Approximate age, 5 to 7 months. A, Nonpneumococcal vaccine antigens, B, Pneumococcal serotypes. For age at first vaccination, geometric mean ratios (GMRs) greater than 1.0 indicate that children who are older at the time of their first vaccination have higher antibody concentrations after the third dose. For maternal antibodies, GMRs greater than 1.0 indicate that children with higher antibody concentrations prior to vaccination have higher antibody concentrations after the third dose. Data are displayed in eTable 4 in the Supplement for model 1, which includes maternal antibody in the model as a continuous variable (see Statistical Analysis subsection of Methods for definition of model 1). FHA indicates filamentous hemagglutinin; PRN, pertactin; PRP, polyribosylribitol phosphate; and PT, pertussis toxoid.
Effect on antibody concentrations after the third dose (approximate age, 5-7 months) with 2-, 3-, and 4-month schedules and 2-, 4- and, 6-month schedules. Geometric mean ratios (GMRs) less than 1.0 indicate that children with higher antibody concentrations prior to vaccination have lower antibody concentrations after the third dose. Data are from model 1, which includes maternal antibody in the model as a continuous variable and was unadjusted for the schedule of administration (see Statistical Analysis subsection of Methods for definition of model 1). FHA indicates filamentous hemagglutinin; PRN, pertactin; PRP, polyribosylribitol phosphate; and PT, pertussis toxoid. 4,6B, 9V, 14, 18C, 19F, 23F are serotypes of pneumococcus.
For maternal antibodies, geometric mean ratios (GMRs) less than 1.0 indicate that children with higher antibody concentrations prior to vaccination have lower antibody concentrations after the booster dose. For age effects, GMRs greater than 1.0 indicate higher antibody concentrations after boosters in those who were older at the time of priming or booster vaccination. Data are from model 1, which includes maternal antibody in the model as a continuous variable and is displayed in eTable 7 in the Supplement (see Statistical Analysis subsection of Methods for definition of model 1). FHA indicates filamentous hemagglutinin; PRN, pertactin; PRP, polyribosylribitol phosphate; and PT, pertussis toxoid.
For maternal antibodies, geometric mean ratios (GMRs) less than 1.0 indicate that children with higher antibody concentrations prior to vaccination have lower antibody concentrations after the booster dose. For age effects, GMRs greater than 1.0 indicate higher antibody concentrations after boosters in those who were older at the time of priming or booster vaccination. Data are from model 1, which includes maternal antibody in the model as a continuous variable and is displayed in eTable 7 in the Supplement (see Statistical Analysis subsection of Methods for definition of model 1).
A delay in first immunization results in a beneficial effect that is the combination of less maternal antibody due to antibody decay, and older age when first immunized. Solid lines indicate estimates based on maternal antibody half-lives as detailed in eMethods. Dotted lines indicate sensitivity analyses using half-lives of 7 days longer and 7 days shorter. A 5-fold increase in infant levels of maternal pertussis antibodies due to a prenatal immunization program could be offset with a 5.04-week delay in first immunization assuming pertussis antibodies decay with a half-life of 36 days (solid dark blue line). Using a range of half-life estimates of 29 to 43 days gives a range of 4.6- to 5.4-weeks’ delay required (dotted dark blue line). Details of calculations are provided in the eMethods in the Supplement.
eMethods. Secondary and Subgroup Analyses
eTable 1. Laboratory Assays
eTable 2. List of Included Studies
eTable 3. Antibody Concentration at One-Month Post Priming, in Those Who Did and Did Not Have Detectable Levels of Maternal Antibody Prior to Immunization
eTable 4. Influence of Age at First Vaccination and Pre-Existing Antibody Concentration Prior to Vaccination, on Antibody Concentration Post-3rd Priming Dose (Age 5 ~ 7 Months)
eTable 5. Influence of Pre-Existing Pneumococcal, Diphtheria And Tetanus Antibody Concentrations Prior to Vaccination, on Pneumococcal Antibody Concentration Post-3rd Priming Dose (Age ~ 5-7 Months)
eTable 6. Influence of Age at First Vaccination, Pre-Existing Antibody at First Vaccination, and Age at Booster, on Persistence of Antibody to Pre-Booster Time Point (Age ~ 12-24 Months)
eTable 7. Influence of Age at First Vaccination, Age at Booster Vaccination, and Pre-Existing Antibody at First Vaccination, on Antibody Concentrations Post-Booster (Age ~ 12-24 Months)
eFigure 1. Influence of Age at First Vaccination and Pre-Existing Antibody Concentration Prior to Vaccination, on Antibody Concentration Post-3rd Priming Dose (Age ~ 5 − 7 Months)
eFigure 2. Influence of Age at First Vaccination, Pre-Existing Antibody at First Vaccination, and Age at Booster, on Persistence of Antibody to Pre-Booster Time Point (Age ~ 12-24 Months)
Customize your JAMA Network experience by selecting one or more topics from the list below.
