Articles••

Incidence of persistent and high-risk human papillomavirus infection and associated factors among HIV-positive women in China, 2015–2016

•,,,.

...

Abstract

Background Persistent human papillomavirus (HPV) infection is the necessary factor for cervical cancer. Human immunodeficiency virus (HIV)-positive women are a high-risk population for cervical cancer, but little is known about persistent HPV infection among HIV-positive women in China. We aimed to investigate the persistence and incidence of infection with high-risk HPV (hrHPV) and associated factors among HIV-positive women in China.

Method Using a prospective observational study design, we collected data from 670 HIV-positive women in five counties of three provinces through interviews, medical record reviews, health examination, and laboratory tests at a baseline survey in 2015 and a follow-up survey in 2016. We tested HIV-positive women for hrHPV infection using the Cobas 4800 HPV test, and analyzed factors associated with persistence and new incidence of hrHPV infection using logistic regression.

Results After an average of 15 months at follow-up, the persistence and new incidence rates of hrHPV infection were 38.0% (46/121, 95% CI: 29.2%–46.8%) and 8.3% (35/420,95% CI: 5.7%–11.1%), respectively. HIV-positive women in Yunnan and Guangxi were more likely to have persistent hrHPV infections than those in Xinjiang (aOR = 3.18, 95% CI: 1.00–10.14, p = 0.05; aOR = 6.38, 95% CI: 2.29–17.77, p < 0.001). Migrant women had a higher rate of new hrHPV incidence than local women (aOR = 4.48, 95% CI: 1.65–12.15, p = 0.003). HIV-positive women from Xinjiang were 6.5 times more likely to be newly infected with hrHPV than those from Yunnan (95% CI: 2.40–17.60, p < 0.001). HIV-positive women with HBV infection had a higher risk of being infected with hrHPV than those without HBV infection (aOR = 3.48, 95% CI: 1.13–10.71, p = 0.029).

Conclusions The persistence and new incidence rates of hrHPV infection among HIV-positive women were high and varied significantly between regions. HBV infection and being among the migrant population were associated with new hrHPV infection.

Highlights

HIV-positive women are at high risk of infection or developing a persistent hrHPV infection.
The rates of persistent hrHPV infection and new infections varied significantly by region.
HBV infection and being among the migrant population were associated with new hrHPV infection.

1 Introduction

Human papillomavirus (HPV) is a common sexually transmitted infection. The virus infects epithelial cells and has the ability to proliferate and differentiate through small wounds in the skin and mucosa. Currently, more than 100 HPV genotypes have been found, of which at least 14 are related to the occurrence of cervical cancer and precancerous lesions; these are called high-risk human papillomavirus (hrHPV).¹ While most cervical HPV infections last for a short period, some individuals develop persistent HPV infections, defined as positive results of two consecutive HPV tests with at least a 6-month interval. Persistent hrHPV infection is a necessary factor for cervical cancer²^,³ and is closely associated with cervical cancer and precancerous lesions in middle-aged women.⁴^,⁵

While the majority of women with HPV do not develop persistent infections, women with human immunodeficiency virus (HIV) infections are particularly susceptible to HPV infection and have a reduced ability to clear the virus due to immune suppression.⁶ Previous studies have shown that the risk of persistent HPV infection is higher in HIV-positive women than in HIV-negative women,⁷ and HIV-positive women are 4–5 times more likely than HIV-negative women to develop cervical cancer.⁸ It has been reported that persistent hrHPV infection in HIV-infected women is associated with antiretroviral treatment, having multiple sexual partners, and long follow-up time.^9–11

As of the end of 2015, there were 577,423 reported HIV/acquired immunodeficiency syndrome (AIDS) patients in China, of whom ∼165,000 were female (28.6%). Among the 115,465 newly-reported HIV/AIDS patients in 2015, ∼25,100 cases were female (21.7%).¹² HIV-positive women are mainly concentrated in several areas, including Yunnan, Guangxi, Xinjiang, and Henan. Due to the negative physical, psychological, and social effects of HIV infection, HIV-positive women face more severe reproductive health problems than HIV-negative women. As a result, HIV-positive women are extremely vulnerable.

