Article Text

An open-label, phase IV randomised controlled trial of two schedules of a four-component meningococcal B vaccine in UK preterm infants
  1. Anna Calvert1,2,
  2. Nick Andrews3,
  3. Sheula Barlow4,
  4. Ray Borrow5,
  5. Charlotte Black6,
  6. Barbara Bromage7,
  7. Jeremy Carr6,8,
  8. Paul Clarke9,10,
  9. Andrew C Collinson7,
  10. Karen Few9,
  11. Naomi Hayward1,2,
  12. Christine E Jones11,12,
  13. Kirsty Le Doare1,13,14,
  14. Shamez N Ladhani1,2,3,
  15. Jennifer Louth5,
  16. Georgia Papadopoulou4,
  17. Michelle Pople15,
  18. Tim Scorrer15,
  19. Matthew D Snape4,6,
  20. Paul T Heath1,2
  1. 1Centre for Neonatal and Paediatric Infection and Vaccine Institute, St George's, University of London, London, UK
  2. 2St George's University Hospitals NHS Foundation Trust, London, UK
  3. 3Immunisation and Vaccine Preventable Diseases Division, UK Health Security Agency, London, UK
  4. 4Oxford University Hospitals NHS Foundation Trust, Oxford, UK
  5. 5Vaccine Evaluation Unit, UK Health Security Agency, Manchester Royal Infirmary, Manchester, UK
  6. 6Oxford Vaccine Group, Department of Paediatrics, University of Oxford, Oxford, UK
  7. 7Royal Cornwall Hospitals NHS Trust, Truro, Cornwall, UK
  8. 8Monash University, Clayton, Victoria, Australia
  9. 9Neonatal Intensive Care Unit, Norfolk and Norwich University Hospital NHS Trust, Norwich, UK
  10. 10Norwich Medical School, University of East Anglia, Norwich, UK
  11. 11Faculty of Medicine and Institute for Life Sciences, University of Southampton, Southampton, UK
  12. 12NIHR Southampton Clinical Research Facility and NIHR Southampton Biomedical Research Centre, University Hospital Southampton NHS Foundation Trust, Southampton, UK
  13. 13Makerere University Johns Hopkins University, Kampala, Uganda
  14. 14Pathogen Immunology Group, UK Health Security Agency, Salisbury, UK
  15. 15Portsmouth Hospitals University NHS Trust, Portsmouth, UK
  1. Correspondence to Dr Anna Calvert; acalvert{at}


Objective To compare immunological responses of preterm infants to a four-component meningococcal B vaccine (4CMenB; Bexsero) following a 2+1 vs a 3+1 schedule, and to describe reactogenicity of routine vaccines.

Design An open-label, phase IV randomised study conducted across six UK sites.

Setting Neonatal units, postnatal wards, community recruitment following discharge.

Participants 129 preterm infants born at a gestation of <35 weeks (64 in group 1 (2+1), 65 in group 2 (3+1)) were included in the analysis. Analysis was completed for postprimary samples from 125 participants (59 in group 1, 66 in group 2) and for postbooster samples from 118 participants (59 in both groups).

Interventions Infants randomised to 4CMenB according to a 2+1 or a 3+1 schedule, alongside routine vaccines.

Main outcome measures Serum bactericidal antibody (SBA) assays performed at 5, 12 and 13 months of age: geometric mean titres (GMTs) and proportions of infants achieving titres ≥4 compared between groups.

Results There were no significant differences in SBA GMTs between infants receiving a 2+1 compared with a 3+1 schedule following primary or booster vaccination, but a significantly higher proportion of infants had an SBA titre ≥4 against strain NZ98/254 (porin A) at 1 month after primary vaccination using a 3+1 compared with a 2+1 schedule (3+1: 87% (95% CI 76 to 94%), 2+1: 70% (95% CI 56 to 81%), p=0.03).

At 12 weeks of age those in the 3+1 group, who received a dose of 4CMenB, had significantly more episodes of fever >38.0°C than those in the 2+1 group who did not (group 2+1: 2% (n=1); 3+1: 14% (n=9); p=0.02).

Conclusions Both schedules were immunogenic in preterm infants, although a lower response against strain NZ98/254 was seen in the 2+1 schedule; ongoing disease surveillance is important in understanding the clinical significance of this difference.

Trial registration number NCT03125616.

  • paediatrics
  • infectious disease medicine

Data availability statement

Data are available on reasonable request.

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


  • Routine vaccination against Neisseria meningitidis serogroup B (MenB) was introduced in the UK in 2015 using a 2+1 schedule at 8 and 16 weeks, followed by a booster at 12 months of age.