Voysey M, Kelly DF, Fanshawe TR, et al. The Influence of Maternally Derived Antibody and Infant Age at Vaccination on Infant Vaccine Responses : An Individual Participant Meta-analysis . JAMA Pediatr. 2017;171(7):637–646. doi:10.1001/jamapediatrics.2017.0638
What is the potential for and extent of maternal antibody interference in infant vaccine responses?
In this individual participant data meta-analysis of the serum of 7630 infants, maternal antibody concentrations and the infant's age at first vaccination both influenced infant vaccine responses. These effects are seen for almost all antigens contained in global immunization programs, are not removed by more widely spaced schedules, and influence immune response for some vaccines even at the age of 24 months.
Prenatal immunization programs containing multicomponent vaccines have the potential to interfere with current immunization programs; however, a delayed start to infant immunization may mitigate these inhibitory effects.
The design of infant immunization schedules requires an understanding of the factors that determine the immune response to each vaccine antigen.
Deidentified individual participant data from GlaxoSmithKline clinical trials were obtained through Clinical Study Data Request. The data were requested on January 2, 2015, and final data were received on April 11, 2016.
Immunogenicity trials of licensed or unlicensed vaccines administered to infants were included if antibody concentrations in infants were measured prior to the first dose of vaccine.
Data Extraction and Synthesis
The database was examined; studies that appeared to have appropriate data were reviewed.
Main Outcomes and Measures
Antigen-specific antibody concentration measured 1 month after priming vaccine doses, before booster vaccination, and 1 month after booster vaccine doses.
A total of 7630 infants from 32 studies in 17 countries were included. Mean (SD) age at baseline was 9.0 (2.3) weeks; 3906 (51.2%) were boys. Preexisting maternal antibody inhibited infant antibody responses to priming doses for 20 of 21 antigens. The largest effects were observed for inactivated polio vaccine, where 2-fold higher maternal antibody concentrations resulted in 20% to 28% lower postvaccination antibody concentration (geometric mean ratios [GMRs], type 1: 0.80; 95% CI, 0.78-0.83; type 2: 0.72; 95% CI, 0.69-0.74; type 3: 0.78; 95% CI, 0.75-0.82). For acellular pertussis antigens, 2-fold higher maternal antibody was associated with 11% lower postvaccination antibody for pertussis toxoid (GMR, 0.89; 95% CI, 0.87-0.90) and filamentous hemagglutinin (GMR, 0.89; 95% CI, 0.88-0.90) and 22% lower pertactin antibody (GMR, 0.78; 95% CI, 0.77-0.80). For tetanus and diphtheria, these estimates were 13% (GMR, 0.87; 95% CI, 0.86-0.88) and 24% (GMR, 0.76; 95% CI, 0.74-0.77), respectively. The influence of maternal antibody was still evident in reduced responses to booster doses of acellular pertussis, inactivated polio, and diphtheria vaccines at 12 to 24 months of age. Children who were older when first immunized had higher antibody responses to priming doses for 18 of 21 antigens, after adjusting for the effect of maternal antibody concentrations. The largest effect was seen for polyribosylribitol phosphate antibody, where responses were 71% higher per month (GMR, 1.71; 95% CI, 1.52-1.92).