Prior surveys with small sample sizes in Hubei and Yunnan reported that the persistent hrHPV infection rate was higher in HIV-positive women than in HIV-negative women.^13–16 However, there are very few studies exploring persistent hrHPV infection and associated factors among HIV-positive women in China, and no such research on new hrHPV infection rates was found. Considering the large number of HIV-positive women in China who are vulnerable to HPV infection and cervical cancer, it is necessary to address those issues. Therefore, this study aimed to determine the rate of new and persistent hrHPV infections, and factors associated with them, among HIV-infected women in China. This will ultimately provide scientific evidence for strategies to prevent and treat HPV infection and cervical cancer among HIV-positive women in China.

2 Method

2.1 Study setting and population

We conducted a prospective observational study in China between 2015 and 2016. The baseline survey was carried out between May and August 2015. Sample size calculations and the recruitment of HIV-positive women in the baseline survey were reported previously.¹⁷ The follow-up survey followed the same procedure and was conducted 15 months later.

We utilized random cluster sampling and selected three provinces/autonomous regions: Yunnan province, Guangxi Zhuang Autonomous Region, and Xinjiang Uygur Autonomous Region. These regions had high incidence of new HIV infections in women aged 15–49 before the research conducted, representing areas where the AIDS epidemic emerged earlier and hit harder than in other areas of China.¹⁸

We defined our study population as HIV-positive women, meaning they had confirmed HIV status by laboratory test and were aged 18–49 years. We excluded HIV-positive women who: 1) were ill at the time of the study contacted; 2) could not be contacted after meeting the requirements of the study; 3) had no sexual intercourse experience, or 4) had a history of uterectomy or high-grade squamous intraepithelial lesion (HSIL) or above.

2.2 Data collection and procedures

The field team consisted of trained healthcare staff in local maternal and child health (MCH) hospitals. After obtaining informed consent, the field team conducted face-to-face interviews with participants to collect data related to sociodemographic characteristics and sexual practices using a structured questionnaire. The field team reviewed the HIV-positive women's medical records to extract data on antiretroviral treatment (ART) and laboratory results.

Trained clinicians conducted gynecological physical examinations as follows: 1) observed the vulva, vagina, cervix, and vaginal discharge for abnormalities; 2) collected vaginal discharge and exfoliative cervical cell specimens; 3) conducted a pelvic examination, and provided B-ultrasound to those with abnormal uterus or uterine adnexa. Blood samples of participants were also collected.

The study team sent cervical cell specimens for cytological examination and HPV DNA analysis with the Cobas 4800 HPV test (Roche Molecular Diagnostics [Shanghai] Co. Ltd., Shanghai, China) in Beijing Di'an Laboratory. The HPV test produced individual results for two hrHPV strains, HPV16 and HPV18, and a pooled result for the other 12 hrHPV types (31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, and 68b).

Data related to CD4 lymphocyte count tests and HIV viral load tests in the six months prior to fieldwork were obtained from clinical records. For participants who had not CD4 lymphocyte counts or HIV viral load results in the prior six months, serum samples were sent to local prefecture CDC laboratories for the appropriate test(s). Other tests were conducted at the county level MCH hospital laboratories, including microscopic examination of vaginal secretions with amine test and pH measurement as appropriate, syphilis treponema pallidum particle agglutination (TPPA) and syphilis toluidine red unheated serum test (TRUST) for syphilis diagnosis, enzyme-linked immunosorbent assay (ELISA) for presence of hepatitis B surface antigen (HBsAg).

HIV-positive women with cytological abnormity or hrHPV infection received a biopsy with colposcopy for cervical cancer screening, following the technical protocol for rural women cervical cancer screening.¹⁹ We referred the HIV-positive women with HSIL and above to appropriate treatment.

