Background: The Gen-Probe APTIMA Combo 2 (AC2) assay is a second-generation transcription-mediated amplification assay for the detection of Chlamydia trachomatis (CT) and Neisseria gonorrhoeae (NG).
Goal: The goal of this study was to evaluate AC2 performance of endocervical (cx) swabs for the detection of CT and NG using either a specimen or an infected patient standard.
Study design: In a multicenter clinical study, we compared AC2 with Abbott's ligase chain reaction (LCR) and Roche's polymerase chain reaction (PCR; Amplicor or COBAS) for CT, and we compared AC2 with Abbott's LCR and culture for NG. A total of 1569 females were enrolled in the study; we collected cx and first-catch urine (FCU) specimens.
Results: CT prevalence was 13.3% for cx specimens and 13.7% for FCU specimens. NG prevalence was 8.7% and 7.9% for cx and FCU specimens, respectively. When based only on cx specimens, AC2, LCR, and PCR sensitivities for CT were 99.4%, 95.6%, and 95.6%, respectively. However, cx sensitivity for CT was reduced to 92.1%, 86.6%, and 87.1% for each respective assay when based on both cx and FCU specimen results (infected patient standard). NG sensitivities for AC2, LCR, and culture based solely on cx specimen results were 99.2%, 96.1%, and 85.9%, respectively. Based on infected patient standard, the sensitivities of each respective assay were 98.5%, 93.9%, and 84.0%.
Conclusions: The infected patient standard reduces the sensitivity of the endocervical evaluation because some infected patients are positive only with FCU. The reduction in sensitivity is greater when testing for CT. Specificities improved slightly, because some unique cx positives, initially classified as false-positive were confirmed by a positive FCU result. Sensitivity of AC2 was higher than LCR, PCR, and culture. Specificity was slightly lower, but discrepant analysis (using alternate TMA targets) of apparent AC2 false-positives showed that 75% to 80% were true-positives.