Conclusions and Relevance
Maternal antibody concentrations and infant age at first vaccination both influence infant vaccine responses. These effects are seen for almost all vaccines contained in global immunization programs and influence immune response for some vaccines even at the age of 24 months. These data highlight the potential for maternal immunization strategies to influence established infant programs.
It is difficult to overstate the global effect of routine infant immunization on reducing morbidity and mortality from infectious diseases. However, the design of immunogenic schedules for routine immunization is increasingly challenging owing to the variety of vaccines in use, the use of prenatal immunization programs for some pathogens, and differences in the epidemiology of infections. A detailed understanding of the variables that influence immunogenicity is therefore of great importance. Two key factors that can affect the immunogenicity of infant vaccines are (1) the residual concentration of maternal placentally transferred antigen-specific antibodies at the time of immunization and (2) the age at which immunizations are given.
Maternal antibodies are transferred to an infant via the placenta during the third trimester of pregnancy and are important in providing protection against infection during the first months of life.1-3 The amount of antibody transferred from mother to infant depends primarily on the gestational age at birth, placental function, and serum antibody concentration in the mother.4,5 Following birth, the concentration of maternally derived antibodies declines over weeks to months while infant immune competence gradually increases.1
High concentrations of maternally acquired antibodies have been shown to inhibit the immune response to infant vaccination against some antigens.4,6-12 However, owing to the small size of most studies, there is inconsistent evidence on the extent of this inhibition and a lack of consensus regarding which antigens or vaccines are affected. In addition, the magnitude of the infant’s antibody response to vaccination is affected by the age of the infant, with older infants generally responding better to vaccines than younger infants.1,13,14 Since transplacentally acquired antibody decays over time, the interference with vaccine responses reduces with age and observed age effects on vaccine responses may therefore be due to reduced maternal antibody interference. Previous studies have not been able to separate the influence of these interrelated components on the infant immune response, leading to the observation that, “multivariate analyses on a large number of infants are required to identify the main determinants of vaccine antibody responses.”15(p32)
Reduced response to infant vaccination is important in the context of prenatal immunization programs. Currently, vaccination with a pertussis-containing vaccine in the second or third trimester of pregnancy is recommended in many countries.16-20 Maternal immunization with diphtheria-tetanus-pertussis-polio vaccine during pregnancy effectively protects infants against pertussis disease in the first few months of life.21,22 However, subsequent diminished responses to infant vaccination may leave children more susceptible to disease in later infancy and increase the transmission rates to unvaccinated cohorts.12
In this investigation, we combined serologic data from multiple randomized clinical trials to create a large cohort of infants vaccinated at various ages under different schedules from whom serum antibody concentrations were measured both before and after vaccination. We conducted a meta-analysis and calculated the effects of preexisting antibody and age at vaccination on antibody responses to priming and booster vaccines. From these results, we investigated to what extent a delay in vaccination would offset the maternal inhibition of infant antibody responses and whether maternal antibody inhibition was less pronounced in schedules with a wider spacing of doses.
Fully deidentified data from clinical trials were available through Clinical Study Data Request (https://clinicalstudydatarequest.com/). Data were obtained as part of a wider series of studies on immune responses to infant vaccination.23 Studies included in the present report are a subset of immunogenicity trials in which antibody concentrations were measured prior to the first dose of vaccine. All trials from which these data were obtained were sponsored by GlaxoSmithKline and assays were conducted in the GlaxoSmithKline central laboratory. The outcomes were antigen-specific IgG concentrations at 1 month after the third priming dose (approximate age, 5-7 months) and both before and 1 month after the booster vaccination.
Antibody concentrations were log2-transformed and analyzed using mixed-effects models for each antigen and time point separately. Infants were included if antibody concentrations were available at both the prevaccination visit (maternal antibody) and the postvaccination time point. Children who received a placebo or no vaccine were excluded. Some studies were conducted in multiple countries. Instead of an analysis grouped at the study level, a country cohort variable was created that was unique to countries within studies. Maternal antibody was included in models as a continuous covariate (model 1) and, for comparison, sensitivity analyses were conducted with maternal antibody as a binary variable (above the lower limit of assay detection or not; model 2). Mixed-effects models contained country cohort–level random intercepts and were adjusted for age at first vaccination (months). Adjustment for additional factors known to influence antibody responses (eg, sex, vaccine received, and the scheduled spacing of doses) were included in models to account for their influence; however, coefficients from these parameters are not presented.
