Objectives Human papillomavirus (HPV) testing is an important part of cervical cancer screening and management of women with atypical screening results. This study was conducted to evaluate the analytical and clinical performance of the Abbott RealTime High-Risk HPV assay (RealTime) in a referral population, in comparison to the Qiagen Hybrid Capture 2 High-Risk HPV DNA Test (hc2).

Methods RealTime is a new polymerase chain reaction assay that detects 14 high-risk HPV genotypes with simultaneous differentiation between HPV 16 and HPV 18. Five hundred forty-five routine cervical smear samples (ThinPrep) from women who were referred to 2 German colposcopy clinics were included in the study. All samples were tested with both assays for the detection of high-risk HPV DNA. Specimens with repeatedly discordant results were genotyped by Linear Array (Roche) and in-house polymerase chain reaction assays.

Results Both assays showed excellent overall agreement (92.8%; κ = 0.86) on 545 samples. Analytical sensitivity of RealTime was comparable to that of hc2 (97.6% vs 95.1%, P = 0.189), whereas RealTime demonstrated significantly higher analytical specificity compared with hc2 (100% vs 93.1%, P < 0.0001). RealTime showed no cross-reactivity with untargeted HPV genotypes in this study. The clinical performance of the assays was evaluated based on histology results available from 319 women (90 nonpathological, 73 cervical intraepithelial neoplasia [CIN] 1, 75 CIN 2, 74 CIN 3, and 7 invasive cancers). High-risk HPV detection rates observed in women with CIN 1, CIN 2+, and CIN 3+ diagnosis, respectively, were comparable for both assays: 47.9%, 92.3%, and 97.5% (RealTime) and 47.9%, 92.3%, and 93.8% (hc2). Detection of HPV 16/18 with RealTime was highly correlated with severity of dysplasia: less than CIN 2, 30.5%; CIN 2+, 59.0%; CIN 3+, 71.6%.

Conclusions These results support the use of RealTime for routine detection of HPV infections in a referral population.