Register-based evaluation of HPV vaccination programs

Abstract: The studies included in this thesis examined the population-level effect of human Papillomavirus (HPV) vaccination on incidence of HPV-related disease outcomes, and participation to cervical screening following vaccination. In study I, a cohort of young Swedish girls ages 10 to 24 was followed for HPV vaccination and condyloma to investigate the effect of vaccination on condyloma by vaccine dose. The results showed greatest protection against condyloma following administration of three HPV vaccine doses. Considerable protection against condyloma was also seen after vaccination with two doses of the quadrivalent HPV (qHPV) vaccine. Risk reductions of 71% and 82% following vaccination with two and three doses, respectively, were seen when vaccination was initiated prior age 17. Greater protection against condyloma was seen in those younger at vaccination initiation. In study II, a birth cohort of women at cervical screening ages (born between 1977 and 1987) was followed for HPV vaccination, invitation to cervical screening, and attendance to screening. The results showed that, compared to unvaccinated women, women HPV vaccinated through opportunistic vaccination were equally likely, if not more likely to attend organized cervical screening following an invitation letter to cervical screening. In study III, a cohort of young Swedish girls and women at ages 23 to 29 was followed for HPV vaccination and cervical lesions, i.e. cervical intraepithelial neoplasia (CIN) stage 2 or worse (CIN2+) and CIN stage 3 or worse (CIN3+), to investigate the effect of vaccination on incidence of cervical lesions after three-dose vaccination. The results showed reductions in risk for CIN2+ and CIN3+ following vaccination in girls and young women that initiated vaccination up until age 29. Greater reductions in risk for cervical lesions were seen in those younger at vaccination initiation. A maximum reduction of 75% and 84% in risk for CIN2+ and CIN3+, respectively, were seen when vaccination was initiated prior to age 17. In study IV, we assessed the incidence of condyloma following the introduction of qHPV in Sweden. During the study period, girls were mainly vaccinated via opportunistic HPV vaccination, and vaccination coverage remained rather low. Declines in condyloma incidence in girls below age 20 were seen following the introduction of qHPV vaccination in Sweden and confirm anticipated effects. In addition, we observed declines in incidence of condyloma among men and in women age 20 and above indicates possible herd protection. In conclusion, the results of these studies contribute to the existing evidence on the population level effect of HPV vaccination and the disease monitoring of HPV related disease in an era where opportunistic HPV vaccination was available. The results of study I have also contributed to the discussions on reduced vaccine dosing schedules. Future monitoring of the disease burden over time, as well as observational studies comparing vaccinated and unvaccinated individuals, are necessary to evaluate whether the organized school-based vaccination program has the anticipated effect in the population.