The antilog (2x) of the coefficients from models and their 95% CIs are presented as geometric mean ratios (GMRs). These ratios are interpreted as the relative increase (fold-rise) in antibody response that is associated with a 1-unit change in the given covariate. For age at vaccination, the GMR is the relative increase in antibody response for a 1-month older child. For log2-transformed maternal antibody concentrations, the GMR is the relative increase in antibody response associated with a doubling in maternal antibody concentrations. Further details of statistical analyses are contained in the eMethods in the Supplement. A list of all assays and assay thresholds is included in eTable 1 in the Supplement.
Antibody data were available for 21 antigens from 19 child cohorts (7630 infants) from 32 studies in 17 countries (eTables 2 and 3 in the Supplement). Children from different studies varied in age at the time of enrollment, resulting in a total age range of 5 to 20 weeks, with mean (SD) age at baseline 9.0 (2.3) weeks; 3906 (51.2%) of the infants were boys. Children received vaccines according to 1 of 4 schedules: 6, 10, and 14 weeks; 2, 3, and 4 months; 2, 4, and 6 months, or 3, 4, and 5 months. Studies were conducted in Europe (Spain24-26 France,27 Germany,28,29 Greece,24 Poland27,30,31 Czech Republic,32 Finland27), Africa (Mali and Nigeria),33 Latin America (Chile,34 Nicaragua,35 Argentina,35 Dominican Republic36) East Asia (The Philippines30,37 and Korea38,39), Russia,40 and Australia.41
Preexisting antibody in the infant inhibited the response to a priming series of vaccinations for 20 of the 21 measured antigens, the exception being serotype 7F pneumococcal polysaccharide. A similar result was obtained with preexisting antibody modeled as a binary variable with reduced antibody response seen for 18 of the 21 antigens assessed, the exceptions being antibody responses to serotype 7F and 4 pneumococcal polysaccharides and Haemophilus influenzae type b polysaccharide capsule (polyribosylribitol phosphate [PRP]). The largest effects were observed for inactivated polio vaccine responses, where 2-fold higher preexisting antibody resulted in 20% to 28% lower postvaccination antibody GMRs (type 1: GMR, 0.80; 95% CI, 0.78-0.83; type 2: GMR, 0.72; 95% CI, 0.69-0.74; and type 3: 0.78; 95% CI, 0.75-0.82) (Figure 1 and eTable 4 in the Supplement). When preexisting antibody was modeled as a binary variable, infant seropositivity to polio vaccines was associated with a 36% to 60% decrease (type 1: GMR, 0.57; 95% CI, 0.50-0.65; type 2: GMR, 0.40; 95% CI, 0.35-0.46; and type 3: GMR, 0.64; 95% CI, 0.55-0.73) in postvaccine responses (eFigure 1 and eTable 4 in the Supplement).
For acellular pertussis antigens (pertussis toxoid, filamentous hemagglutinin, and pertactin), 2-fold higher preexisting antibody was associated with 11% lower postvaccination antibody for pertussis toxoid (GMR, 0.89; 95% CI, 0.87-0.90) and filamentous hemagglutinin (GMR, 0,89; 95% CI, 0.88-0.90) and 22% lower pertactin antibody (GMR, 0.78; 95% CI, 0.77-0.80). For tetanus and diphtheria, these estimates were 13% (GMR, 0.87; 95% CI, 0.86-0.88) and 24% (GMR, 0.76; 95% CI, 0.74-0.77), respectively (Figure 1; eTable 4 in the Supplement).
In the binary analysis, infants who were seropositive for pertussis antigens had 23% lower postvaccination pertussis toxoid antibody, 27% lower filamentous hemagglutinin antibody, and 50% lower pertactin antibody. For both tetanus and diphtheria, seropositive infants had 53% lower postvaccine antibodies (eFigure 1 and eTable 4 in the Supplement).
