Comparison of Hybrid Capture 2 HPV Borderline or Low Positive Results with the Cobas® HPV Test Results in the ATHENA Trial
Arundhati Rao, Stephen Young, Barbara Body, Carol Eisenhunt, Abha Sharma, Catherine Behrens. Scott and White Hospital, Temple, TX; TriCore Reference Laboratories/University of New Mexico Health Science Center, Albuquerque, NM; LabCorp, Greensboro/Winston-Salem, NC; Diagnostic Cytology Lab, Indianapolis, IN; Roche Molecular Systems, Pleasanton, CA
Background: The Digene Hybrid Capture 2 assay (HC2, Qiagen, MD) is often used as a standard for new HPV detection tests, but at the lower limits of detection the reproducibility, sensitivity and specificity for detection of cervical intraepithelial neoplasia (CIN) is suspect and samples with RLU values of greater than or equal to 1 but less than 2.5 have be retested. In this study we compare the performance of the new FDA- approved Cobas® HPV test (Roche, CA) against the HC2 assay in the borderline or low positive range.
Design: This study is taken from the baseline data of the ATHENA HPV study which enrolled 47,208 women 21 years and older between 2008-2009 who had cytology, HPV testing with the HC2 test, Roche Cobas® HPV Test and Roche Linear Array High Risk Genotyping assay (LA) performed in 5 accredited clinical labs. This sub-analysis includes 331 patients with RLU/CO values on the HC2 test between 1 and 5.0 RLU. When repeat testing for HC2 did not yield a definitive result or had inadequate volume, results were reported as indeterminate (59 patients total, RLU available for 55). For these 331 patients, cytology results were as follows: 194 Negative, 111 ASC-US and 26 ≥ASC-US.
Results: There were no indeterminates or inadequates with the Cobas® HPV Test compared to 55 indeterminates for the HC2 test. Of the 56 negative samples by the LA assay, 2 were negative by the HC2 Test compared to 48 negative by the cobas HPV Test (Negative Percent Agreement [NPA] of 3.6% for HC2 compared to 85.7% for the Cobas® HPV Test). Sequencing revealed NPA of 4.3% for HC 2 compared to 52.2% for the Cobas® HPV test. Results from biopsies adjudicated by 3 expert pathologists revealed sensitivities in this low range was 96.8% for the HC2 test and 90.3% for the Cobas test, but the specificity of the cobas® HPV Test was 24.3% compared to 1.6% for HC2.
Conclusions: In the low RLU positive range, the Roche Cobas® HPV Test showed comparable sensitivity, greater specificity and better NPA than HC2 without the need for retesting when HPV detection or biopsy-confirmed CIN is considered as an end-point. Additionally, the Cobas® HPV Test provides HPV16/18 genotyping result which eliminates the need for reflex testing in patients who are HPV positive and cytology negative.
Category: Gynecologic & Obstetrics
Wednesday, March 21, 2012 1:00 PM
Poster Session VI # 189, Wednesday Afternoon