Summary High-risk human papillomaviruses (HR-HPV) are the necessary cause of cervical carcinomas. To determine whether HPR-HPV DNA detection in primary routine screening could represent a sensitive and reliable technique for the detection of high-grade squamous intraepithelial lesions (HGSIL), laboratory analysis using 2 cytologic techniques (conventional and liquid-based), HPV testing with Hybrid Capture II assay (HC-II), followed by colposcopic examination of women with abnormal cervical finding and/or persistent HR-HPV infection, was conducted in 7932 women who had routine cervical examination. The sensitivity of HPV testing for detecting a histologically proven HGSIL was 100%, higher than that of conventional (68.1%) and liquid-based (87.8%) cytology. The low specificities of 85.6% and 87.3% of HPV testing slightly increased to 88.4% and 90.1% if HPV testing was reserved for woman >30 years old. The quantitative approach provided by the HC-II assay for the assessment of the viral load was not reliable for predicting HGSIL in normal smears. HR-HPV testing could be proposed in primary screening in association with cytology. With conventional cytology it significantly improves the detection of HGSIL. With the use of the same cervical scrape for HPV testing and liquid-based cytology, HR-HPV testing would allow to select positive samples treated in a second time for cytology which gives a good specificity.

Hybrid Capture II (HC-II) is a commercial human papillomavirus (HPV) detection test designed to detect 18 HPV types divided into high-risk and low-risk groups. We have tested 1647 scrapes from 1518 unselected women attending routine cytological screening by this assay for the detection of histologically proven high-grade lesions. The reliability of this test was also evaluated on 117 fresh cone biopsy samples. HPV DNA has been detected in 400 scrapes (24.3%), 296 containing a high-risk HPV (18.0%). All the smears evocative of high-grade lesions were positive for high-risk HPV, and high-risk HPV were detected in all the 34 cases presenting a histologically proven high-grade lesion and in 68 (97.1%) of the 70 cone biopsy samples showing a high-grade lesion or an invasive carcinoma. Thus, the sensitivity was superior to the sensitivity of cytology (85.3%). Nevertheless, the quantitative approach provided by the HC-II assay for the assessment of the viral load could not clearly distinguish among cases with or without high-grade lesions. Thus this assay is recommended for the screening of high-grade lesions on a large scale, in association with classic cytology. © 1999 Cancer Research Campaign

No abstract

No abstract