Accuracy of Allplex HPV HR Detection Full Genotyping Assay on Cervical Samples Compared to Vaginal Self-Samples

Vaginal self-sampling is increasingly integrated into cervical cancer screening programs, requiring evidence of accurate HPV testing on self-collected samples. This study evaluated the clinical accuracy of Allplex HPV HR Detection (Allplex) within VALHUDES, a framework established to assess the clinical performance of HPV assays on self-collected vaginal versus clinician-collected cervical samples. A total of 523 women, recruited at Belgian colposcopy clinics, provided vaginal self-samples using either the Evalyn Brush or Qvintip whereas gynaecologists subsequently collected cervical samples with a Cervex Brush. Both vaginal and cervical specimens were resuspended in 20 mL of ThinPrep medium. Paired specimens from 499 women were tested with Allplex. The relative clinical sensitivity of vaginal self-samples collected with Evalyn Brush and Qvintip for high-grade cervical intraepithelial neoplasia (CIN2+) versus clinician-collected cervical samples was 0.96 (95% CI: 0.91-1.02) and 0.91 (95% CI: 0.82-1.01), respectively. Both self-sampling devices demonstrated higher specificity for < CIN2 compared to clinician-collected cervical samples, with relative specificity of 1.13 (95% CI: 1.02-1.25) on vaginal self-samples collected with Evalyn Brush and 1.09 (95% CI: 0.99-1.19) for those collected with Qvintip. Median Ct-values for overall hrHPV, HPV16, 31, 45, 52, 58 and human Beta-globin gene were significantly lower in clinician-collected cervical compared to vaginal self-samples. In conclusion, Allplex demonstrated comparable clinical sensitivity and higher specificity on vaginal self- versus clinician-collected cervical samples, without the need for cut-off optimization.