Transition from Hybrid Capture 2 to Cobas 4800 in Hpv detection: sensitivity and specificity for Cin2+ in two time periods.

Background: High-risk (HR) Human Papilloma Virus (HPV) Tests for HPV detection differ in sensitivity and specificity. In this study, we evaluated the sensitivity and specificity of the HC2 HR HPV Test and the Cobas 4800 HPV Test in consecutive cervical samples collected from a referral population with a high prevalence of disease, using CIN2+ histology as clinical outcome.
Methods: Ten thousand two-hundred and thirteen consecutive cervical samples were assayed for HR-HPV in the Laboratory Medicine Division of IEO: 5140 from January 2012 to June 2013 with HC2 and 5073 from July 2013 to December 2014 with the Cobas HPV Test. These two assays differ in terms of target genes and testing methods.
Results: The test positivity rates for HC2 and Cobas 4800 were 29.5% (1515/5135, 95% CI 28.3-30.8%) and 23.9% (1212/5069, 95% CI 22.7-25.1%), respectively. The detection rates of CIN2+ in the two time periods were 2.8% (145/5140, 95% CI 2.4-3.3%) and 1.6% (79/5073, 95% CI 1.2-1.9%), respectively. The sensitivity for CIN2+ for HC2 and Cobas 4800 was 95.2% (138/145, 95% CI 91.7-98.7%) and 93.7% (74/79, 95% CI 88.3-99.0%), respectively. The specificity for CIN2+ for HC2 and Cobas 4800 was 72.4% (3613/4990, 95% CI 71.2-73.6%) and 77.2% (3852/4990, 95% CI 76.0-78.4%), respectively. There were 23 cases of cancer in each of the two time periods. HC2 detected 100% (23/23). Cobas 4800 detected 82.6% (19/23).
Conclusions: The detection rate of CIN2+ was higher in the first period than in the second period. There was no significant difference in sensitivity of HC2 and Cobas 4800 in women with CIN2+. The specificity of CIN2+ using Cobas 4800 in the second period was higher than HC2 in the first period, probably due to the lower prevalence of CIN2+ in the second period.