HPValidate-human papillomavirus testing with DNA and mRNA assays on self-collected samples in cervical screening: comparison of test characteristics on three self-sampling devices.

Background: Relative test accuracy of human papillomavirus (HPV) testing on self vs. clinician-collected samples may depend on the specific combination of a self-sampling device and HPV assay.

Methods: Five self-sampling workflows were studied within the routine English cervical screening programme; the cobas HPV DNA and APTIMA HPV mRNA assays with the Evalyn brush, Self Vaginal FLOQSwabs (FLOQSwabs) and the Multitest kit. To study test sensitivity, women were recruited at routine colposcopy appointments; to study test specificity, women were recruited at routine screening appointments.

Results: The estimated conditional relative sensitivity for high-grade cervical intraepithelial neoplasia (CIN2+) was 0.90 (90% CI: 0.84-0.94) for the Evalyn + cobas workflow, 0.94 (0.90-0.97) for FLOQSwabs + cobas, 0.77 (0.69-0.83) for Evalyn + APTIMA, 0.92 (0.85-0.96) for FLOQSwabs+APTIMA and 0.92 (0.86-0.96) for Multitest+APTIMA. The estimates of the relative specificity were 0.96 (0.95-0.98), 0.91 (0.90-0.93), 0.99 (0.97-1.01), 0.89 (0.87-0.92) and 0.87 (0.85-0.89), respectively. The specificity estimates were sensitive to the inclusion of certain subgroups of women. HPV detection rates were higher for all self-sample than clinician-sample workflows.

Conclusions: The relative test sensitivity of four self-sampling workflows including both DNA and mRNA HPV assays was relatively close to that associated with clinician-collected samples.