  • This schedule has been shown to be immunogenic and effective in term infants, but there is no evidence about the use of this schedule in preterm infants who are known to have reduced responses to some vaccines.


  • This study is the first to investigate the immunogenicity of a recombinant protein-based, four-component outer membrane protein meningococcal B (Bexsero) vaccination in preterm infants, comparing two vaccination schedules: one administered at 8 and 16 weeks followed with a booster at 12 months of age (2+1 schedule), and the other administered at 8, 12 and 16 weeks with a booster at 12 months of age (3+1 schedule).

  • The 2+1 schedule was immunogenic in preterm infants, although following the primary series, the proportion of infants with titres ≥4 against strain NZ98/254 (porin A (PorA)) was lower compared with those receiving the 3+1 schedule.


  • It is important that surveillance continues to monitor cases of invasive meningococcal disease so that any increase in cases caused by PorA expressing strains, particularly in preterm infants, can be identified.


Invasive meningococcal disease (IMD) is most common in infants <1 year, with Neisseria meningitidis serogroup B (MenB) predominating.1 A four-component MenB vaccine (4CMenB; Bexsero, GlaxoSmithKline, Belgium), was licensed for protection against MenB by the European Medicines Agency in 2013 to be given according to a three-dose priming schedule in infants followed by a booster in the second year of life.2 Routine MenB vaccination was introduced in the UK in 2015, using a two-dose priming schedule at 8 and 16 weeks with a booster at 1 year. Two studies have since demonstrated acceptable immunogenicity of this schedule in term infants.3 4

Preterm infants may mount suboptimal immune responses to vaccines compared with term infants,5 although most infants achieve responses above putative protective thresholds.6–11 There are, however, no data on 4CMenB immunogenicity in preterm infants when used at the reduced 2+1 schedule. We conducted a randomised controlled trial to assess the immunogenicity of 4CMenB 1 month after a two-dose or three-dose priming schedule in preterm infants followed by a booster at 1 year, and compared 4CMenB antigen-specific antibody responses at 5, 12 and 13 months. We also assessed the reactogenicity of routine immunisation schedules including 4CMenB in preterm infants.


Study design

This was an open-label, phase IV randomised trial conducted across six UK sites. The study was prospectively registered on in April 2017 (NCT03125616). Recruitment was between September 2017 and September 2018.


Preterm infants born at <35 weeks (50% <30 weeks), with no contraindications to vaccination according to the UK guidance,12 no life-limiting congenital abnormality or prior diagnosis of an immunodeficiency syndrome.


Infants were randomised to receive 4CMenB using a 2+1 (8 weeks, 16 weeks and 1 year) or 3+1 (8 weeks, 12 weeks, 16 weeks and 1 year) schedule alongside routine vaccinations (online supplemental table 1). Randomisation was performed by the study statistician and used two computerised block randomisation lists, one for participants born at <30 weeks of gestation and one for those born at 30–34+6 weeks of gestation. Participants were enrolled by a member of the research team. Allocation was according to a 1:1 ratio, was not blinded and vaccines were given by a member of the research team. As per national guidelines, prophylactic paracetamol was recommended following 4CMenB vaccination at 8, 12 and 16 weeks of age. Blood sampling was performed at 5, 12 and 13 months of age.

Main outcome measures

Immunogenicity was assessed using serum bactericidal antibody (SBA) assays. These were performed at UK Health Security Agency Vaccine Evaluation Unit Manchester.13 Both geometric mean titre (GMT) and proportions achieving a titre ≥4 against the tested strains (44/76-SL (factor H binding protein (fHbp)), 5/99 (Neisseria adhesin A (NadA)) and NZ98/254 (porin A (PorA))) were calculated for each sampling time point. Reactogenicity was assessed by the caregiver completing a 7-day diary after each vaccine; infants vaccinated as inpatients had their cardiorespiratory status recorded for 24 hours before and 72 hours after vaccination by nursing staff.

Statistical analysis

The primary outcome was antigen-specific responses as measured by SBA assays at 5 months (postprimary vaccines) in preterm infants receiving two or three priming doses. Secondary outcomes included antigen-specific responses prior to and 1 month following the booster dose, and reactogenicity assessment following each vaccine.

The SD of the GMT responses to the three antigens was expected to be around 1.0 log units. A sample size of 60 per group was predicted to allow a 1.7-fold difference between groups to be detectable at 80% power at the 5% significance level. A drop-out rate of around 10% was expected so the minimum sample size was 132.

