Kinetics of the long-term antibody response after meningococcal C vaccination in patients with juvenile idiopathic arthritis: a retrospective cohort study
- Susanne P Stoof1,2,
- Marloes W Heijstek2,
- Karen M Sijssens2,
- Fiona van der Klis1,
- Elisabeth A M Sanders2,
- Peter F M Teunis3,4,
- Nico M Wulffraat2,
- Guy A M Berbers1
- 1Laboratory for Infectious Diseases and Screening, Centre for Infectious Disease Control Netherlands, National Institute of Public Health and the Environment (RIVM), Bilthoven, The Netherlands
- 2Department of Paediatric Immunology, Wilhelmina Children's Hospital, University Medical Centre Utrecht, Utrecht, The Netherlands
- 3Epidemiology and Surveillance Unit, National Institute of Public Health and the Environment (RIVM), Bilthoven, The Netherlands
- 4Hubert Department of Global Health, Rollins School of Public Health, Emory University, Atlanta, Georgia, USA
- Correspondence to Dr Marloes W Heijstek, Department of Paediatric Immunology, Wilhelmina Children's Hospital, University Medical Centre Utrecht, Room number KC 03.063.0, PO Box 85090, Utrecht 3508 AB, The Netherlands;
- Accepted 24 February 2013
- Published Online First 16 March 2013
Objectives The kinetics of the antibody response induced by meningococcal serogroup C (MenC) conjugate vaccination was analysed in patients with juvenile idiopathic arthritis (JIA) to assess their long-term protection against MenC disease.
Methods In The Netherlands, a nationwide catch-up campaign was performed in 2002 during which children aged 1–19 years, including JIA patients, received the MenC conjugate vaccination. From 127 JIA patients, IgG antibody concentrations against MenC-polysaccharide were determined by a fluorescent-bead-based immunoassay in 402 serum samples collected between 2002 and 2010. Using a hierarchical linear regression model, the 8 years course of MenC-specific antibodies was analysed in four age groups (13–19, 9–12.9, 5–8.9 and 1–4.9 years), and in patients starting with methotrexate or biologicals. In 65 randomly selected samples, the correlation of MenC-specific IgG concentrations with serum bactericidal assay (SBA) titres was assessed. MenC-specific IgG concentrations at 4.2 years after vaccination were compared with those of 1527 age-matched healthy controls.
Results MenC-specific IgG concentrations postvaccination were highest in patients aged 13–19 years at time of vaccination. Antibodies gradually waned over time in patients, but their estimated concentrations at 4.2 years postvaccination were similar to those measured in controls. MenC-specific IgG concentrations correlated well with SBA titres (r=0.72, p<0.001). By contrast with methotrexate, starting treatment with biologicals induced a trend towards accelerated decline of MenC-specific antibodies.
Conclusions Persistence of MenC-specific IgG antibodies in JIA patients is similar to healthy controls, but treatment with biologicals may induce accelerated antibody waning, resulting in unprotected patients who may need revaccination.
Patients with juvenile idiopathic arthritis (JIA) may be at increased risk of infections due to the immunosuppressive effect of the inflammatory disease or its treatment.1 ,2 Therefore, bacterial infections, like Neisseria meningitidis infections causing meningococcal disease, may occur more frequently in patients with JIA, especially in patients treated with high-dose immunosuppressive drugs or biologicals.2 ,3 In order to prevent meningococcal disease by N meningitidis serogroup C, effective vaccination inducing sustained long-term antibody responses is essential.4
In healthy individuals, vaccination with the meningococcal serogroup C conjugate (MenCC) vaccine is immunogenic and effective in preventing infections.5 ,6 However, in children below 5 years of age at time of vaccination, MenC-specific antibodies wane rapidly after vaccination.4 ,7 In children older than 5 years, antibody responses upon vaccination are higher, and circulating IgG concentrations persist longer, with a positive correlation between persistence of MenC-specific antibodies and age at time of MenC vaccination.7 ,8
For patients with JIA, vaccination against meningococcal disease is recommended in the European League Against Rheumatism (EULAR) guidelines.9 The MenCC vaccine has proven to induce MenC-specific antibodies with bactericidal capacity in all JIA patients at 3 months after vaccination.10 However, the long-term persistence of MenC-specific antibodies is still uncertain.9 Like several other vaccine-specific antibodies,11 the persistence of MenC-specific antibodies may be lower in JIA patients compared with healthy controls. Low circulating antibody levels may place vaccinated subjects at risk of infection.4 Furthermore, it is unknown whether the height and persistence of MenC-specific antibody concentrations in JIA are similarly affected by age at time of vaccination as in healthy controls, and whether the use of antirheumatic drugs influences the kinetics of MenCC vaccine-induced antibodies.