In an analysis of a subset of participants who received pneumococcal conjugate vaccines (PCV7, n = 466 and PCV10, n = 1457), the effect of preexisting antibody to related carrier proteins (diphtheria and tetanus) was assessed simultaneously, while also adjusting for preexisting pneumococcal antibodies. For PCV7 vaccine, conjugated to diphtheria cross-reacting material, preexisting diphtheria antibodies were associated with reduced responses to serotypes 4 (GMR, 0.93; 95% CI, 0.87-0.98), 6B (GMR, 0.88; 95% CI, 0.80-0.98), and 9V (GMR, 0.91; 95% CI, 0.86-0.97) pneumococcal polysaccharides, whereas preexisting tetanus antibodies had no effect. In contrast, in PCV10 vaccine recipients, preexisting diphtheria antibodies were associated only with reduced responses to serotype 19F antibodies (the only serotype conjugated to diphtheria protein) (GMR, 0.84; 95% CI, 0.80-0.89), whereas preexisting tetanus antibodies affected both serotypes 18C and 19F pneumococcal antibodies (18C conjugated to tetanus protein) (GMR, 0.94; 95% CI, 0.91-0.99; and 0.92; 95% CI, 0.88-0.96, respectively) (eTable 5 in the Supplement).
Similarly, in recipients of Haemophilus influenzae type b vaccine conjugated to tetanus toxoid, a small degree of inhibition of PRP responses was associated with preexisting PRP antibodies (GMR, 0.96; 95% CI, 0.93-0.99) and preexisting tetanus antibodies (GMR, 0.97; 95% CI, 0.94-0.996). Preexisting diphtheria antibodies had no influence on PRP responses (GMR, 0.96; 95% CI, 0.94-0.99).
For antibodies against 18 of the 21 antigens studied, children who were older when initially immunized had between 10% and 71% higher postvaccination antibody per additional month of age (Figure 1). These effects remained after adjusting for preexisting antibody. The largest effect was seen for PRP antibody, where responses were 71% higher per month (GMR, 1.71; 95% CI, 1.52-1.92). No effect of age at first vaccination was seen for capsular group C meningococcal antibodies and serotypes 14 and 23F pneumococcal antibodies (Figure 1 and eTable 4 in the Supplement, model 1).
Twelve studies were conducted using a 2-, 4-, and 6-month schedule, and 3 studies used a 2-, 3-, and 4-month schedule. When models were fitted within these subgroups separately, the effect of maternal antibodies in terms of proportionate reduction in postvaccination responses was similar for both schedules (Figure 2).
A subset of 12 trials administered booster vaccinations between ages 12 and 24 months. At prebooster time points, lower responses associated with preexisting maternal antibodies at the time of the first vaccination were still evident for 16 of the 20 antigens. Age at first vaccination also had a significant effect on antibody levels prior to booster vaccination for all antigens except tetanus and serotype 23F pneumococcal antibodies (eFigure 2 and eTable 6 in the Supplement).
The concentration of maternal antibodies prior to the first vaccination was associated with reduced responses to booster doses for diphtheria, pertussis antigens, inactivated polio, and 5 of 10 (50%) serotypes of pneumococcus (Figure 3 and Figure 4; and eTable 7 in the Supplement).
For diphtheria, the pertussis antigens pertussis toxoid and filamentous hemagglutinin, inactivated polio type 3, and 5 of 10 (50%) serotypes of pneumococcus, the age at first vaccination had a persistent positive effect on postbooster antibody levels. The largest differences were observed for diphtheria, pneumococcal serotypes 5 and 18C, and polio type 3, where antibody levels were between 28% and 48% higher for each month of delay in the initial vaccination. In contrast, the age at booster vaccination was associated with higher antibody concentrations only for tetanus and PRP antibodies (9% and 15% increase per month, respectively).
A delay in age at the first immunization resulted in a combined beneficial effect on immunogenicity due to (1) a reduction in maternal antibody interference associated with antibody decay and (2) an increase in infant age. Assuming a half-life of 26 days for pertussis toxoid maternal antibodies, the total benefit for acellular pertussis was a 1.06-fold increase (GMR, 1.06; sensitivity range, 1.05-1.06, using half-lives of 29 to 43 days) in pertussis toxoid antibody concentrations per week of delay in the first infant dose of vaccine.