2.3 Study variables

We defined hrHPV infection as a positive result of the Cobas 4800 HPV test for HPV16, HPV18, or any of the other 12 hrHPV types. “Persistent hrHPV infection” refers to positive results of HPV DNA tests at both the baseline and follow-up surveys. “Persistent hrHPV infection rate” refers the proportion of women with persistent hrHPV infection (established at the follow-up survey) among those with hrHPV infection at baseline. “ New hrHPV infection” refers to a negative HPV test result at the baseline survey but a positive result at a follow-up survey. “New hrHPV infection rate” refers to the proportion of women with hrHPV infection at the follow-up among those who were hrHPV-negative at the baseline.

Trichomonas vaginitis, vulvovaginal candidiasis, bacterial vaginosis, and other infections were collectively referred to as vaginitis. Cervicitis was diagnosed depending on vaginal discharge examination and particular signs found in gynecological physical examinations. Syphilis infection was diagnosed when the results for both TPPA and TRUST were positive. Degrees of cervical lesions were defined as Normal, low-grade squamous intraepithelial lesion (LSIL), or HSIL.

2.4 Statistical analysis

Before data entry, we reviewed the questionnaire data and undertook checks for quality assurance. We double entered and validated the data using Epidata software (version 3.1, The EpiData Association, Odense Denmark), and used SPSS software (version 23.0, IBM Corp, Armonk, NY, USA) to analyze the data.

Categorical variables (sociodemographic characteristics, sexual practices, and laboratory characteristics) were presented as frequencies and proportions. Continuous variables with normal distribution were presented as mean ± standard deviation (SD). Persistent and new hrHPV infections are described by frequencies and proportions, as appropriate. Pearson's chi-squared test and Fisher's exact test as appropriate was used to determine the association of categorical and continuous variables with hrHPV infection categories. Factors with a p-value of <0.1 were entered into a binary logistic regression to build the final model. A p-value of <0.05 was defined as statistically significant, and Odds Ratios (OR) and 95% confidence intervals (95% CI) were calculated as indicators of the magnitude of association and statistical significance of hrHPV infection.

The Institutional Review Board of the National Centre for Women and Children's Health, China CDC reviewed and approved the study (No. FY2015–014). All participants provided informed consent to participate in the study.

3 Results

The recruitment, hrHPV test, and follow-up processes are shown in Fig. 1. During the baseline survey, 695 HIV-positive women were recruited, and 25 women were diagnosed with HSIL, excluding them from the study. Therefore, 670 women were enrolled in the study following the baseline survey. A total of 541 women (80.7%) participated in the follow-up survey after 14.5 ± 0.7 months from the baseline.

Fig. 1

Request permissions

Diagram showing processes for recruitment, hrHPV test, and follow-up survey processes among HIV-positive women from high HIV-burden areas in China, 2015–2016.

Among the 670 women enrolled in the study, 33.9% (227) were Uighur, 30.7% (206) were Jingpo and Dai, and 45.1% (302) were farmers (Table 1). Among the 659 women who reported their ART status, 95.0% (626) were on ART at the baseline. The distribution of ethnic group, marital status, and ART status was significantly different between women who completed the follow-up survey (“follow-up women”) and women who did not (“lost-follow-up women”; p < 0.05). The proportions of women who were Uighur, not on ART, and widowed/single/divorced among lost-follow-up women were higher than those among follow-up women. We found no statistically significant differences in the distribution of other sociodemographic characteristics and hrHPV infection at baseline between these two groups of women.

Table 1

•

Characteristics of followed and lost follow-up HIV-positive women enrolled in the study from HIV high-burden areas in China, 2015–2016.

3.1 hrHPV infection

Among the 670 women in the baseline survey, 157 (23.4%) were infected with hrHPV. Among the 541 women at the follow-up survey, the hrHPV infection rate was 15.0%, lower than that at the baseline (χ² = 12.046, p < 0.001). Among 121 follow-up women who were positive for hrHPV at the baseline survey, 46 women tested positive for hrHPV at the follow-up survey, for a hrHPV persistent infection rate of 38.0% (95% CI: 29.2%–46.8%). Among 420 women who were hrHPV-negative at the baseline survey, 35 were hrHPV-positive at the follow-up survey, making the new hrHPV infection rate 8.3% (95% CI: 5.7%–11.1%).