For each schedule, GMTs of the SBA titres against three strains (44/76-SL (fHbp), 5/99 (NadA) and NZ98/254 (PorA)) were calculated with 95% CI 1 month after primary vaccination. Titres were compared using a Kruskall-Wallis test because of non-normal distribution of the log-titres. Proportions with titres ≥4 were also calculated with exact binomial 95% CI and compared using Fisher’s exact test. The analysis was conducted using a modified intention-to-treat approach, including all participants for whom at least one antibody result was available. To investigate protection against individual strains at different time points, analysis was performed using paired postprimary and postbooster samples to assess protective thresholds (SBA ≥4) in individual infants and the percentage of infants with SBA >4 at neither time point, one time point and both time points were calculated. Reactogenicity rates between groups were compared using Fisher’s exact test. Statistical analysis was performed using STATA V.17 (StataCorp, College Station, Texas, USA).


136 infants were recruited: 3 died, 2 were lost to follow-up and 7 withdrew from the study, leaving 124 infants who completed all visits (figure 1). For seven babies who did not complete the study there was no confirmation of consent that data could be used, and analysis therefore included 129 infants. Table 1 summarises the baseline characteristics of these infants.

Figure 1

Consolidated Standards of Reporting Trials flow diagram of participants.

Table 1

Baseline characteristics of infants included in phase IV randomised controlled trial of two schedules of a four-component meningococcal B vaccine in UK preterm infants


Following both primary immunisation and booster, there were no significant differences between schedules in SBA GMTs against strains 44/76-SL (fHbp), 5/99 (NadA) or NZ98/254 (PorA) (figure 2; online supplemental table 2). Prior to boosting, however, the SBA GMT against strain NZ98/254 (PorA) was lower for infants receiving the 2+1 compared with the 3+1 schedule (p=0.002).

Figure 2

GMTs and 95% CIs for serum bactericidal antibody against three strains postprimary, prebooster and postbooster vaccination. Blood sampling was performed at 5 months (postprimary), 12 months (prebooster) and 13 months (postbooster). fHbp, factor H binding protein; GMT, geometric mean titre; NadA, Neisseria adhesin A; PorA, Porin A. **P=0.002.

There were no significant differences in the percentage of infants with an SBA titre ≥4 against strain 44/76-SL (fHbp) or 5/99 (NadA) at any time point. A higher percentage of infants receiving the 3+1 schedule had SBA titres ≥4 against strain NZ98/254 (PorA) after primary immunisation and before the booster, but not after the booster (table 2).

Table 2

Percentages of participants with SBA titres ≥4 for each of the tested antigens postprimary series and prebooster and postbooster vaccination

All infants achieved a protective titre against strain 5/99 (NadA) at both time points. Fewer infants receiving a 2+1 schedule reached a protective threshold against strain NZ98/254 (PorA) compared with those receiving a 3+1 schedule (2+1: both time points=67.3% (n=35), postbooster only 19.2% (n=10), neither time point=13.4% (n=7); 3+1: both time points=87.0% (n=47), postbooster only=7.4% (n=4), neither time point=5.8% (n=3)). The results were similar between schedules for strain 44/76-SL (fHbp) (online supplemental table 3).

Comparison of fold-changes between postprimary and postbooster doses identified modest responses against strain 44/76-SL (fHbp) in both groups, lower than for the other vaccine antigens (figure 2, online supplemental table 4).


Overall, 84 vaccinations in 56 participants took place in a neonatal unit (online supplemental table 5). In the 72 hours following first and second vaccinations, there was no significant difference in rates of apnoea, bradycardia or desaturation compared with the 24-hour periods preceding (online supplemental figures 1 and 2). There were no episodes of apnoea or bradycardia reported before or after the third vaccines, and one infant experienced a single desaturation following these. Local reactions (online supplemental figures 3-7) and non-febrile systemic reactions (figure 3) after routine vaccines were common in both groups at every time point.

Figure 3

Systemic reactions following four-component meningococcal B vaccination alongside routine immunisations at 8, 12 and 16 weeks followed by a booster at 1 year of age.

