Assessing sex-differences and the effect of timing of vaccination on immunogenicity, reactogenicity and efficacy of vaccines in young children: study protocol for an individual participant data meta-analysis of randomised controlled trials

INTRODUCTION: Disease incidence differs between males and females for some infectious or inflammatory diseases. Sex-differences in immune responses to some vaccines have also been observed, mostly to viral vaccines in adults. Little evidence is available on whether sex-differences occur in response to immunisation in infancy even though this is the age group in which most vaccines are administered. Factors other than sex, such as timing or coadministration of other vaccines, can also influence the immune response to vaccination. METHODS AND ANALYSIS: Individual participant data meta-analysis of randomised controlled trials of vaccines in healthy infants and young children will be conducted. Fully anonymised data from ∼170 randomised controlled trials of vaccines for diphtheria, tetanus, Bordetella pertussis, polio, Haemophilus influenzae type B, hepatitis B, Streptococcus pneumoniae, Neisseria meningitidis, measles, mumps, rubella, varicella and rotavirus will be combined for analysis. Outcomes include measures of immunogenicity (immunoglobulins), reactogenicity, safety and disease-specific clinical efficacy. Data from trials of vaccines containing similar components will be combined in hierarchical models and the effect of sex and timing of vaccinations estimated for each outcome separately. ETHICS AND DISSEMINATION: Systematic reviews of published estimates of sex-differences cannot adequately answer questions in this field since such comparisons are never the main purpose of a clinical trial, thus a large degree of reporting bias exists in the published literature. Recent improvements in the widespread availability of individual participant data from randomised controlled trials makes it feasible to conduct extensive individual participant data meta-analyses which were previously impossible, thereby reducing the effect of publication or reporting bias on the understanding of the infant immune response. Preliminary results will be available in 2016 with final results available in 2019. No ethics review is required for secondary analyses of anonymised data.