Cervical cancer remains a significant public health concern in sub-Saharan Africa where organized screening is limited. This secondary analytic observational study evaluated the short-term reproducibility and comparative screening performance of paired cervical swab HPV DNA tests among women in Senegal. The analysis used existing data from two cohort studies conducted between 1998 and 2006 to examine repeat testing within a maximum interval of 119 days.
The study analyzed data from two previously approved Senegalese cohort studies (referred to in the source as EPI/BIO and Developing New Approaches for Cervical Cancer Control). A total of 768 women contributed paired cervical swab specimens collected within 119 days of one another. The present report is a secondary analysis of those preexisting data and specimens.
Agreement between the first and second swab HPV DNA tests was assessed using simple percent agreement and Cohen’s kappa (κ). Analyses were performed for detection of any high-risk HPV (hrHPV), low-risk HPV, and for genotype-specific detection (notably HPV16). Screening performance was compared across four paired-testing strategies; one key comparison contrasted a single-swab strategy with a strategy that classifies a woman as positive if either swab is positive. Exploratory logistic regression evaluated factors associated with discordant paired hrHPV results.
Reproducibility for any hrHPV detection was reported as substantial, with κ = 0.74 (95% CI: 0.69–0.79). Genotype-specific reproducibility for HPV16 was higher, with almost perfect agreement (κ = 0.85, 95% CI: 0.79–0.91). These statistics indicate consistent short-term detection of carcinogenic HPV types in this sample when swabs were repeated within the defined interval.
Agreement for hrHPV detection remained substantial across multiple subgroups examined in the analysis, including age strata, HIV infection status, education level, marital status, lifetime number of sexual partners, parity, contraceptive use, and categories of cervical disease. Exploratory analyses identified timing between sample collections as a factor: discordance in paired hrHPV results was more likely when samples were collected 30–59 days apart than when collected within 0–29 days. Discordance was less frequent among women who had CIN2+, invasive cervical cancer (ICC), or who were living with HIV infection.
The study compared four paired-testing approaches. A clear finding was that classifying a woman as positive if either swab was positive increased sensitivity for detecting CIN2+ and ICC by about 7–8% compared with a single-swab strategy. The source reports this approximate increase in sensitivity but does not provide additional numeric details of specificity, predictive values, or exact strategy definitions in the abstract. These results suggest modest gains in case detection when using an either-swab positive rule, at the likely cost of decreased specificity (specificity data were not reported in the abstract).
This analysis used previously collected study data and specimens. Ethical oversight for the parent studies was provided by the Senegal Ministry of Health through the Comité National de Recherche en Santé (CNRS)/Direction de la Santé; the Developing New Approaches study additionally received approval from the University of Washington Human Subjects Review Committee. The manuscript is a medRxiv preprint and has not been peer reviewed; the authors note competing interests for one investigator and declare no other competing interests. The source states that the deidentified minimal dataset and analytic code will be made available according to applicable data-governance requirements prior to publication.
Limitations reported or implied by the source include that this is a secondary analysis of data collected between 1998 and 2006 (study time frame reported), that the analysis focuses on short-term reproducibility only (within 119 days), and that full screening performance metrics beyond the stated sensitivity gains were not detailed in the abstract. As a preprint, findings should be interpreted with caution until peer review is completed.
The study demonstrates substantial short-term reproducibility of paired cervical hrHPV DNA testing among women in Senegal, with particularly high agreement for carcinogenic types such as HPV16. The modest increase in sensitivity for CIN2+ and ICC when classifying either swab as positive suggests that paired testing can detect additional high-grade disease; however, the abstract emphasizes that single hrHPV DNA testing remains a reliable screening strategy in this high-burden setting. The findings also underscore the need for cautious interpretation of discordant repeat results, especially among women without evidence of high-grade cervical disease.