A larger proportion of infants experienced fever within 7 days of each subsequent vaccination (first vaccines: 6%, second vaccine 2+1: 2%, second vaccines 3+1: 14%, third vaccines: 2+1: 20%, second vaccines 3+1: 14%, fourth vaccines 26%) (online supplemental table 6). Infants receiving the 3+1 schedule had significantly more fever episodes at the time of the second vaccines than those receiving the 2+1 schedule (2+1: 1/58 (2%) vs 3+1: 9/64 (14%); p=0.02), although the proportion of infants experiencing a grade 3 fever (≥39.0°C) was similar (2+1: 0/58 (0%) vs 3+1: 1/64 (2%); p=1.0), as was the proportion of infants investigated for sepsis after the second vaccines (2+1: 2/63 (3.2%) vs 3+1: 2/65 (3.1%); p=1.0) (online supplemental table 7). Most infants (V1: 104/114, 91.2%, V2 3+1: 56/59, 94.9%, V3: 95/99, 96%, V4: 86/91, 94.5%) received at least one paracetamol dose on the day of 4CMenB vaccination (online supplemental table 8). The proportion of babies receiving paracetamol remained high following the booster vaccines at 12 months; indications included prophylaxis in 76.7% (n=46), fever in 18.3% (n=11), both in 3.3% (n=2) and an unrelated symptom in 1.7% (n=1).


There were 100 serious adverse events (SAEs) reported (online supplemental table 9), mainly in infants born at <30 weeks of gestation (73/100, 73%). Six SAEs were related to vaccination: 4 probably related (3 apnoea, 1 fever), all following vaccines including 4CMenB, and 2 definitely related (elective hospitalisation for monitoring because of apnoea following previous vaccination). Three infants died during the study (two in the 3+1 group and one in the 2+1 group); none was assessed to be vaccine-related (RSV bronchiolitis, rhinovirus bronchiolitis, sudden unexpected collapse in hospital).


We found that both 2+1 and 3+1 schedules for 4CMenB were immunogenic in preterm infants, although the GMT against strain NZ98/254 (PorA) at 12 months, and the proportions of infants with postprimary and prebooster SBA titres ≥4 were significantly higher with the 3+1 than 2+1 schedule. The vaccine antigen PorA is the immunodominant component of outer membrane vesicle (OMV) vaccines. The immunogenicity of these vaccines is known to be greater when a 3+1 compared with a 2+1 schedule is used.14 15 The clinical significance of a lower proportion of infants with protective titres against one strain is not clear and, importantly, there was no statistically significant difference after the booster dose meaning that protection after the booster is likely to be similar in the two cohorts. A higher proportion of infants receiving an additional dose of 4CMenB at 3 months experienced fever, although this is unlikely to be of clinical significance since rates of high fever and number of infants investigated for sepsis following vaccination were similar in the two groups.

The GMTs for preterm infants receiving a 2+1 schedule in this trial are similar to those found in term infants following the primary series when a 2+1 schedule was used and when analysis was performed in the same laboratory.4 16 After the booster dose, the reported GMTs in this preterm cohort were similar to those reported by Davis et al,4 but the fHbp GMTs were slightly lower compared with those found by Valente Pinto et al.16 These two studies in term infants also investigated proportions of preterm infants with putative protective titres after the primary vaccine series and the booster dose, and reported similar results to those found in our preterm population.4 16 Another study using a 2+1 schedule in term infants found slightly higher proportions protected after the booster dose compared with our preterm population, but this study was performed in a different country, using a different assay.17 Overall, our results indicate that preterm infants receiving the 2+1 schedule had similar immune responses to term infants receiving the same schedule.

In line with the findings of Davis et al and Valente Pinto et al in term infants,4 16 we found a relatively low response following booster vaccination for fHbp in preterm infants who received a 2+1 schedule. While this might suggest the superiority of a 3+1 schedule in priming, this is not supported by the findings of Martinon-Torres et al.17 Although it is noteworthy that in that study both of the reduced schedules were administered slightly later (3.5 and 5 months or 6 and 8 months) compared with our study, or with the schedules used in the studies by Davis et al and Valente Pinto et al (all at 2 and 4 months), which may explain this difference. The similarity of fHbp responses after the booster irrespective of the priming schedule in preterm infants is reassuring and suggests that both groups would likely be similarly protected in the medium-to-long term.

The protection offered by 4CMenB involves immune responses against four major meningococcal surface protein antigens and many minor antigens in the OMV, which is also part of the vaccine.18 How the observed immunogenicity is translated into clinical protection against IMD is complex because the vaccine antigens are not present on the surface of all meningococci and, when present, their expression on the meningococcal surface varies considerably between strains. Consequently, clinical protection depends on the characteristics of circulating strains causing IMD in different populations.