We analysed the kinetics of MenC-specific antibodies in various age groups of patients with JIA, and compared the height of MenC-specific IgG concentrations between patients and age-matched healthy controls at approximately 4 years post-MenCC vaccination. Furthermore, the effect of antirheumatic treatment (methotrexate, biologicals) on the persistence of MenC-specific antibodies was evaluated.
The MenCC vaccine was implemented in the Dutch national immunisation programme in September 2002 for all children at the age of 14 months. In addition, a catch-up campaign was conducted between June and November 2002 during which all children aged 1–19 years, including JIA patients, were invited to receive a single MenCC vaccination. National vaccination coverage was 94%.12
In this study, two aspects of the MenC-specific antibody response in JIA patients were retrospectively assessed using stored serum samples obtained during routine outpatient visits after vaccination during the catch-up campaign. First, the long-term persistence of MenC-specific antibodies after MenCC vaccination was assessed in a retrospective cohort of patients. As the long-term persistence is known to be influenced by age at time of vaccination in healthy children,7 this analysis was performed in four different age cohorts. The persistence of MenC-specific antibodies in patients was compared with that of healthy age-matched controls selected from a Dutch nation-wide cross-sectional study performed between February 2006 and June 2007 (ISRCTN 20164309).7 MenC-specific antibody levels from these healthy controls had been determined earlier as previously reported.7 Second, the effects of treatment with methotrexate and biologicals on the long-term persistence of MenC-specific antibodies were analysed in the retrospective JIA cohort.
As MenC-specific serum bactericidal antibody assay (SBA) titres express the functional capacity of MenC-specific antibodies,13 ,14 the correlation between MenC-specific immunoglobulin G (IgG) antibody concentrations and SBA titres was assessed in a randomly selected subset of 65 serum samples to confirm the functional capacity of MenC-specific antibodies in JIA patients.
Setting and participants
Patients with a confirmed JIA diagnosis according to the International League of Associations for Rheumatology criteria15 were recruited from a tertiary centre for paediatric rheumatology (Wilhelmina Children's Hospital, University Medical Centre, Utrecht, The Netherlands). Patients were eligible for inclusion when (1) they were born between September 1983 and February 2001; (2) they had been vaccinated against MenC during the catch-up campaign (2002); (3) at least one serum sample was obtained from them after the catch-up campaign in the period 2002–2010. In total, 402 stored serum samples were available from 127 patients. The study was approved by the medical ethics committee of the University Medical Centre, Utrecht. Informed consent was obtained to use stored samples.
As described above, the persistence of MenC-specific IgG antibody concentrations in patients was assessed by comparing estimated values from patients with measured values from a cohort of healthy controls from a Dutch nationwide cross-sectional study performed in 2006–2007.7 From this cohort, individuals born between September 1983 and February 2001 and vaccinated during the catch-up campaign in 2002 were selected (n=1527 healthy controls). Their MenC-specific IgG concentrations had been measured previously in the same laboratory as the antibody concentrations of JIA patients.
MenC conjugate vaccine
The NeisVac-C vaccine (Baxter Healthcare, Vienna, Austria) used in the Dutch national immunisation programme, contains the N meningitidis serogroup C polysaccharide (strain C11, 10 μg) conjugated to tetanus toxoid (10–20 μg). As the exact vaccination date could not be retrieved for all participants, the vaccination date for all participants was set at 1 September 2002.
MenC polysaccharide-specific IgG antibody levels were determined by a fluorescent-bead-based immunoassay, as previously described.16 MenC-specific functional antibody titres were determined with the SBA using baby rabbit complement and serogroup C strain C11.17 SBA titres were expressed as the reciprocal of the final serum dilution with ≥50% killing after 60 min. An SBA titre ≥8 is the accepted correlate of protection against MenC disease.13 ,14
To assess the effect of age at time of MenCC vaccination on MenC antibody responses, the JIA cohort was divided into four cohorts: (1) patients born between September 1983 and December 1989, aged 13–19 years (adolescents, n=21); (2) patients born between January 1990 and December 1993, aged 9–12.9 years (preadolescents, n=41); (3) patients born between January 1994 and December 1997, aged 5–8.9 years (young children, n=48) and (4) patients born between January 1998 and January 2001, aged 1–4.9 years (toddlers, n=17).
