INTERACTION BETWEEN MENINGOCOCCAL SEROGROUP B VACCINE (BEXSERO™) AND ROUTINE INFANT VACCINES ON THE RISK OF OCCURRENCE AND RECURRENCE OF ADVERSE EVENTS FOLLOWING IMMUNIZATION.

Proposal
1224

Title of Proposed Research

Lead Researcher

Gaston De Serres

Affiliation

Institut National de Santé Publique du Québec(INSPQ)andLaval University

Funding Source

Joseline Zafack will do this work as part of her PhD in Epidemiology for which she receives a scholarship of the CIRN network (Canadian Immunization Research Network) funded by the Canadian Institute of Health Research.

Potential Conflicts of Interest

Gaston De Serres received research grant from GSK and was reimbursed travel expenses to travel to an ad hoc advisory board meeting by GSK but received no honoraria.

Data Sharing Agreement Date

4 Feb 2016

Lay Summary

In 2014, Bexsero™ a four component serogroup B meningococcal vaccine (4CMenB) has been licensed in Europe, Canada, and Australia. It has also been used as an Investigational New Drug for outbreak control in the United States. The vaccination schedule of infants includes three visits at 2, 4 and 6 months of age for primary immunization and a booster dose at 12 months. To avoid additional visits, the doses of 4CMenB would ideally be administered concomitantly with routine vaccines. In infants under one year of age, prelicensure studies and post marketing surveillance data showed that 4CMenB has a high reactogenicity [1-4]. Pain at 4CMenB injection site occurs in 58-87% of vaccinees [1-3]. Fever (>=38°C) occurs in 71% in those who received 4CMenB alternatively with routine vaccines but in 76-80% of children who received 4CMenB concomitantly with routine vaccines [1,2]. While this suggests that coadministration causes an interaction resulting in a greater risk of adverse events following Immunization (AEFI) only the AEFI after the 4CMenB dose and not those occurring after routine vaccine immunizations were reported, underestimating the total risk associated with separate visits. For financial and practical reasons, coadministration of infant vaccines is preferred to separate visits. Separate visits may however be preferred if the sum of the AEFI risk at each visit is significantly smaller than the risk with coadministration and/or if the AEFI has a lesser severity.
For children who suffered an AEFI with a dose of 4CMenB, parents and providers are worried about the risk of recurrence and increased severity of the AEFI with administration of the next dose(s) of vaccine. This concern may lead to delay or cessation of immunization which can leave the child with suboptimal vaccine protection. While the 4CMenB clinical trials report the frequency of AEFI, they did not report their risk of recurrence and factors affecting this risk.
Given the high reactogenicity of Bexsero™ it is important to recalculate the risk of occurrence and severity of AEFI with the coadministration of Bexsero™ and routine vaccines compared to separate injections to assess the interaction occurring with the co-administration. We also need to estimate the risk of recurrence of AEFI at subsequent immunizations with the 4CMenB and assess if this risk varies with separate or coadministration with routine vaccines. This information will guide vaccine providers and public health authorities in both decisions regarding the regular vaccination program and for the management of safety concerns that will arise in patients presenting adverse events following 4CMenB immunization. To evaluate this, we propose to perform a secondary analysis of existing databases of three randomized trials (RCT)conducted in infants and toddlers by Novartis. The results of this study will be part of a PhD thesis in epidemiology and a manuscript will be submitted to a peer reviewed journal.

Study Data Provided

[{ "PostingID": 3726, "Title": "GSK-V72P13", "Description": "A Phase 3, Partially Blinded, Randomized, Multi-Center, Controlled Study to Evaluate Immunogenicity, Safety and Lot to Lot Consistency of Novartis Meningococcal B Recombinant Vaccine When Administered With Routine Infant Vaccinations to Healthy Infants" },{ "PostingID": 3727, "Title": "GSK-V72P13E1", "Description": "A Phase 3, Open Label, Multi-Center, Extension Study to Evaluate the Safety, Tolerability and Immunogenicity of Novartis Meningococcal B Recombinant Vaccine When Administered as a Booster at 12 Months of Age or as a Two-dose Catch-up to Healthy Toddlers Who Participated in Study V72P13" },{ "PostingID": 3728, "Title": "GSK-V72P12", "Description": "A Phase 2b, Open Label, Randomized, Parallel-Group, Multi-Center Study to Evaluate the Safety, Tolerability and Immunogenicity of Novartis Meningococcal B Recombinant Vaccine When Administered With or Without Routine Infant Vaccinations to Healthy Infants According to Different Immunization Schedules." }]