3.2 Factors associated with persistent hrHPV infection

In the univariate analyses, persistent hrHPV infection rates varied significantly by province, ethnic group, rural/urban status, and ART status during the follow-up period (Table 2). We entered the factors at p < 0.1 in univariate analyses into the logistic regression model and found that provinces were statistically associated with persistent hrHPV infection; compared with women from Xinjiang, women from Yunnan and Guangxi were more likely to have persistent hrHPV infections (aOR = 3.18, 95% CI: 1.00–10.14, p = 0.05; aOR = 6.38, 95% CI: 2.29–17.77, p < 0.001).

Table 2

•

Persistent hrHPV infection among HIV-positive women from HIV high-burden areas in China, 2015–2016.

3.3 New hrHPV infections and associated factors

Like persistent hrHPV infection rates, new hrHPV infection rates were significantly different in women from different provinces and ethnic groups, with additional significant differences based on occupation (Table 3). We entered the factors at p < 0.1 in univariate analyses into the logistic regression model and found that compared with women from Yunnan, women from Xinjiang were more likely to become newly infected with hrHPV (aOR = 6.50, 95% CI: 2.40–17.60, p < 0.001). Women who were migrant were more likely to be newly infected with hrHPV during the follow-up period (aOR = 4.48, 95% CI: 1.65–12.15, p = 0.003). HBsAg-positive women had a higher risk of hrHPV infection compared with HBsAg-negative women (aOR = 3.48, 95% CI: 1.13–10.71, p = 0.029).

Table 3

•

New hrHPV infection among HIV-positive women from HIV high-burden areas in China, 2015–2016.

4 Discussion

Persistent and new hrHPV infection rates were studied in HIV-positive women from high HIV-burden areas in China. The persistent hrHPV infection rate was 38.0% (95% CI: 29.2%–46.8%) and the new hrHPV infection rate was 8.3% (95% CI: 5.7%–11.1%) at a 15-month follow-up. In a previous study conducted in Wuhan, China, the persistent HPV infection rates among HIV-positive and HIV-negative women were 47.5% (48/101) and 21.1% (8/38), respectively, after a 12-month follow-up.¹⁶ We did not find statistically significant differences between the persistence rates in our study and the previous study (χ² = 2.039, p = 0.153). The new hrHPV infection rate reported here is similar to results reported by Ahdieh et al. that hrHPV among HIV-infected women in the United States occurred at a rate of 8.05/100 person-years.¹¹ Konopnicki et al. also reported the hrHPV incidence was 13/100 person-years after 66 months among 651 African HIV-infected women in Belgium, which was higher than the rate (5/100 person-years) in HIV-negative women.²⁰ Consistent with these prior studies, our results indicate that HIV-positive women had an increased susceptibility to infection with hrHPV, and that HIV-positive women with hrHPV infection were at higher risk of developing persistent infection compared to HIV-negative women.

We found that the rates of persistent and new hrHPV infections among HIV-infected women in different research sites were different. This variation was likely associated with the variation in sociodemographic characteristics of HIV-infected women in different sites. It has been previously reported that the hrHPV infection rate was higher among Uighur women than among other ethnic groups.²¹ Of the 36 baseline hrHPV-infected women lost to follow-up, 27 were from Xinjiang. The lost follow-up indicated poor adherence and less healthcare-seeking behavior of the participants. The low ART treatment rate of lost-follow-up women also increased the risk of persistent infections. A lower follow-up rate in Xinjiang than in the other two provinces may explain the low persistent hrHPV infection rate among HIV-infected women in Xinjiang reported here. We also found that the new hrHPV infection rate in HIV-infected women from Xinjiang was higher than in participants from other regions. Our findings indicate a need to address hrHPV infection in HIV-positive Uighur women in Xinjiang.

