Authors’ reply to Fiala

The findings in table 3 are not unexpected.1 2 As Brown and colleagues reported,3 and we stated in our paper,2 the FUTURE I and II clinical trials showed that vaccination reduced the incidence of cervical intraepithelial neoplasia grade 3 or worse associated with 10 non-vaccine human papillomavirus (HPV) types by 32.5% (95% CI 6.0% to 51.9%). The analyses in table 3 suggest that this is due to protection against HPV types that are phylogenetically related to the quadrivalent vaccine types.

As we also stated,2 another study has shown that the quadrivalent vaccine generates an anamnestic response in women aged 15-26 who are seropositive before vaccination,4 5 and that the vaccine prevents reinfection or reactivation of disease that is related to vaccine HPV types.6

Only two women in the vaccine arm developed cervical disease related to the quadrivalent vaccine HPV types after surgery, compared with nine in the placebo arm, which corresponds to a vaccine efficacy of 79%. Six of these nine women in the placebo arm had a vaccine HPV type that was not detected in the original surgical specimen, which suggests that the post-surgery lesion was caused by a new infection and not residual disease. Four of these six women also reported at least one new partner during the interval between the first surgery and the subsequent diagnosis of disease.

For these reasons, we conclude that the observed efficacy is related to vaccination. Although the exact underlying mechanism is not fully understood, these reductions are thought to be clinically significant.