To compare the long-term persistence of MenC-specific IgG concentrations between patients and healthy controls, controls were divided into the same four cohorts based on age at time of MenCC vaccination.
In the treatment of JIA, methotrexate is the cornerstone disease-modifying antirheumatic drug.18 Additional antirheumatic drugs, such as TNFα, IL-1R or IL-6 antagonists, are started in patients who insufficiently respond to methotrexate.19 To evaluate the effects of these antirheumatic drugs on the long-term persistence of MenC-specific IgG concentrations, the starting dates of methotrexate and/or biologicals were retrieved from the medical charts and the duration between treatment start and MenCC vaccination was calculated.
MenC-specific IgG concentrations were log-transformed and analysed with a multilevel linear regression model in a Bayesian framework, using Markov chain Monte Carlo methods.20 ,21 In short, this approach creates a linear decay curve for every study participant separately by fitting a regression equation to the measured MenC-specific IgG values, and estimating all missing values using Bayes rule. The variation among all participants within the same age group is taken into account in this process. In detail, observed log antibody concentrations Y at time T following vaccination are considered to be normally distributed with mean μ and SD σ (representing measurement error)
The mean log concentration is assumed to decline with time as
Parameter α represents the initial antibody concentration after vaccination, β the rate of decay (β<0). All parameters are estimated iteratively, resulting in a set of values (a Monte Carlo sample) that eventually converge around the best estimates. Rapid convergence (within a few hundred iterations) and Gelman Rubin statistics that approximate 1 indicate that the data allow accurate estimation of the model's parameters. To describe their variation among individuals, the parameters α and β are assumed random, and normally distributed
with mean and SD stratified by age, thus allowing for individual variation within each age group. Parameters are subsequently used to estimate IgG concentrations. If there is a good model fit, the estimated values are close to the observed values. The model was implemented in WinBUGS, and after a burn-in of 10 000 iterations, results were obtained from the subsequent 10 000 iterations.
The correlation between MenC-specific IgG concentrations and SBA titres was calculated using Spearman's rank correlation.
The mean time elapsed between MenCC vaccination and serum sample collection in healthy controls was 4.2 (±0.4) years. To compare the persistence of MenC-specific IgG concentrations between patients and healthy controls, MenC-specific IgG concentrations in patients were estimated at 4.2 years after MenCC vaccination using the constructed statistical model.
The impact of starting methotrexate or biologicals on the decay of MenC-specific IgG concentrations was evaluated by incorporating a change in decay rate starting from the onset of treatment into the regression model. If treatment started at time Tt after MenCC vaccination, the mean antibody concentration was calculated as
where the min function returns the minimum of its arguments, and the step function (WinBUGS) returns 0 when T−Tt≤0 and 1 elsewhere. This results in a linear decay curve starting from μ=α at T=0 and decay rate β1 for 0<T≤Tt, switching to a decay rate β2 for T>Tt. If treatment had started prior to vaccination, Tt=0 was used, whereas for patients who never received treatment during follow-up, Tt=100 was used. Source code for the regression models is available from the authors on request. See online supplementary table S1–S4 for exact numbers used for the models.
In total, 127 patients with JIA, aged 8.9 (±3.7) years, and 1527 healthy controls, aged 9.1 (±5.1) years, at time of MenCC vaccination were included (table 1). Age and gender did not differ significantly between patients and controls. Serum samples from patients were available from the period between 2 weeks and 8 years after MenCC vaccination.
After MenCC vaccination, methotrexate was started in 66 (52%) patients and biologicals in 53 (42%). The duration between MenCC vaccination and initiation of treatment was 3.1 (±1.9) years for methotrexate and 4.8 (±1.8) years for biologicals. None of the patients used rituximab at time of vaccination or during follow-up.
Medication use (table 2) reflected the standard treatment modalities in an academic referral centre. After MenCC vaccination, methotrexate was started in relatively more patients in the younger age groups (65–71%) than in the older age groups (29–42%; p=0.008). Biologicals were also started relatively more frequent in young patients (53%) than in adolescent patients (24%), albeit non-significant (p=0.282, table 2). The mean time between MenCC vaccination and starting methotrexate (p=0.926) or biologicals (p=0.140) did not differ between age groups.
