Can we increase the cervical cancer screening interval with an HPV test for women living with HIV? Results of a cohort study from Maharashtra, India

We are reporting (a) updated incidence of cervical intraepithelial neoplasia (CIN) among women who did not have colposcopic or histopathological disease at baseline and (b) disease outcomes among women treated for CIN and their follow‐up HPV status; in a cohort of women living with HIV (WHIV). The median overall follow‐up was 3.5 years (IQR 2.8‐4.3). The incidence of any CIN and that of CIN 2 or worse disease was 16.7 and 7.0 per 1000 person‐years of observation (PYO), respectively. Compared with women who were HPV negative at baseline, women who cleared HPV infection had 23.95 times increased risk of incident CIN 2 or worse lesions (95% CI 2.40‐661.07). Women with persistent HPV infection had 138.18 times increased risk of CIN 2 or worse lesions (95% CI 20.30‐3300.22). Complete disease regression was observed in 65.6% of the HPV positive women with high‐grade CIN and were treated with thermal ablation but HPV persistence was seen in 44.8% of those with high‐grade disease. Among those who did not have any disease at baseline and were also HPV negative, about 87% (95% CI 83.79‐89.48) women remained HPV negative during consecutive HPV test/s with the median interval of 3.5 years. Long‐term surveillance of WHIV treated for any CIN is necessary for the prevention of cervical cancer among them. Our study provides an early indication that the currently recommended screening interval of 3 to 5 years among WHIV may be extended to at least 5 years among HPV negative women. Increasing the screening interval can be cost saving and improve scalability among WHIV to support WHO's cervical cancer elimination initiative.


What's new?
Women living with HIV (WHIV) are at disproportionately high risk of human papillomavirus (HPV) infection and subsequent cervical cancer. The extent to which cervical intraepithelial neoplasia (CIN) affects WHIV and outcomes among those treated for CIN are not fully known. Here, among women in India, WHIV with persistent HPV infection had a 138.18-fold increase in risk of CIN2 or worse. Two-thirds of HPV-positive WHIV treated with thermal ablation for highgrade CIN experienced complete disease regression. Nonetheless, HPV persisted in nearly 45% of ablation-treated WHIV. A significant majority of WHIV who were HPV-negative at baseline remained HPV-free in subsequent screenings.

| INTRODUCTION
Cervical cancer is eminently preventable by human papillomavirus (HPV) vaccination and early detection and treatment of precancerous lesions, but it is still a major public health problem in the resource limited countries. The International Agency for Research on Cancer (IARC) of the World Health Organization (WHO) estimated that there were 604 127 new cases and 341 831 deaths due to cervical cancer globally in 2020. 1 China and India account for more than a third of the global burden of cervical cancer. 1 About 58.5% deaths due to cervical cancer are from Asia. 2 Cervical cancer disproportionately affects women living with human immunodeficiency virus (WHIV). 1 Many of the low-and middle-income countries (LMICs) bearing high cervical cancer burden are also the countries with high HIV burden. The risk factors for HIV and HPV often overlap and WLHIV have a significantly higher risk of persistent HPV infection, making them more susceptible to develop cervical cancer. [3][4][5] WLHIV have a 6-fold increased risk of cervical cancer as compared with women in the general population. 5 Improved access to antiretroviral therapy (ART) has increased the life expectancy in HIV-infected individuals. 6 Early ART initiation and sustained adherence is likely to reduce the incidence of cervical cancer in WHIV provided cervical cancer screening and management of precancer is integrated into the ART program. 7 However, compared with the general population WLHIV require more intensive screening and long-term follow-up, if they are HPV positive and/or treated for high-grade CIN. 8,9 The WHO released the guidelines for cervical cancer screening and precancer treatment in the year 2021 that also specify the screening and follow-up algorithms for WLHIV. 10 However, many of the recommendations are based on "weak evidence" as not many studies have systematically followed up WLHIV, especially those positive on HPV test and/or treated for precancer.
Considering the need to advance the understanding of secondary prevention of cervical cancer in WLHIV, we initiated a cohort study in 2010 that enrolled 1153 WHIV from Maharashtra, India. Based on the cohort, we have previously reported the performance of different screening tests (visual inspection with 3-5% acetic acid [VIA], visual inspection with Lugol's iodine [VILI], high-risk HPV testing and cytology), 11 HPV genotype distribution among those with and without CIN, 12 incidence of CIN according to the HPV genotypes 13

| METHODS
The study was conducted at a designated cervical cancer screening clinic at Prayas, Pune, India, a nongovernmental, nonprofit organization. We enrolled 1153 WLHIV between 2010 and 2011 in a longitudinal cohort study and the study procedures and the baseline results have been published earlier. [11][12][13][14][15][16] The study procedures at baseline and subsequent follow-up visits are explained in Figure 1. Consecutive, serologically confirmed WHIV, regardless of their CD4+ cell counts and ART status, were recruited after a written informed consent if they were aged between 21 and 60 years, had an intact uterus, did not have significant utero-vaginal prolapse, who were not pregnant and were not diagnosed with CIN or cervical cancer in the past. The enrolled women were interviewed by a female social worker and a structured questionnaire was used to collect information on sociodemographic, sexual, reproductive, medical and HIV infection related characteristics. History regarding the time since diagnosis of HIV infection, nadir CD4 count and time since ART was noted based on their available medical records.
Then a trained nurse collected a cervical specimen in the Digene specimen transport medium (STM) for Hybrid Capture 2 (HC2) assay, another for HPV genotyping in Preservcyt medium and for cytology.
The nurse also performed VIA and recorded the findings. A trained colposcopist performed colposcopy on all the women at baseline.
When colposcopy was suggestive of a low-grade or a high-grade lesion and the woman was eligible for ablative treatment, thermal ablation was performed in the same sitting after obtaining punch biopsies. If the lesion was not eligible for thermal ablation, large loop excision of the transformation zone (LLETZ) was performed later.
Women treated with thermal ablation or LLETZ were recalled after 6 to 12 weeks to exclude treatment complications and to review the histopathology report and then after 1 year. Women who were not treated at baseline were called initially after 6 months (to review the HPV and cytology reports) and every year thereafter for the first 3 years. Both the treated and the untreated women underwent colposcopy at the yearly visits. If colposcopy was suggestive of any low-grade or worse lesion, directed punch biopsies were collected from abnormal areas and immediate treatment with thermal ablation was offered when eligible or LLETZ at a later date.
All the women were followed up with yearly colposcopy for the first three consecutive years, following which yearly follow-up was replaced by 3-yearly HPV DNA test (HC2) at least for additional two rounds. Women who tested HPV negative, were called to the clinic for a repeat HPV test after 3 years. If the HPV test was positive at any of the visits, the women were called for colposcopy, biopsy and treatment when indicated. Women with persistent HPV and no lesion on colposcopy had random biopsies. Repeat treatment with thermal ablation was offered when eligible. An endocervical curettage was done when the squamocolumnar junction was not visible and there was no visible lesion on colposcopy.

