Clinical and analytical evaluation of the RealTime High Risk HPV assay in Colli-Pee collected first-void urine using the VALHUDES protocol

Abstract

ObjectiveUrine self-sampling has gained increasing interest for cervical cancer screening. In contrast to analytical performance, little information is available regarding the clinical accuracy for high-risk Human Papillomavirus (hrHPV) testing on urine. MethodsVALHUDES is a diagnostic test accuracy study comparing clinical accuracy to detect high-grade cervical precancer (CIN2+) of HPV testing on self-collected compared to clinician-collected samples (NCT03064087). Disease outcome was assessed by colposcopy and histology. The Abbott RealTime High Risk HPV assay performance was evaluated on Colli-Pee collected first-void urine with cervical outcomes as comparator. ResultsAs no assay cut-off for urine has been clinically validated, we used the predefined cut-off for cervical samples (CN ≤ 32). Using this cut-off, hrHPV testing was similarly sensitive (relative sensitivity 0.95; 95% CI: 0.88–1.01) and specific (relative specificity 1.03; 95% CI: 0.95–1.13) for detection of CIN2+ compared to testing cervical samples. In the subgroup of women of 30 years and older, similar relative sensitivity (0.97; 95% CI: 0.89–1.05) and specificity (1.02; 95% CI: 0.93–1.12) was found. Additionally, an exploratory cut-off (CN ≤ 33.86) was defined which further improved sensitivity and analytical test performance. ConclusionHrHPV-DNA based PCR testing on home-collected first-void urine has similar accuracy for detecting CIN2+ compared to cervical samples taken by a clinician.