The number of weeks’ delay required to offset a 2- to 5-fold increase in infant levels of maternal pertussis antibodies due to prenatal immunization was between 2.2 and 5.04 weeks. For diphtheria and tetanus, these estimates were 2.6 to 5.9 and 1.7 to 3.9, respectively (Figure 5).
Our analysis of the serum of more than 7600 infants from 17 countries surmounts the problem of insufficient statistical power inherent in previous attempts to investigate maternal antibody interference with infant vaccine immunogenicity, which has led to inconsistent and/or nonsignificant findings. Our analysis comprehensively models the effects of maternal antibody inhibition and infant age at vaccination on the majority of vaccine antigens contained in current global infant immunization programs and reveals that, for almost all antigens, transplacental antibody inhibits the antibody response to priming vaccinations and these effects are not diminished by administration of a booster dose. These analyses further reveal the benefit of infants being older when first immunized, an association that remains after adjusting for waning maternal antibody levels.
In contrast to previous reports,10 the effects of maternal antibodies and the infant’s age when first immunized are not only seen in response to a priming series of vaccines, but continue to affect antibody responses to booster vaccinations at ages 12 to 24 months for many antigens. This finding suggests the importance of the quality of the immune response to the first dose of antigen, regardless of subsequent doses.
The influence of maternal antibodies on measles vaccine immunogenicity has been widely studied to determine the optimal age for vaccination. Younger infants (approximately 6 months) have higher maternal antibody concentrations and lower antibody responses to measles vaccine compared with older infants (approximately 9-12 months), but also produce lower neutralizing antibody responses in the absence of maternally derived measles antibodies.42 In contrast, there is little consensus in the medical literature regarding the potential for, extent of, or clinical implications of maternal antibody interference with vaccines received in early infancy. O’Brien et al10 reported maternal antibody interference with priming doses of PCV7 in infants, and Englund et al8 reported inhibition for whole cell pertussis vaccine but observed no association between maternal antibody to pertussis toxoid and infant responses to 13 different acellular pertussis vaccines in a combined, yet unadjusted, analysis. Other studies are limited in size or incorporate only historical control methods prone to bias. In a small study of 36 infants, Jones et al43 showed that high levels of maternal antibody inhibited responses to pneumococcal conjugate vaccine as well as to tetanus toxoid, but no effect was seen against acellular pertussis toxoid or H influenzae type b, and other small studies have failed to find significant effects.44-46 Our analysis reveals that, not only are infant responses inhibited by maternal antibody to the same antigen, but responses to conjugate vaccines in infancy are also affected by antibody to related carrier proteins.
Schedules with a wider spacing of doses delivered the second and third doses of vaccine after a longer interval, allowing for maternal antibody decay. In our analyses, substantial inhibition was associated with preexisting antibodies for both the 2-, 4-, and 6-month and 2-, 3-, and 4-month schedules. The interference due to maternal antibodies is thus predominantly mediated through the effect on response to the first dose of vaccine.
In settings where prenatal immunization programs are well established with high coverage, maternal antibodies provide protection against pertussis for most infants in the first few months of life.21 In such settings, a delay in administration of the first vaccination would increase responses to infant vaccination if vaccine coverage is high, and the current schedule already includes an early first dose (≤2 months). We calculated the magnitude of the expected total increase in vaccine antibody due to a delay in vaccination by combining model parameters with the half-life of maternal antibodies. The magnitude of the increase in maternal antibody concentrations in infants due to prenatal immunization programs has not been well established, but estimates from a small randomized trial45 and nonrandomized studies47,48 suggest that, for pertussis, between 2-fold and 5-fold higher antibody concentrations may be induced. Our models show that the inhibitory effect of a 2-fold to 5-fold increase in maternal antibody concentrations in infants can be offset by a delay in vaccination of between 2.2 and 5.0 weeks—time periods that are short enough to make delayed-start vaccination a feasible option. Since current prenatal immunization programs use diphtheria and tetanus–containing vaccines, a delay in immunization may improve responses to other vaccines containing these carrier proteins.