Estimation of likely vaccine strain coverage in a population is possible using the Meningococcal Antigen Typing System (MATS), which assesses whether a particular meningococcal strain expresses antigens which cross-react with those in 4CMenB.19 In the light of the most recent UK MATS data, it is reassuring that a high proportion of the preterm infants in this study had titres ≥4 against strain 44/76-SL (fHbp) after both primary and booster vaccinations as MATS coverage is often due to fHbp (59% of isolates with matched antigens). Also, as MATS coverage was less dependent on PorA (16% of isolates with matched antigens, and PorA as the single matched antigen in <1% of isolates), it is perhaps less concerning that 13.4% of infants vaccinated according to a 2+1 schedule did not reach a protective threshold for this antigen either after primary or booster vaccination.18 Similar results have been observed in a Canadian study.20

Overall, routine vaccination including 4CMenB was well tolerated. The number of reported SAEs reflects the vulnerability of the preterm population, especially concerning increased hospital admissions related to infections unrelated to vaccination, for example, respiratory tract infections.21 22 Fever after 4CMenB administration, especially when given with other routine infant immunisations, is a particular concern. Reassuringly, preterm infants in the current study experienced fewer fever episodes than previously reported in term infants, irrespective of whether or not they had received prophylactic paracetamol.23–27 Additionally, we observed no differences between study groups in proportions with fever ≥39.0°C or number of infants investigated for sepsis. Systemic and local reaction rates were similar to those previously reported in term infants, apart from local tenderness and vomiting, which were reported more frequently in preterm infants than term infants in some previous studies.23–27

Strengths and limitations

A major strength of this study was the ability to recruit a large and broad sample of preterm infants. One potential limitation of the study is that we did not collect baseline prevaccination antibody titres, but most infants are likely to have very low antibody titres,26 which should be similar between the two study groups. We also did not include a term comparator group and, therefore, our comparisons with vaccine responses in term infants should be interpreted with caution, although two of the UK trials in term infants were conducted in the same reference laboratory using the same methodology and testing protocols.4 16 The study was not blinded, which is particularly relevant for reactogenicity assessment at 12 weeks when infants in group 2 received an additional dose of 4CMenB. It is possible that this increased reporting of reactogenicity in these infants. Finally, we did not follow-up the participants beyond 13 months and, therefore, we cannot comment on long-term immune responses.


The 2+1 schedule was immunogenic in preterm infants, although proportions of infants with putative protective titres against strain NZ98/254 (PorA) were lower postprimary and prebooster in infants after a 2+1 compared with a 3+1 schedule, similar to previous studies in term infants and consistent with previous studies of OMV vaccines.14 15 Postbooster responses against strain NZ98/254 (PorA), however, were similar after the two schedules. Our findings support the current 2+1 4CMenB schedule for both preterm and term infants.

Data availability statement

Data are available on reasonable request.

Ethics statements

Patient consent for publication

Ethics approval

Ethical approval was received from York and Humber Research Ethics Committee (17/YH/0150) and the Health Research Authority. Participants gave informed consent to participate in the study before taking part.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • X @drpaulclarke, @drchrissiejones, @shamezladhani

  • Contributors AC, PTH and SNL devised the study. AC, SB, CB, BB, JC, PC, ACC, KF, NH, CEJ, GP, MP, TS and MS were responsible for study set up in individual sites, participant recruitment, study procedures and follow up, data collection and safety monitoring. RB and JL coordinated the laboratory analysis. AC and NA performed the statistical analysis. AC wrote the first draft of the paper. PTH, SL, CEJ, KLD and NA provided significant input into the first draft of the paper and all authors reviewed the manuscript, made improvements and approved the final version. AC, PTH and NA had full access to the data. AC and PTH are guarantors.

  • Funding This work was funded by Meningitis Now and GlaxoSmithKline.

  • Competing interests JL and RB perform contract research on behalf of UKHSA for GlaxoSmithKline (GSK), Pfizer and Sanofi. JC works for an institution which conducts meningococcal vaccine research on behalf of GSK; he receives no personal payment or inducement of any kind. MS was an employee of the University of Oxford and Oxford University Hospitals Foundation NHS trust up until September 2022, and in this role acted as an investigator for clinical research studies funded or otherwise supported by the vaccine manufacturers GSK, Janssen, AstraZeneca, Pfizer, Novavax and MCM vaccines. He received no personal financial benefit for this work. As of September 2022, MS has been an employee of Moderna UK and holds equity in this company; however, all study activities and data analysis were completed before this date.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Author note Both Meningitis Now and GlaxoSmithKline were provided with the opportunity to review the final version of the manuscript and the supplementary materials for factual accuracy but the authors are solely responsible for the content, interpretation and conclusions from the work.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.