AbstractsWomens Studies

Testing for serum antibodies to HPV 6 and 16 was performed with an ELISA using recombinant HPV capsids. These assays were applied to 3,533 sera taken from a cohort of women followed for up to 12 years at the Seattle-King County STD clinic. The HPV serum antibody data were matched with the demographic, clinical, and other laboratory data collected from the 1,212 women enrolled in the study cohort. This study was undertaken to test the ability of HPV serum antibodies to predict the development of biopsy-confirmed CIN 2-3 in a subset of the study women (n = 528), and to also describe the association between HPV 6 serum antibodies and the history or presence of genital warts in a separate subset of the study women (n = 731).HPV 16 and HPV 6 seropositivity during the 6-month period preceding and including a study visit were both associated with an increase in the risk of biopsy-confirmed CIN 2-3. However, only HPV 6 seropositivity was associated with a statistically significant increase in risk. In a multivariate model, the relative risk (RR) associated with HPV 16 serum antibody positivity was 1.5 (95% confidence interval (CI) 0.8, 3.0), and that associated with HPV 6 serum antibody positivity was 2.4 (95% CI 1.2, 4.7). Other covariates associated with an increased risk of biopsy-confirmed CIN 2-3 included HPV DNA positivity, a history of low-grade squamous intraepithelial lesions (LSIL), and the diagnosis of cervicitis during follow-up.The 328 women with a history of or evidence of genital warts at enrollment were more likely to be HPV 6 seropositive at enrollment than were the 403 women without genital warts at enrollment ($\chi\sp2$ 21.7, P $<$ 0.001). Using a generalized estimating equation (GEE) model, recurrent warts were found to be more likely among women who were younger at their first reported episode of genital warts, were observed to be temporally clustered, and were less likely to occur after HPV 6 seroconversion.These results extend our understanding of the natural history of the immune response to genital HPV infection, but also demonstrate that the relationship between HPV biomarkers and clinical disease is one marked by complexity.