The natural history of human papillomavirus infection as measured by repeated DNA testing in adolescent and young women

Objectives: The objectives of this study were to describe the early natural history of human papillomavirus (HPV) infection by examining a cohort of young women positive for an HPV test and to define within this cohort (1) the probability of HPV regression, (2) the risk of having a squamous intraepithelial lesion, and (3) factors that were associated with HPV regression. Study design: The study was a cohort analytic design. An inception cohort of 618 women positive for HPV participated. HPV testing, cytologic evaluation, and colposcopic evaluation were performed at 4-month intervals. HPV testing was characterized for two groups: low risk (five types rarely associated with cancers) and high risk (nine types most commonly associated with cancers). Results: Estimates provided by Kaplan-Meier curves showed that ~70% of women were found to have HPV regression by 24 months. Women with low-risk HPV type infections were more likely to show HPV regression than were women with high-risk HPV type infections (log rank test p = 0.002). The relative risk for the development of high-grade squamous intraepithelial lesion (HSIL) was 14.1 (95% confidence interval: 2.3, 84.5) for women with at least three positive tests for high-risk HPV preceding the development of the HSIL compared with that for women with negative tests for high-risk HPV. However, 88% of women with persistent positive HPV tests have not had HSIL to date. No factors associated with high-risk HPV type regression were identified except for a negative association with an incident history of vulvar condyloma (relative risk = 0.5 [95% confidence interval: 0.3 to 0.8]). Conclusion: Most young women with a positive HPV test will become negative within a 24-month period. Persistent positive tests with oncogenic HPV types represented a significant risk for the development of HSIL. However, we found that most young women with persistent positive HPV tests did not have cytologically perceptible HSIL over a 2-year-period. Factors thought to be associated with the development of HSIL were found not to be important in HPV regression.