The impact of the distribution of human papillomavirus types and associated high-risk lesions in a colposcopy population for monitoring vaccine efficacy

Context.-Impact studies of the new human papillomavirus (HPV) vaccines will be biased unless local baseline distribution studies are conducted. Vaccine cross protection for other important oncogenic HPV types and the emergence of potential genotype replacements require the knowledge of the prevaccine epidemiology of HPV. Objective.-To determine the prevaccine distribution of HPV types in Saskatchewan, using a subpopulation of women referred to a colposcopy clinic. Design.-One thousand three hundred fifty-five specimens obtained during colposcopic examination were typed for HPV using L1 or El gene polymerase chain reaction and direct sequencing. HPV-16 and HPV-31 infections were confirmed with real-time E6 polymerase chain reac reaction. Indeterminate samples were analyzed using Luminex technology. Correlations of the HPV type and histology were examined for statistical significance. Results.-The most commonly identified genotype in patients with cervical intraepithelial neoplasia grade 2 or worse was HPV-16 (46.7%) followed by HPV-31 (14.7%) and then HPV-18 (3.9%). Fifteen of 330 specimens that were positive for HPV-16 or HPV-31 were further resolved to be mixed HPV-16/HPV-31 infections by real-time polymerase chain reaction. The risk of cervical intraepithelial neoplasia associated with HPV-18 infection (0.4-1.7) is substantially lower than with either HPV-16 (3.6-11.0) or HPV-31 (1.8-12.6). Conclusions.-HPV-31 is contributing significantly to the proportion of women with cervical intraepithelial neoplasia in our population and shows a higher prevalence than HPV-18 in high-grade lesions. The clinical significance of HPV-31 may be underestimated and its continued significance will depend on the level of cross protection offered by the new vaccines.