Infection with human papillomavirus (HPV) has been detected in 95 to 100% of cervical cancers, the second most common female cancer worldwide (30), and persistent infection with high-risk HPV (HR-HPV) types is known to be the first step in the process of carcinogenesis. Several epidemiologic and molecular studies suggest the use of HPV testing in order to improve the efficacy of population-based screening programs for cervical cancer. HPV testing has been found to be useful for triaging minor cytological abnormalities and in the follow-up of treated cervical intraepithelial neoplasia (CIN) (2,4,7,27). HR-HPV DNA testing has been shown to be more sensitive than cytology and may improve patient management when used in addition to cytology (3, 4, 11); such testing is also under investigation as a primary screening tool (1, 10, 26). The negative predictive value of HR-HPV testing is very high (99.0%), which may allow screening intervals to be increased for women found to be negative by both cytology and HPV testing (16).The Digene Hybrid Capture 2 assay (HC2) (Digene, Gaithersburg, MD) is commonly used for HPV testing to detect 13 common HR-HPV types (types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68). Recently, the Roche AMPLICOR HPV test (AMP) (Roche Molecular Systems, CA) was made available for the same purpose. AMP detects the same 13 HR-HPV types detected by HC2 and, moreover, uses amplification of the -globin gene as an internal measure of sample integrity and adequacy. Since it is a PCR-based method, the assay can be performed on small aliquots of samples in liquid cytology media or on samples obtained from archival paraffin-embedded tissue, and it is extremely sensitive (detects Ͻ100 copies of HPV DNA/PCR). In addition, incorporation of AmpErase (uracil-N-glycosylase [UNG]) into the master mix allows selective destruction of carryover products (containing deoxyuridine) from previous amplification reactions. However, evaluation of both analytical and clinical sensitivity is needed to support the use of AMP in clinical practice and cervical cancer screening programs.In the present study, we assessed the prevalence of HR-HPV in samples from subjects with cytological abnormalities (atypical squamous cells of undetermined significance [ASC-US], atypical glandular cells of undetermined significance [AGC-US], low-grade squamous intraepithelial lesions [LSIL], and high-grade squamous intraepithelial lesions [HSIL]) as well as in samples from women with normal cytology; we also evaluated the analytical agreement between AMP and HC2 in detecting HR-HPV, and for analytical determinations, genotyping was carried out on discordant samples. In addition, the clinical sensitivity and specificity of AMP for histologically * Corresponding author. Mailing address: Centro per lo Studio e la Prevenzione