A novel methodology based on serological data for improving precision in the influenza vaccine response of non-seroprotected

Vaccine immunogenicity trial is a widely accepted approach for estimating effects of influenza vaccine campaign since it allows reducing timing and costs with respect to clinical trials. When evaluating the influenza vaccine immunogenicity in elderly, the geometric mean of post-vaccination titers (post-GMT) of seronegatives, a population for which the pre-vaccination titer is unique and known a-priori, is the statistic that the Committee for Medical Products for Human Use (CHMP) recommends to report together with its measurement uncertainty. Since in elderly population the seronegatives represent a small fraction, this low number affects the precision of the estimate. Thus, statistical approaches that reduce significantly the corresponding measurement uncertainty are adopted. Besides seronegatives, even the immunological response of non-seroprotected people is significant for characterizing a vaccine. However, this sub-population includes people with more than one single pre-vaccination state, and thus the statistical methodology usually applied to seronegatives for improving the estimate precision cannot be applied automatically. In this paper, a novel approach is described for improving the precision of post-GMT estimate from serological data of non-seroprotected elderly, a population for which the influenza vaccination is strongly recommended by CHMP. The precision of the proposed methodology is characterized in terms of type-A uncertainty on elderly Umbrian (Italian) population and results confirm its effectiveness. Due to the increased precision of the obtained post-GMT estimate, also the quality of comparative studies, that otherwise could be highly affected by the choice of evaluating vaccine effects based on serological parameters, improves with respect to the use of standard methods indicated in the scientific literature.