High-risk human papillomavirus detection in self-collected vaginal samples compared with healthcare worker collected cervical samples among women attending gynecology clinics at a tertiary hospital in Pretoria, South Africa

Background In 2017, the South African National Department of Health (NDoH) Cervical Cancer Prevention and Control Policy was revised. Human papillomavirus (HPV) testing on self-collected samples may offer improved screening uptake. The objectives of the study were to compare the positivity of high-risk (hr)-HPV deoxyribonucleic acid (DNA) and hrHPV viral messenger ribonucleic acid (mRNA) between healthcare worker-collected cervical and self-collected vaginal samples and investigate the accuracy of the applicator-tampon-based self-collected samples in detecting hrHPV DNA and hrHPV mRNA. Methods A total of 527 women aged 18 years and older and seeking gynecology services at a tertiary hospital in Pretoria, South Africa, were enrolled. Vaginal samples were self-collected using SelfCerv applicator tampon, followed by cervical samples collected by a healthcare worker using a Cervex Brush® Combi. Both samples were tested with the Abbott m2000 analyzer for 14-hrHPV types and 285 paired samples were tested for hrHPV E6/E7 mRNA using the Aptima HR-HPV mRNA assay. The prevalence of hrHPV DNA and hrHPV E6/E7 mRNA was estimated and the positivity between the two collection methods was compared for the total group as well as per age group. Results HrHPV prevalence was 48.0% (95% CI 43.7–52.4) among healthcare worker collected samples and 47.6% (95% CI 43.3–52.0) among self-collected samples. There was no difference in positivity between healthcare worker collection (48.0%) and applicator-tampon-based self-collection, 47.6% (p-value = 0.90). The proportions of hrHPV were equal between the age groups as shown by the McNemar test (p = 0.9036) results for correlated proportions. The prevalence of hrHPV mRNA was 78.6% (95% CI 73.4–83.2) and 58.6% (95% CI 52.6–64.4) for healthcare worker- and self-collection, respectively. The McNemar test for correlated proportions was highly significant (p < 0.0001), indicating that the hrHPV mRNA proportions are not comparable, although this differed between age groups. Conclusions Applicator-tampon-based self-collection has a comparable hrHPV DNA positivity rate as healthcare worker collection but different positivity rates for hrHPV mRNA. Self-sampling showed high concordance with healthcare worker-collected sampling for hrHPV DNA detection, especially regarding HPV 16/18 detection. HrHPV DNA was equally detected between the total group as well as per age group. Implementation of self-sampling using an applicator tampon as a primary screening tool may be considered.