It was previously reported that a CD4 lymphocyte count ≥500/mm³ for more than 18 months and an HIV viral load <50 copies/mL for more than 40 months were protective factors against persistent hrHPV infection.⁹ However, our study did not find correlations between CD4 lymphocyte count level and either persistent or new hrHPV infection, which was consistent with the conclusions of previous studies.²⁰ Previous research reported that the rate of persistent HPV infection among HIV-positive women dropped from 67.7% at six months to 45.2% at 36 months, indicating that HPV infection persistence decreased over time.¹¹ Our study was followed up for only 15 months and therefore a prolonged observation to explore the relationship between persistent hrHPV infection and CD4 lymphocyte count levels may be necessary.

We found that women with hepatitis B infection had an increased risk of hrHPV infection compared with hepatitis B virus (HBV)-negative women. Siu et al. reported that women with high-grade cervical lesions were more likely to have HBV infection.²² Liu et al. analyzed the data of 4,290 women in northern China who were co-infected with HPV, hepatitis B, HIV, or other infections, and found that HPV infection was closely linked to HBV infection.²³ Immunogenic responses can eliminate most infections with HPV or HBV, but HIV/AIDS weakens the body's ability to clear infections, and thus the probability of long-term infection increases. However, the interaction between HBV and HPV infections is not yet understood, and further research is still needed.

We found that the new hrHPV infection rate in HIV-infected women from the migrant population was higher than that in HIV-infected women from the local population. Increased attention should be paid to migrant HIV-positive women in the future. HPV infection relates to having multiple sexual partners, which might increase the risk of HPV exposure.²⁴ However, no association between HPV infection and sexual behavior was observed in this study. This is likely because sexual topics are sensitive; as sexual behavior was directly related to their HIV infection, some HIV-positive women did not respond to questions about high-risk sexual behaviors such as multiple sexual partners. Nevertheless, our findings indicate that health education targeted to HIV-positive women, especially the migrant population, is necessary to improve their awareness of the role of safe sexual behavior in preventing cervical cancer.

The study is subject to several limitations. First, most lost-follow-up women had not received ART, had low CD4⁺ T lymphocyte counts and had high HIV viral load levels. Hence the rates of persistent and new HPV infection reported here may be lower than the reality. Second, limited by the hrHPV test method, this study did not provide type-specific persistent hrHPV infection rates. Third, limited by the short follow-up period, our results could not provide evidence-based suggestions related to HPV screening frequency for HIV-positive women. Hence, long-term follow-up of HIV-infected women should be conducted to obtain more evidence on HIV infection-related factors and screening strategies.

5 Conclusion

This research suggests that HIV-positive women are at higher risk of infection or developing a persistent hrHPV infection than. The rates of persistent hrHPV infection and new infections varied significantly by region. The migrant population and HBV-infected women had a higher risk of new hrHPV infection than others.

Declaration of interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

☆: This study was supported by grants from the UNICEF China (IR-5.2 PMTCT and Pediatric Services).

