Background

The AS04-adjuvanted bivalent L1 virus-like-particle (VLP) vaccine (Cervarix™) against infection with human papillomavirus (HPV) types 16/18 holds great promise to prevent HPV16/18 infections and associated neoplasias, but it is important to rule out significant co-factors of the neoplasias like smoking. Methods

We conducted a pilot study to compare the quantity and quality of HPV16/18 antibody response at baseline and 7 months post vaccination in 104 non-smoking and 112 smoking female participants vaccinated at 0, 1 and 6 months with Cervarix™ (55 and 48 study participants) or with Hepatitis A vaccine (HAVRIX™) (48 and 64 participants, respectively). These 216 women were a sub-sample of 4808 baseline 16- to 17-year old Finnish women initially enrolled in the double-blind, randomized controlled phase III PATRICIA trial. Following end-of-study unblinding in 2009 they were randomly chosen out of all the participants of the three major Finnish PATRICIA study sites in the Helsinki metropolitan area (University of Helsinki, N = 535, and Family Federation Finland, N = 432) and Tampere (University of Tampere, N = 428). Following enrolment, serum samples were collected at month 0 and month 7 post 1st vaccination shot, and were analysed for levels and avidity of IgG antibodies to HPV16 and HPV18 using standard and modified (4 M urea elution) VLP ELISAs. Results

We found that at month 7 post vaccination women who smoked (cotinine level > 20 ng/ml) had levels of anti-HPV16/18 antibodies comparable to those of non-smoking women. Low-avidity HPV16/18 IgG antibodies were observed in 16% of the vaccinated women, and active smoking conferred a three-fold increased risk (95% CI 1.0-9.3) of having the low-avidity antibodies. Conclusion

Our data suggest that while smoking does not interfere with the quantity of vaccine-induced peak IgG levels, it may affect the avidity of IgG induced by HPV16/18 vaccination.