Investigator
Unknown Institution
Urine human papillomavirus testing for cervical screening in a UK general screening population: a diagnostic test accuracy study
Background Cervical screening uptake is decreasing in the UK, with only 67.5% of those eligible under 50 years old attending in 2022. Barriers include restricted access to screening appointments and poor acceptability of the speculum examination. Urine self-sampling is an alternative cervical screening method that has the potential to improve uptake. Aim To determine the clinical performance and acceptability of human papillomavirus (HPV)-tested urine for cervical screening in a UK general screening population. Design and setting Prospective, cross-sectional diagnostic test accuracy study in North-West England. Method Urine was self-collected using a first-void urine (FVU) collection device (DNA Genotek Colli-Pee® 10 ml with urine conservation medium) before obtaining matched routine cervical screening samples. HPV testing used Roche Cobas® 8800 at cervical sample thresholds. A questionnaire evaluated urine self-sampling acceptability. HPV-positive cervical samples underwent reflex cytology, managed under standard NHS protocols, and clinical outcomes were collected. Results In total, 1517 participants provided matched urine and cervical samples. There were 207 of 1517 (13.6%) cervical and 245 of 1517 (16.2%) urine samples that were HPV positive with a 1.6% ( n = 25/1517) incidence of cervical intraepithelial neoplasia (CIN)2+ following colposcopic assessment ( n = 80). The specificity of urine was non-inferior ( P = 0.0004) to the specificity of cervical samples at 85.19% (95% confidence interval [CI] = 83.28 to 86.95) versus 87.80% (95% CI = 86.03 to 89.42), giving a relative specificity of 0.97 (95% CI = 0.95 to 0.99). Urine detected 24 of 25 (96.0%) participants with CIN2+. In the future, 41.6% ( n = 575/1382) of participants would prefer current cervical screening, compared with 30.0% ( n = 414/1382) with no preference and 28.4% ( n = 393/1382) preferring urine self-sampling. Conclusion HPV-tested urine showed non-inferior specificity to cervical samples in a general screening population. Urine self-sampling was acceptable to current attenders but some prefer traditional screening, making choice an important consideration for policymakers.
