High-risk HPV Testing With Pap Smear Triage for Cervical Cancer Screening in Women Aged 50 and Older in Ethiopia (GOLDEN-Et)

Background: Cervical cancer is the second most common cancer in Ethiopia, with an estimated 8,168 new cases and 5,975 deaths each year. Screening programs worldwide usually target women aged 30 to 49. Women aged 50 and older are increasingly diagnosed at advanced stages and have higher mortality, partly because hysterectomy rates are falling and life expectancy is rising. Ethiopia's 2025 national guideline recognizes unscreened women over 50 as a high-risk group for the first time, but evidence to guide their screening is limited. Visual inspection with acetic acid performs poorly in older women because lesions are harder to see. Cytology is also less reliable because smears from older women are often hypocellular and atrophic. hrHPV testing is more sensitive in this group, since a positive result more reliably reflects persistent infection, but hrHPV-positive women still need triage to identify disease. Cytology remains the recommended triage method above age 50, and no direct evidence exists on how these methods perform in this age group in Ethiopia.

Design: GOLDEN-Et (Genotyping and Outcome of hrHPV with cytology triage for Detection and Evaluation in older Ethiopian womeN) is an observational, prospective, cross-sectional, multicenter, quantitative study with two linked components: a prevalence component and a diagnostic accuracy component. It is exploratory rather than confirmatory. No intervention or treatment is assigned to participants, and there is no randomization. All consenting participants follow the same testing pathway. Two nested analysis sets are used. The prevalence analysis set includes all consenting, eligible women with a valid hrHPV result. The diagnostic accuracy analysis set includes hrHPV-positive women from that set who also consent to and complete cytology, colposcopy, and biopsy.

Procedures: Trained data collectors and nurses recruit women at the study sites, either among women attending the facility for other reasons or through dedicated screening outreach. Study staff, not treating clinicians, obtain consent. Each participant receives a pseudonymized study ID and completes a baseline questionnaire. Each participant then collects a self-sample for hrHPV testing with the ScreenFire RS assay, which reports positivity and genotype in four risk channels: HPV16; HPV18/45; HPV31/33/35/52/58; and HPV39/51/56/59/68. Recruitment for the diagnostic accuracy component runs on a rolling basis. As each hrHPV result becomes available, women who test positive are approached individually for additional consent and scheduled for cytology, colposcopy, and biopsy. Cytology is a conventional Pap smear read according to the Bethesda System. The biopsy result (CIN2+ present or absent) is the reference standard. A pathologist blinded to the hrHPV and cytology results interprets all biopsy specimens.

Baseline variables and confounders: Baseline demographic and clinical data are collected for all participants to describe the study population. Age, HIV status, prior screening history, and parity are considered as potential confounders, because they are related to both HPV persistence and disease progression.

Statistical analysis: Continuous variables are summarized with means and standard deviations or medians and interquartile ranges, and categorical variables with counts and percentages. hrHPV prevalence is reported with a 95% confidence interval, and genotype distribution as counts and percentages among hrHPV-positive women. Cytology is compared with the biopsy reference standard in a 2x2 table, with 95% confidence intervals for the accuracy measures listed under Outcome Measures. The two referral strategies described in the Brief Summary are applied retrospectively to the same hrHPV-positive women and compared descriptively. Because the same women contribute to both strategies, a McNemar test compares CIN2+ detection between them at a two-sided significance level of 0.05. This is the only pre-specified hypothesis test. Its result is interpreted together with the absolute differences between strategies. No correction for multiple testing is applied, and all other analyses are descriptive. No formal power calculation is applied to the McNemar test.

Missing data: Women with an unsatisfactory hrHPV result are excluded from the prevalence analysis set. Unsatisfactory cytology specimens are excluded from the relevant diagnostic accuracy calculations. The number and proportion of unsatisfactory specimens are documented and reported. Missing data are handled by complete-case analysis, with sensitivity analyses considered if missingness is substantial. Analyses are performed in SPSS.

Data management: Data are collected in REDCap using pseudonymized IDs. Before analysis, data are checked for completeness and consistency, including verification that hrHPV, cytology, and biopsy results are linked to the correct participant ID. Directly identifiable data and the code list are stored separately with restricted access.