Review of the literature: Agreement between cobas, CLART, APTIMA, and Hybrid Capture 2 assays.

The list of prior studies was compiled by searching for English-language papers in PubMed using keywords “cobas”, “CLART”/”Clinical Arrays”, or “APTIMA”, each in combination with “HPV”; the search was undertaken in June 2013. Studies were selected if cobas, CLART or APTIMA were compared to Hybrid Capture 2.
Abbreviations: hr = high risk not otherwise specified; NR = not reported; NA = not available.
a Primary samples from women aged 30 to 65 years.
b Women without cervical intraepithelial neoplasia grade two or worse.
c Numbers of samples having an invalid outcome were not reported by pair of assays. Hence, numbers of samples testing positive on both assays were estimated from the reported totals of assay specific frequencies of positive outcomes, and the reported frequencies of discordant samples for pairs of assays; numbers testing negative on both assays were not calculated.
d Non-denatured clinical specimens. Of the 568 samples, 56 were from women below age 30 years, and 512 from women aged ≥30 years who had human papillomavirus testing performed as part of routine screening in conjunction with a liquid based Pap.
e Primary samples from women aged 30 to 65 years having atypical squamous cells of undetermined significance or worse on cytology (all recommended for colposcopy or for repeated testing).
f Based on the total sample, n = 356.
g Estimated based on the proportion of samples testing positive on cobas if Hybrid Capture 2 positive (90.6%; n tested in Predictors 2 study = 668), and the proportion of samples testing positive on Hybrid Capture 2 if cobas positive (95.1%; n tested in Predictors 1 and 2 studies = 1,199).
h Interquartile range.
i Estimated based on the proportion of samples testing positive on APTIMA if Hybrid Capture 2 positive (86.6%; n tested in Predictors 1 and 2 studies = 1,225), and the proportion of samples testing positive on Hybrid Capture 2 if APTIMA positive (98.6%; n tested in Predictors 1 and 2 studies = 1,199).