Alliance for Cervical Cancer Prevention (ACCP). Planning and Implementing Cervical Cancer Prevention and Control Programs: A Manual for Managers, Seattle, ACCP 2004;
Available: here
Google Scholar
Khan M.J., Castle P.E., Lorincz A.T., et al. The elevated 10-Year risk of cervical precancer and cancer in women with human papillomavirus (HPV) type 16 or 18 and the possible utility of type-specific HPV testing in clinical practice. J Natl Cancer Inst 2005; 97:1072–1079.
Google Scholar
Bosch F.X., Lorincz A., Muñoz N., et al. The causal relation between human papillomavirus and cervical cancer. J Clin Pathol 2002; 55:244–265.
Google Scholar
Peto J., Gilham C., Deacon J., et al. Cervical HPV infection and neoplasia in a large population-based prospective study: the Manchester cohort. Br J Canc 2004; 91:942–953.
Google Scholar
Maucort-Boulch D., Franceschi S., Plummer M., et al. International correlation between human papillomavirus prevalence and cervical cancer incidence. Cancer Epidemiol Biomark Prev 2008; 17:717–720.
Google Scholar
Conley L.J., Ellerbrock T.V., Bush T.J., et al. HIV-1 infection and risk of vulvovaginal and perianal condylomata acuminata and intraepithelial neoplasia: a prospective cohort study. Lancet 2002; 359:108–113.
Google Scholar
Tornesello M.L., Duraturo M.L., Giorgi-Rossi P., et al. Human papillomavirus (HPV) genotypes and HPV16 variants in human immunodeficiency virus-positive Italian women. J Gen Virol 2008; 89:1380–1389.
Google Scholar
Frisch M., Biggar R.J., Goedert J.J., et al. Human papillomavirus-associated cancers in patients with human immunodeficiency virus infection and acquired immunodeficiency syndrome. J Natl Cancer Inst 2000; 92:1500–1510.
Google Scholar
Konopnicki D., Manigart Y., Gilles C., et al. Sustained viral suppression and higher CD4+ T-cell count reduces the risk of persistent cervical high-risk human papillomavirus infection in HIV-positive women. J Infect Dis 2013; 207:1723–1729.
Google Scholar
Zhang Y.X., Xiong Y., Gui X.E., et al. Epidemiologic risk profile of human papillomavirus infections in human immunodeficiency virus-positive Chinese women. Jpn J Infect Dis 2011; 64:411–416.
Google Scholar
Ahdieh L., Klein R.S., Burk R., et al. Prevalence, incidence, and type-specific persistence of human papillomavirus in human immunodeficiency virus (HIV)–Positive and HIV-negative women. J Infect Dis 2001; 184:682–690.
Google Scholar
National health and family plan commission. 2015 China AIDS response progress report. 2015;
Available: here
Google Scholar
Zhang H.Y., Tiggelaar S.M., Sahasrabuddhe V.V., et al. HPV prevalence and cervical intraepithelial neoplasia among HIV-infected women in yunnan province, China: a pilot study. Asian Pac J Cancer Prev APJCP 2012; 13:91–96.
Google Scholar
Zhang H.Y., Fei M.D., Jiang Y., et al. The diversity of human papillomavirus infection among human immunodeficiency virus-infected women in Yunnan, China. Virol J 2014; 11:1–10.
Google Scholar
Zhang Y.X., Gui X.E., Zhong Y.H., et al. Cancer in cohort of HIV-infected population: prevalence and clinical characteristics. J Canc Res Clin Oncol 2011; 137:609–614.
Google Scholar
Zhang Y.X., Xiong Y., Gui X.N., et al. Analysis of cervical HPV infection in HIV-positive women. Chin J Obstet Gynecol 2012; 3:185–190.
Google Scholar
Qiao Y.P., Wang A.L., Fang L.W., et al. High-risk human papillomavirus infection and associated factors among HIV-positive women in high HIV-burden areas of China. Biomed Environ Sci 2020; 33:206–212.
Google Scholar
Chen F.F., Guo W., Qin Q.Q., et al. Spatial-temporal distribution of newly detected HIV/AIDS cases among aged 15 years or older women in China, 2010-2016. Chin J Epidemiol 2018; 39:739–744.
Google Scholar
National Health Commission. The technical protocol for rural women cervical cancer screening. 2009;
Available: here [Accessed 2009]
Google Scholar
Konopnicki D., Manigart Y., Gilles C., et al. High-risk human papillomavirus infection in HIV-positive African women living in Europe. J Int AIDS Soc 2013; 16:18023.
Google Scholar
Chen Z.F., Ma X.H., Ding Y., et al. The risk factors and genotype distribution of high risk human papillomaviral infection among Uighur women in Xinjiang. Chin J Prev Med 2011; 45:371–373.
Google Scholar
Siu S.S., Cheung T.H., Chan P.K., et al. Patients with malignant or pre-malignant cervical lesion have increased risk of becoming hepatitis B carrier. J Exp Clin Canc Res 2007; 26:77–81.
Google Scholar
Liu J., Liu W., Liu Y., et al. Prevalence of microorganisms co-infections in human papillomaviruses infected women in Northern China. Arch Gynecol Obstet 2016; 293:595–602.
Google Scholar
Zhang R., Shi T.Y., Lu H., et al. Risk factors for human papillomavirus infection in Shanghai suburbs: a population-based study with 10,000 women. J Clin Virol 2013; 58:144–148.
Google Scholar