A Study of Amplicor Human Papillomavirus DNA Detection in the Atypical Squamous Cells of Undetermined Significance-Low-Grade Squamous Intraepithelial Lesion Triage Study

We analyzed the performance of Amplicor for detecting carcinogenic human papillomavirus (HPV) infections and cervical precancer in women with an atypical squamous cells of undetermined significance (ASCUS) Pap and compared the results with Hybrid Capture 2 (hc2) in the ASCUS and low-grade squamous intraepithelial lesion (LSIL) triage study (ALTS). Baseline specimens collected from women referred into ALTS based on an ASCUS Pap result were prospectively tested by hc2 and retrospectively tested by Amplicor (n = 3,277). Following receiver-operator-characteristics curve analysis, Amplicor performance was analyzed at three cutoffs (0.2, 1.0, and 1.5). Paired Amplicor and hc2 results were compared for the detection of 2-year cumulative cervical intraepithelial neoplasia (CIN) grade 3 and more severe disease outcomes (CIN3+) and for the detection of 13 targeted carcinogenic HPV types. Amplicor at the 0.2 cutoff had a higher sensitivity for the detection of CIN3+ (95.8% versus 92.6%, P = 0.01) but a much lower specificity (38.9% versus 50.6%, P < 0.001) than hc2. Amplicor at the 1.5 cutoff had an identical sensitivity for the detection of CIN3+ (92.6%) and a slightly lower specificity (47.5%; P < 0.001). The positive predictive value of hc2 was higher at all Amplicor cutoffs, whereas referral rates were significantly lower (53.2% for hc2 versus 64.1% at the 0.2 cutoff and 56.0% at the 1.5 cutoff, P < 0.001). Amplicor was more analytically specific for detecting targeted carcinogenic HPV types than hc2. Amplicor at the 1.5 cutoff had comparable performance with hc2. Whereas Amplicor missed more disease related to nontargeted types, hc2 was more likely to miss disease related to targeted types. (Cancer Epidemiol Biomarkers Prev 2009;18(5):1341-9)