CI, confidence interval; CIN, cervical intraepithelial neoplasia;
1 Manufacturer’s positivity threshold for all hrHPV
types in cervical samples and vaginal self-sample: Ct ≤ 40;
2 New a posteriori cut-offs vaginal self-sample: HPV16
Ct ≤ 39, HPV18 Ct ≤ 37, other hrHPV16 Ct ≤ 38.
3.4 hrHPV positivity and concordance
Out of 449 women with matched cervical and vaginal specimens, 256
(57.0%) cervical and 270 (60.1%) vaginal specimens were
hrHPV-positive. Out of 447 matched cervical and FVU samples, 250
(55.9%) cervical swabs and 256 (57.3%) FVU tested hrHPV-positive.
Moderate to excellent agreement with Kappa values ranging from 0.53 to
1.00 between both vaginal and FVU self-collected samples and cervical
scrapes was demonstrated (Tables 4 and 5). In general, vaginal samples
showed higher test agreement with cervical specimens than FVU.
In matched cervical and vaginal hrHPV-positve samples, median Ct values
were significantly lower in vaginal samples compared to cervical for 11
other hrHPV, but not for HPV16 or HPV18 (Supplementary Figure 1). In
matched cervical and FVU hrHPV-positive samples, median Ct values were
always higher in FVU as compared to cervical samples. However, the
difference was not significant for HPV18 (Supplementary Figure 2).
Table 4: Concordance between vaginal self-collected and
clinician-collected cervical samples using manufacturer’s cut-offs.