| HPV DNA testing
The cervical specimens in STM were tested at the Nargis Dutt Memorial Cancer Hospital (NDMCH), Barshi, India by HC2 assay for 13 high-risk HPV types (HPV 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59 and 68). Specimens with a ratio of relative light unit to a positive control (RLU/PC) of 1 or more (corresponding to 5000 or more viral copies) were considered to be HC2 positive. HPV test in this manuscript refers to the HC2 test unless specified.

| Histopathology reporting
Cervical biopsy specimens were processed at a local laboratory, and the histopathologic findings were reported according to the CIN terminology. 15 Table 1.

| Statistical analysis
The median duration of follow-up was 3.5 years (IQR 2.8-4.3).   F I G U R E 2 Follow-up outcomes of women enrolled in the longitudinal cohort study 2/3 cases detected among those who were positive for HPV at baseline were 3/30 (10%).  WHIV are based on low-certainty evidence due to the lack of longitudinal data. 10 Our study involving a large cohort of WHIV followed systematically over several years in India has addressed some of these knowledge gaps.
Our analysis shows that the incidence of CIN 2 or worse disease in women who were free of disease at baseline is 7 per 1000 PYO.
Although the number of person-years of observations has increased after we published the incidence of CIN 2 or worse disease earlier, 13 there has not been any change in the incidence with additional  19 Early initiation of ART and sustained adherence can reduce the incidence of CIN and progression to cervical cancer provided screening and treatment of precancer is integrated into the program. 7 In our study, although ART for more than 2 years had a protective effect on incidence of any CIN, we did not observe the same protective effect on incident CIN 2 or worse disease and it is possibly because of small number of observations. In addition, when we initiated the study in 2010, about 24% of women were not on ART 11 and the ART guidelines have changed many times since we enrolled the cohort. 20 We initiated our study in 2010 while the WHO's "test and treat" strategy for HIV-infected individuals was introduced in 2016. 21 The high negative predictive value of the HPV test provides reassurance while allowing the screening interval to be increased 22 among women in the general population, even beyond 5 years. 23  The strongest risk factor for CIN 2 or worse disease was persistent HPV infection. Women who cleared HPV infection were also at a substantial risk of developing CIN 2+ lesions. Posttreatment HPV clearance among the women was low (44.8%) in our study and was very similar to that observed in a Zambian study. 27 Persistence of HPV infection following treatment which increases the risk of subsequent CIN is a concern among WHIV.
Our study has a few limitations. When the study was initiated in 2010, HIV viral load testing was not available at the National AIDS Control Organization's ART centers in India and was unaffordable to the majority of the WHIV outside the public health settings. Therefore, majority of the participants were monitored immunologically with CD4 counts as per the national guidelines at the ART centers prevailing time to time. The long-term follow-up was particularly affected by the first wave of Covid-19 pandemic and extended lockdown in India. Even then, the outcomes of our study are highly reliable. Based on rigorous evaluation of the cohort with multiple tests at baseline as well as subsequent follow-up visits, our study provides the updated estimates of the incidence of CIN among WHIV and outcomes after treatment of CIN. The observation that the HPV negative women have very low risk of CIN 3 or invasive cancer within a followup period exceeding 5 years provides us confidence to increase the HPV screening interval safely to 3 years or more pragmatically to 5 years. The cost of the HPV test is a major factor driving the costeffectiveness analysis of different screening tests for cervical cancer in the low-and middle-income countries. 28 Increasing the screening interval when screened with an HPV test among WHIV will have a huge impact on the cost-effectiveness and scaling-up of the cervical cancer prevention.

AUTHOR CONTRIBUTIONS
Smita Joshi: Participated in the conception of the study and its design, conduct of the study, monitoring, supervision, acquisition and interpretation of the data and the provision of clinical services in the study.
Richard Muwonge: was responsible for the statistical analysis of the data, monitoring the study and the interpretation of the data. Vinay Kulkarni: participated in the conception of the study, conduct of the study and the acquisition and interpretation of the data. Mahesh Mandolkar: was responsible for histopathology reporting, analysis and interpretation of the data. Eric Lucas: monitoring the study, and the interpretation of the data. Sanjay Pujari: acquisition and interpretation of the data. Rengaswamy Sankaranarayanan: had the initial idea and was responsible for the conception, study design and the conduct, monitoring and supervision of the study, acquisition, analysis and interpretation of the data. Partha Basu: acquisition, analysis and interpretation of the data. The work reported in the paper has been performed by the authors, unless clearly specified in the text.