Short-Term Reproducibility and Comparative Screening Performance of Paired Cervical Cancer HPV DNA Tests Among Women in Senegal

Abstract

Cervical cancer remains a major public health challenge in sub-Saharan Africa, where access to effective screening programs remains limited. Human papillomavirus (HPV) DNA testing has emerged as a highly sensitive screening strategy for cervical precancer and cancer, although less is known about the short-term reproducibility of repeat HPV testing in high-burden settings. This analytic observational study used secondary data from two completed Senegalese cohort studies conducted between 1998 and 2006 to evaluate the reproducibility and comparative screening performance of paired cervical HPV DNA tests collected within 119 days among 768 women. Agreement between Swab 1 and Swab 2 was evaluated using percent agreement and Cohen’s kappa (κ) for overall high-risk HPV (hrHPV), low-risk HPV, and genotype-specific detection. Screening performance analyses compared four paired testing strategies: Swab 1 positive, Swab 2 positive, either swab positive, and both swabs positive. Sensitivity and specificity were evaluated for detection of cervical intraepithelial neoplasia grade 2 or worse (CIN2+) and invasive cervical cancer (ICC). Exploratory logistic regression analyses examined factors associated with discordant paired hrHPV results. Overall reproducibility for any hrHPV detection was substantial (κ = 0.74, 95% CI: 0.69–0.79), with almost perfect agreement observed for HPV16 (κ = 0.85, 95% CI: 0.79–0.91). Agreement remained generally substantial across participant subgroups and cervical disease categories. Discordance was more likely when samples were collected 30–59 days apart than within 0–29 days and was less common among women with CIN2+, ICC, or HIV infection. Compared with a single swab strategy, classifying either swab as positive increased sensitivity for detection of both CIN2+ and ICC by approximately 7–8%, although specificity declined modestly. Strengths of this study include the use of paired cervical HPV DNA samples from a well-characterized Senegalese population, inclusion of women across the cervical disease spectrum, and genotype-specific PCR-based HPV testing which enabled direct evaluation of short-term reproducibility in a West African setting. Limitations include the historical nature of the cohorts, potential variability in the timing between paired samples, and sparse observations within some subgroup analyses. Because the study evaluated short-term reproducibility rather than longitudinal persistence, findings should not be interpreted as evidence of HPV clearance or progression over time. Future research should evaluate reproducibility using newer HPV testing platforms, self-collected samples, point-of-care assays, and HPV screening strategies implemented in vaccinated populations.