HPV and the relationship of these strains to severe cervical findings, especially with HPV 16, which demonstrates the need for implementation of screening tests in this population.
ACKNOWLEDGMENTS
This study was supported by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), Brazil (Finance Code 001), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPQ), Brazil (Grant Number 449851/2014-1; 446425/2014-1), and Fundação de Apoio à Pesquisa e à Inovação Tecnológica do Estado de Sergipe (FAPITEC/SE), Brazil (Grant Number PPSUS SERGIPE 8444/2013-02; 019.203.00941/2018-5).
CONFLICT OF INTERESTS
The authors declare that there are no conflict of interests.
AUTHOR CONTRIBUTIONS
BEBS, VSS, MVAB, and LMDL conceived and planned the study. BEBS and IERS conducted study recruitment and collected the specimens. BEBS, IERS, EDA, MVA, LASS, and MVAB contributed to sample preparation and conducted the assays. BEBS, VSS, and LMDL planned and carried out the analyses. BEBS and IERS wrote the first draft of the manuscript with support from VSS, MVAB, and LMDL. All authors provided critical feedback, helped shape the research, analysis and manuscript, and approved the final version to be published.
