HUMAN PAPILLOMAVIRUS TESTING IN PRIMARY CERVICAL SCREENING

Several studies have examined the role of tests for human papillomavir us (HPV) in screening for cervical cancer but as yet the relevance is unclear. We looked at HPV testing for types 16, 18, 31, and 33 on mate rial taken at the time of a cervical smear in 2009 eligible women havi ng routine screening. Women with any degree of dyskaryosis or high lev els of one of these HPV types were referred for colposcopy. 44% of the cervical intraepithelial neoplasia (CIN) lesions of grade 2/3 detecte d had negative cytology and were found only by HPV testing. A further 22% of the CIN 2/3 lesions were positive for HPV but showed only borde rline or mild cytological changes. The positive predictive value of HP V testing was 42%, which was similar to that for moderate dyskaryosis. HPV types 16 and 31 were more sensitive and specific for CIN 2/3 than were types 18 or 33. However, 25% of the CIN 2/3 lesions were not det ected by these four HPV tests. We suggest that HPV testing could usefu lly augment but not replace conventional cytology. These results shoul d stimulate a much larger randomised trial to assess the impact of the se improved CIN 2/3 detection rates on the subsequent incidence of inv asive cancer.