Statistical Analysis Plan

We will perform analyses including all vaccine doses and analyses stratified by the number of 4CMenB dose. Missing data will be deleted if they are categorical or represent less than 5% of the database. For continuous variables, multiple imputation method will be used if missing information is random. Data analyses will be performed using the SAS software available on the Data access system.A covariable will be kept in the final adjusted model if it triggers a change of more than 10% on the effect measured. Statistical testing will be bilateral with statistical significance at p<0.05.a) Analysis of the interaction between Bexsero™ and routine vaccines The assessment of interaction between 4CMenB and routine vaccines will be done on the additive scale. R01 will represent the risk of a given adverse event in children who received 4CMenB alone, R10 the risk with routine vaccines alone, R11 the risk with the coadministration of 4CMenB and routine vaccines and R00 is the baseline risk (no vaccine administered). Our measure of risk will be the incidence rate of the primary outcome and our measure of effect will be the difference between incidence rates (i.e incidence rate of the primary outcome in the strata [R11, R01 or R11] minus the baseline incidence rate R00). There is interaction between 4cMenB and routine vaccines when the observed effect following the coadministration of these vaccines differs from the addition of each product's individual effect. That is :(R11 - R00) is not equal to ([R01 - R00] + [R10 - R00])If (R11 - R00) > ([R01 - R00] + [R10 - R00]) then the interaction increases the risk If (R11 - R00) < ([R01 - R00] + [R10 - R00]) then the interaction decreases the riskWe will estimate the incidence rate of AEFI in each strata of vaccine administration: 4CMenB alone (R10), routine vaccines alone (R01) and coadministration of 4CMenB and routine vaccines (R11) during their respective at-risk period. For new onset systemic reactions (fever and systemic reactions other than fever), the at-risk period for 4CMenB and inactivated routine vaccines will be the first 24 hours following their administration and the control period will be from day 4 to 7, whereas for MMRV immunization the at-risk period will be from day 5 to day 13 following immunization and the control period will be from day 0 to day 3. The baseline risk of systemic reactions will be calculated as the incidence of these reactions during control periods. On their part, all injection site reactions following immunization will be considered vaccine-related regardless of their delay of onset and their baseline risk will be considered null. The presence of interaction on AEFI severity will be evaluated by an analysis of the severity in the various strata.b) Analysis of the risk of recurrenceWe will evaluate the risk of recurrence of AEFI attributable to immunization by calculating the difference between incidence rates of recurrences during the at risk and control periods. Subgroups analyses according to sex, age, randomized controlled trial, study design (open label or observer blinded), AEFI severity, antipyretic prophylaxis and antibody level will be performed. We will compare the severities of initial AEFI and recurrence. Comparison of categorical variables will be performed using Chi square or Fisher exact tests. For continuous variables , we will compare the value of the outcome at first occurrence to that at recurrence with a Student T test or a Mann Withney test. To determine if a history of AEFI significantly increases the risk of AEFI at subsequent immunizations, we will compare the incidence of the primary outcomes in children who had and those who did not have these outcomes at previous doses. We will evaluate the confounding and modifying effects of sex, age, randomized controlled trial, study design (open label or observer blinded), number of recurrences, AEFI severity, antipyretic prophylaxis and antibody titers. To estimate our risk differences, we will fit a binomial model using SAS PROC GENMOD with link= identity[5]. Goodness of fit of the model will be evaluated using the deviance and the Pearson statistics. In sensitivity analyses, the at-risk period for 4CMenB and inactivated routine vaccines will be extended to 48 hours following their administration and for MMRV the at-risk period will be extended to 5 to 19 days following its administration.

Publication Citation

http://dx.doi.org/10.1136/bmjopen-2018-026953