Decisions regarding the timing of infant vaccinations need to consider the length of time during which direct protection afforded by prenatal immunization is maintained in infants, an area in which there are currently few data. Infant immunization programs need to consider the level of disease risk in early infancy and the level of coverage of prenatal programs, as well as balance the complex interaction between the direct protection afforded the infant early in life by maternal vaccination, against the increased risk of disease later in infancy as a result of reduced immunogenicity, and the increase risk of disease in the population due to reduced indirect effects. This tradeoff is minimized if the vaccine schedule can be improved to increase infant responses to vaccination.
The studies included herein were conducted in the absence of prenatal vaccination programs. The effects of antibodies induced by natural exposure and those induced by vaccination during pregnancy may differ. In addition, preexisting antibodies in the infant may be due to exposure of the infant to circulating pathogens, such as pneumococcal or pertussis antigens, in the first few months of life rather than being entirely due to maternal antibody transmission. These models may, therefore, underestimate the effect of maternal antibody interference for widely circulating pathogens.
Older infants mount greater antibody responses to vaccination and maternal transplacental antibodies inhibit infant responses. These effects are observed for both priming and booster vaccine doses and occur for almost all antigens present in global immunization programs. Antibody responses to conjugate vaccines are not only affected by interference from preexisting antibodies to the polysaccharide antigen, but also from antibodies associated with carrier proteins. Therefore, prenatal immunization programs containing multicomponent vaccines have the potential to interfere with the immunogenicity of current immunization programs. However, the clinical relevance in terms of potential reductions in disease due to maternal antibody interference on infant vaccine immunogenicity is not known.
Decisions regarding the design of infant immunization schedules are increasingly complex owing to greater numbers of vaccine antigens, prenatal immunization programs, and the addition of booster doses of vaccines throughout childhood and adolescence. These data on key factors that influence the immune response will help in the planning of immunization schedules and provide important factors for evaluating their cost-effectiveness.
Accepted for Publication: February 24, 2017.
Corresponding Author: Merryn Voysey, MSc, Radcliffe Observatory Quarter, Woodstock Road, Oxford OX2 6GG, England (email@example.com).
Published Online: May 15, 2017. doi:10.1001/jamapediatrics.2017.0638
Author Contributions: Mrs Voysey had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Voysey, Pollard.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Voysey.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Voysey.
Obtained funding: Voysey.
Administrative, technical, or material support: All authors.
Supervision: Fanshawe, Perera, Pollard.
Conflict of Interest Disclosures: Dr Sadarangani has received investigator-initiated research grants from Pfizer. Dr Kelly has previously accepted support from vaccine manufacturers to attend scientific meetings. Dr O’Brien is a member of the World Health Organization’s (WHO’s) Strategic Advisory Group of Experts, has received research grants from Pfizer and GlaxoSmithKline in the past 3 years, and has served in a voluntary capacity on expert advisory committees for GlaxoSmithKline and Merck. Dr Pollard has conducted studies on behalf of Oxford University funded by vaccine manufacturers, but currently does not undertake industry-funded clinical trials. Trials of vaccines or observational studies previously funded by Okairos, Novartis, and Pfizer were completed within the past 3 years. His department received unrestricted educational grants from Pfizer, GlaxoSmithKline, and Astra Zeneca in July 2016 for a course on Infection and Immunity in Children. Dr Pollard chairs the UK Department of Health’s (DH’s) Joint Committee on Vaccination and Immunisation (JCVI) and is a member of the WHO’s Strategic Advisory Group of Experts.
Funding/Support: Mrs Voysey is funded by National Institute for Health Research (NIHR) Doctoral Research Fellowship DRF-2015-08-048. Dr Kelly receives salary support from the NIHR Oxford Biomedical Research Centre.
Role of the Funder/Sponsor: The funder had no involvement in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Disclaimer: The views expressed in this manuscript are those of the authors and do not necessarily reflect the views of the JCVI, the DH, the NIHR, the National Health Service, or the WHO.
Additional Information: All data included in these analyses were provided by GlaxoSmithKline through an independent third party (http://clinicalstudydatarequest.com). The company had no access to the analysis results, was not involved in the decision to submit the manuscript, and did not review the manuscript prior to submission.
Create a personal account or sign in to: