HPV Screening With Triage by HPV Genotyping Versus Visual Inspection With Acetic Acid

Promoting Comprehensive Cervical Cancer Prevention and Better Women's Health in Low- and Medium Resource Settings HPV Screening With Triage by HPV Genotyping Versus Visual Inspection With Acetic Acid: a Randomized Controlled Trial

University Hospital, Geneva, Bafoussam Regional Hospital, Cameroon, Dschang District Hospital, Cameroon, University of Dschang

Cervical cancer is the leading cause of cancer death among women in sub-Saharan Africa, despite the existence of effective prevention and screening methods. Because vaccination rates against human papillomavirus (causing nearly all cervical cancers) are still insufficient in some low-resource countries, early detection and treatment of cervical lesions at risk of progressing to cancer are crucial components of cervical cancer control. Therefore, it is essential to find the most reliable and appropriate screening strategy in the context of low-resource countries in order to identify women in need of treatment and thus prevent the development of cervical cancer. The objective of our study is to compare two different methods of cervical cancer screening adapted to low-resource settings, in two study centers in Cameroon.

HPV used as a stand-alone test has a limited specificity and positive predictive value and as a consequence, a significant number of HPV-positive women have no cervical precancerous lesions or cancer and receive unnecessary workup and treatment. For this reason, the WHO has recommended visual inspection with acetic acid (VIA/VILI) as a triage test of HPV-positive women to identify women requiring treatment. Nevertheless, VIA is a highly subjective procedure dependent on the health care provider's experience, with diagnostic accuracy varying from setting to setting. Triage by HPV genotyping has recently emerged as an alternative to triage by VIA, with immediate treatment of women with a subset of high-risk HPV genotypes only, thus reducing overtreatment rates. However, to date, the triage of HPV-positive women by VIA versus HPV genotyping has not yet been compared. This project aims to implement primary HPV-based screening in Cameroon followed by an immediate offer for treatment by thermal ablation after randomization for triage by HPV genotyping or VIA. More specifically, we aim to determine if triage by HPV genotyping (with immediate treatment of women with HPV types 16, 18, 45, 31, 33, 35, 52 or 58) allows better targeting of women needing treatment and allocation of resources to women at-risk than triage by VIA, as recommended by the WHO. Primary objective: To identify the most efficient screening strategy for cervical cancer in Cameroon, more specifically to determine whether triage by a pool of eight genotypes (HPV types 16, 18, 45, 31, 33, 35, 52 or 58) is more effective than triage by visual inspection with acetic acid for detection of precancerous lesions. ¨ Secondary objectives: To determine the overtreatment rate in each screening group (HPV genotyping and VIA/VILI) To determine the rate of adverse events (e.g. hemorrhage, infection, hospitalization) in each screening group To determine which participant characteristics may be associated with better prediction of CIN2+ for each screening group To assess patient and health care provider acceptability of both screening strategies To create a sustainable structure for the promotion of women health with a priority made in the prevention of cervical cancer West Region of Cameroon To treat all precancerous or cancerous lesions discovered during the screening To inform women and their families about gynecological pathologies, including cervical cancer, sexually transmitted diseases (STD) and HIV To create a database of cervical images for continuous clinical education To develop an Automated VIA/VILI Classifier (AVC) that can help identify cervical precancerous lesions based on a 2-minute video of the cervix during VIA/VILI To assess women's, the community's and healthcare providers' acceptability of the AVC test Study Design: National multicentric open-label two-arm randomized controlled trial Qualitative and quantitative studies for participants and health care providers will be included during the study period addressing preferences and attitudes toward the screening process and treatment.

Allocation to the study arm will be concealed until primary HPV screening by the GeneXpert machine has been performed, in order to ensure that the health care provider enrolling the participant into the study and the participant herself have no a priori knowledge of group assignment. Detected HPV genotypes will only be revealed to the midwife executing the pelvic examination and treatment if the participant is randomized in the intervention arm (triage by genotyping), in order to know whether to treat the participant or not. If the participant is randomized in the control arm (triage by VIA), the midwife will be blinded to the exact HPV genotype channel(s) in order to avoid influence of treatment. The pathologist performing histological analysis (gold standard) will be blinded to the study arm and the on-site diagnosis.

The study intervention will consist in applying HPV genotyping as a triage method of HPV-positive women for cervical cancer screening. After primary screening with the Xpert HPV test, positive women will be sorted according to two groups of genotypes: group 1 (HPV types 16,18,45, 31, 33, 35, 52 and/or 58 obtained from channels 1, 2 and 3) and group 2 (HPV types 51, 59, 39,56, 66 and/or 68 obtained from channels 4 and 5). Women of group 1 will immediately treated, while those of group 2 will not receive immediate treatment and will be followed-up at 12 months. An exception will be made for participants with lesions suspicious of invasive cancer upon examination, which will be referred for further investigations regardless of the HPV type.

The control arm will consist in triage of HPV-positive women by VIA, as currently recommended by the WHO. Women with a positive VIA will be treated immediately, while VIA-negative women will not be treated and will be followed-up at 12 months.

Genotyping will be obtained by the Xpert system which uses 5 color channels containing primers and probes for the detection of specific genotypes or pooled results as follows: i) HPV 16, ii) HPV 18/45 in pooled result, iii) HPV types 31, 33, 35 52, or 58, in pooled result, iv) HPV types 51 or 59, in pooled result, and v) HPV types 39, 56, 66 or 68 in pooled result.

After application of acetic acid and Lugol's iodine, the cervix will be assessed using simplified "ABCD criteria" (A= acetowhite lesion within the transformation zone, B = spontaneous bleeding or upon slight touch, C (optional) = Lugol-positive coloring of acetowhite lesions, D = diameter > 5mm of acetowhite lesion).

Sensitivity of triage by HPV genotyping and VIA/VILI for cervical intraepithelial neoplasia grade 2 or more severe (CIN2+) detection

Sensitivity of triage by HPV genotyping and VIA/VILI for cervical intraepithelial neoplasia grade 2 or more severe (CIN2+) detection at time of screening (first visit), considering histologic results (from cervical biopsy and/or endocervical brushing) as the gold-standard.

Specificity, positive predictive value and negative predictive value of triage by HPV genotyping and VIA/VILI for cervical intraepithelial neoplasia

Specificity, PPV and NPV of triage by HPV genotyping and VIA/VILI for cervical intraepithelial neoplasia grade 2 or more severe (CIN2+) detection at time of screening (first visit), considering histologic results as the gold-standard.

Percentage of participants who have correctly followed the screening, triage and treatment strategy in each study arm

The percentage of participants who have correctly followed the screening, triage and treatment strategy in each study arm will be measured to assess the feasibility of both triage strategies. This will be measured by study case report forms for both arms.

Overtreatment rate in each screening group, considered as treatment of participants with <CIN2 on histology

Odds ratios of correct prediction of CIN2+ for various participant characteristics in each screening group

Estimated number of avoided cases of cervical cancer in each screening group based on identified and treated precancerous lesions and known rate of progression to cervical cancer if no treatment is provided.

Satisfaction with the screening process in both study arms will be measured on a scale from 0 to 10 (on a paper survey) among participants at the end of the first visit, and by a qualitative analysis of data from focus group discussions and individual interviews with health care providers and patients, in order to assess acceptability of both strategies.

Prevalence of HPV infection and cervical pre-cancer and cancer among Cameroonian women, based on HPV results obtained by self-sampling and analysis by the GeneXpert assay and on histological analyses of cervical biopsies and endocervical brushing.

Creation of a database of anonymized cervical images for clinical education, captured by smartphone photography during the VIA/VILI process.

Inclusion Criteria: HIV-negative women aged 30-49 and HIV-positive women aged 25-49 years old Ability to understand study procedures and accepting voluntarily to participate by signing an informed consent form (ICF). Exclusion Criteria: Pregnancy at the time of screening Previous hysterectomy Known cervical cancer Symptoms of cervical cancer (e.g. metrorrhagia, known pelvic mass) Conditions that can interfere with visualization of the cervix Severe pre-existing medical conditions (e.g. advanced cancer, terminal renal failure) Women who are not able to comply with the study protocol.