Comparing Two Platforms of Detection for High Risk Human Papillomavirus in Both SurePath and ThinPrep Gynecologic Specimens
Angelique Levi, Pei Hui, Kara Dutch, Kevin Schofield, David C Chhieng. Yale University, New Haven, CT
Background: HR HPV DNA testing is currently recommended for the triage of all women with ASC-US and in conjunction with Pap tests for women >30. Most of the commercially available methods including the Hybrid Capture 2 (HC2) assay (Qiagen, Gaithersburg, MD) and the more recent Cobas 4800 (C4800) test (Roche, Branchburg, NJ) are approved by the FDA for ThinPrep (TP) (Hologic, Boxborough, MA), but not for SurePath (SP) (BD, Burlington, NC). As part of a validation process, we compared the performance of C4800 test and the HC2 assay in the detection of HR HPV using both TP and SP.
Design: A total of 1,371 liquid based gynecologic preparations, 1,122 SP and 249 TP specimens, were tested for HR HPV DNA using the C4800 and the HC2 assay. For discrepant test results between C4800 and HC2 assay, confirmatory test was performed using Linear Array HPV testing (Roche). The concordance between the CTQ and HC2 assays was analyzed using the kappa statistic.We calculated the diagnostic accuracy of both C4800 and HC2 assays. A sample was treated as a true negative if results of both the C4800 and the HC2 assay were negative and a true positive if results of both the C4800 and the HC2 assay were positive or if the result of the Linear Array test confirmed a positive assay result for cases with discordant results.
Results: Table 1 summarized the results of HR HPV DNA testing using both assays. The kappa value was 0.73. There was no statistical significant difference in the percentage of positive HPV results among SP and TP. A total of 82 (6.0%) cases demonstrate discordant HPV results, 70 (6.2%) cases were SP and 12 (4.8%) cases TP. (p > 0.05) 27 out of 82 (32.9%) cases were C4800-/HC2+ and 55 out of 82 (67.1%) were C4800+/HC2-; 10% C4800-/HC2+ and 70% C4800+/HC2- cases were confirmed positive for HR HPV by Linear Array.
|Type of specimens||Number of cases tested negative for both assays||No. of cases tested positive for both assays||No. of discordant results between the 2 assays||Kappa Value||Diagnostic accuracy for C4800 assay||Diagnostic accuracy for HC II assay|
|SurePath (n=1,122)||949 (84.6%)||103 (9.2%)||70 (6.2%)||0.71||99.0%||95.2%|
|ThinPrep (n=249)||211 (84.7%)||26 (10.4%)||12 (4.8%)||0.79||99.6%||95.6%|
|Total (n=1,371)||1,160 (84.6%)||82 (6.0%)||82 (6.0%)||0.72||99.1%||95.3%|