MenC-specific IgG concentrations after MenCC vaccination in patients with JIA according to age at time of vaccination
MenC-specific IgG concentrations after vaccination in JIA patients showed a clear age-dependent pattern with older children reaching higher antibody levels directly after vaccination (figure 1). The Gelman Rubin statistics approximated 1, and the Markov chains showed rapid stable convergence of the models, indicating that MenC-specific IgG concentrations could be accurately predicted. The estimated antibody concentrations in the 13–19 years age group postvaccination (geometric mean concentrations (GMCs) 8.4 μg/ml (95% predictive interval (PI) 3.9–18.2)) were higher than the concentrations in the younger cohorts, and remained higher for several years. Patients aged 1–4.9 years at MenCC vaccination reached the lowest postvaccination MenC-specific IgG concentrations (GMC 0.6 μg/ml (95% PI 0.3–1.3)). Over time, MenC-specific antibodies gradually declined, resulting in low antibody concentrations 8 years after MenCC vaccination in all patients. The estimated half-life of MenC-specific antibodies was the shortest in the oldest vaccinees (t½=2.3 years; table 2).
The MenC-specific IgG concentrations in the subset of 65 serum samples from 43 patients correlated well with the functional SBA titres (r=0.72, p<0.001), indicating that patients with low MenC-specific IgG concentrations are indeed likely to be unprotected against MenC disease.
Long-term persistence of MenC-specific antibody concentrations in patients with JIA and healthy controls according to age at time of vaccination
At 4.2 years after MenCC vaccination, the estimated MenC-specific IgG concentrations in patients were remarkably similar to levels measured in healthy controls for all age groups (figure 2). Adolescents showed the highest GMCs (patients 2.3 μg/ml (95% PI 1.2–4.7) versus healthy controls 2.3 μg/ml (95% CI 2 to 2.8)). Although the estimated antibody levels in patients aged 9–12.9 years at time of vaccination (GMC 0.7 μg/ml (95% PI 0.4–1.2)) seemed slightly lower compared with healthy controls (GMC 1.0 μg/ml (95% CI 0.8 to 1.2)), the 95% intervals overlapped indicating that these differences were non-significant. The youngest age group showed the lowest MenC-specific IgG concentrations 4.2 years after MenCC vaccination, again with comparable levels between patients (GMC 0.2 μg/ml (95% PI 0.1–0.5)) and healthy controls (GMC 0.2 μg/ml (95% CI 0.2 to 0.3)).
Impact of starting antirheumatic drugs on the persistence of MenC-specific antibodies in patients with JIA
Starting methotrexate treatment did not affect the decline of MenC-specific IgG concentrations (figure 3A). However, starting biological treatment induced a trend towards accelerated decay in MenC-specific antibodies, with a faster predicted decay rate in 92.6% of patients (figure 3B). The Gelman Rubin statistics approximated 1, and the Markov chains showed rapid stable convergence. The good model fit and the ability of the model to accurately predict the effects of biological treatment on MenC-specific antibody decay are illustrated in figure 4.
MenC-specific IgG concentrations were highest in adolescent JIA patients shortly after MenCC vaccination and for several years afterwards. In all age cohorts, MenC-specific IgG concentrations gradually waned over time. However, the long-term persistence of MenC-specific antibodies at approximately 4 years after vaccination was similar in patients and healthy controls. A trend towards accelerated decline of MenC-specific IgG concentrations was observed after starting treatment with biologicals. Methotrexate alone did not affect the decline of MenC-specific antibodies.
The fact that MenCC vaccination induced the highest humoral responses in adolescent patients (figure 1) is in accordance with what is known from healthy controls.4 ,6–8 ,13 ,22–29 The persistence of MenC-specific antibodies at 3–6 years after one dose of MenCC vaccine is known to be much better in older vaccinees compared with children who are younger at time of vaccination.4 ,8 ,13 ,27–29 This might be due to higher initial humoral responses to the MenCC vaccine at older ages,13 as detected in our patient cohort (figure 1). Proposed mechanisms for these higher responses are immunological priming and maturation of the immune system through childhood.29 Additionally, it has been postulated that the superior persistence of MenC-specific antibodies in older vaccinees may be due to a slower decline in antibody concentrations after vaccination.29 However, we did not find a slower rate of decline in MenC-specific antibodies in the eldest patients, as the estimated half-life of MenC-specific antibodies in older vaccinees was lower (t½=2.3 years) than in the younger patients (estimated t½=2.9–6.6 years). This suggests that solely the initial higher humoral responses account for the longer persistence of MenC immunity in older vaccinees. Due to waning immunity in young vaccinees, several countries consider the implementation of a MenC-booster for (pre)adolescents into their immunisation programme.
