Persistence of an Incident Human Papillomavirus Infection and Timing of Cervical Lesions in Previously Unexposed Young Women

Background: We analyzed data from a cohort of 553 women enrolled in the placebo arm of a randomized controlled trial of the human papillomavirus (HPV) 16/18 vaccine to study the timing of the occurrence of squamous intraepithelial lesions (SIL) or cervical intraepithelial neoplasia (CIN) following incident HPV infection and its relation to persistence of the infection. Methods: At entry, women were cytologically negative, HPV 16/18 seronegative, and high-fisk HPV (HR-HPV) DNA negative. Cervicovaginal samples were initially collected at 3-month and cervical samples at 6-month intervals. We estimated the mean time to SIL/CIN, relative risks of SIL/CIN following incident HPV, and odds ratios between persistent HPV and SIL/CIN. Results: The mean time for SIL/CIN detection was 43.3 [95% confidence interval (95% CI), 36.4-50.1] and 46.4 (95% Cl, 42.0-50.7) months from first infection with HPV 16/18 and other HR-HPVs, respectively. Relative risks of SIL/CIN following incident HPV infection were 66.2 (95% CI, 14.9-295.1) for HPV 16/18 and 50.9 (95% CI, 11.5-225.4) for other HR-HPVs. The odds ratios of SIL/CIN for persistent HPV 16/18 infection, defined as a minimum of two and three (6 monthly) visits, were, respectively, 169.0 (95% CI, 37.2-768.6) and 169.1 (95% CI, 31.5-907.4). The majority of women with cervical infection with HPV 16/18 lasting >6 months (33 of 51, 65%) developed SIL and/or CIN. Conclusions: These analyses provide the first actuarial estimate of mean time between incident HR-HPV infection in previously uninfected women and onset of cervical lesion development. Persistent HR-HPV infection, particularly HPV 16/18, is a strong predictor of cervical lesion risk and potentially a reliable end point for clinical HPV research. (Cancer Epidemiol Biomarkers Prev 2009;18(3):854-62)