The safety and short-term immunogenicity of the NeisVac-C vaccine has been studied previously in 234 JIA patients.10 The vaccine was found to be safe with no effect on disease activity, and showed good short-term immunogenicity with all patients having protective functional antibodies (SBA titre ≥8). Other conjugate vaccines (ie, against Haemophilus influenzae type B and pneumococci) also showed to induce a pronounced short-term vaccine-specific antibody response in patients with rheumatic diseases. However, immunosuppressive treatment negatively affected antibody responses to conjugate vaccines,30 ,31 with lower MenC-specific IgG concentrations in patients on high-dose immune suppressive drugs at time of vaccination.10 The current study adds information on the long-term immunogenicity, and shows that starting treatment with biologicals after MenCC vaccination may accelerate the decay rate of MenC-specific IgG concentrations, whereas methotrexate does not.
As described for healthy children,4 ,13 ,27–29 MenC-specific IgG concentrations waned in the majority of patients until 8 years after vaccination. This is worrisome, as the most important predisposing factor for invasive disease is lack of protective functional antibodies.32 We determined the correlation of MenC-specific IgG concentrations and SBA titres, which was high (r=0.72, p<0.001). The low MenC-specific IgG concentrations, therefore, implicate that a substantial proportion of patients is unprotected several years after vaccination, irrespective of age at time of vaccination. Particularly, patients using biologicals, who tend to have lower short-term antibody responses after MenCC vaccination,10 and in whom circulating MenC-specific antibodies tend to wane more rapidly, may eventually not be protected against meningococcal disease.
The age groups in our study differed in the frequency of starting methotrexate or biologicals after MenCC vaccination during follow-up, with the lowest rate of starting methotrexate and biologicals in the adolescent vaccinees. Differences in methotrexate start will not have influenced the estimated waning of antibodies in relation to age, since we found that starting methotrexate after MenCC vaccination did not influence antibody waning. However, the lower frequency of starting biologicals in the eldest patients could have resulted in an underestimation of average antibody decline compared with the younger vaccinees, since starting biologicals resulted in accelerated antibody waning. As the eldest patients already showed the fastest antibody decline, the differences in antibody waning between the eldest patients and younger patients may, in fact, even be larger.
In our study, precise vaccination data were not known in all patients. This is of little consequence since the catch-up campaign was performed in a very short time period between June and November 2002. In addition, the previous study performed in our centre showed that only 3% of patients was not vaccinated against MenC.10 This is approximately similar to the nationwide MenCC vaccination coverage of 94% during the catch-up campaign.12 Therefore, the chance that potential differences in vaccination coverage between patients and controls have biased our results is considered negligible.
In conclusion, adolescent patients with JIA showed the highest MenC-specific IgG antibody responses upon MenCC vaccination. Persistence of MenC-specific IgG concentrations was similar in patients and healthy controls. Irrespective of age, MenC-specific antibodies waned over time, with a trend towards accelerated decline after starting biologicals. Especially in these JIA patients, monitoring MenC-specific antibodies or routine revaccination in endemic areas seems warranted for better protection against meningococcal serogroup C disease.
Handling editor Tore K Kvien
SPS, MWH shared first author, respectively NMW, GAMB shared last author.
Contributors Study design: SPS, MWH, EAMS, GAMB, NMW. Data collection: SPS, KMS, MWH. Statistical analyses: SPS, MWH, PFMT. Manuscript drafting, revision and approval of final version: SPS, MWH, KMS, FK, EAMS, PFMT, GAMB, NMW. All authors, external and internal, had full access to all data (including statistical reports and tables) in the study and can take responsibility for the integrity of the data and the accuracy of the data analysis.
Funding NMW received grants from the Dutch Arthritis Association (Project number 07-02-403). Funders were not involved in the study design; in the collection, analysis and interpretation of data; in writing of the report; or in the decision to submit the article for publication.
Competing interests None.
Ethics approval University Medical Centre, Utrecht.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement Source code for the regression models is available from the authors on request.