A phase IV, multi-centre, randomized clinical trial comparing two pertussis-containing vaccines in pregnant women in England and vaccine responses in their infants

In this phase IV, multi-centre, randomized clinical trial, pregnant women were randomized to receive either REPEVAX-IPV or BOOSTRIX-IPV (Additional file 1: Protocol). A contemporaneous non-randomized control group of infants born to women who had not received a pertussis-containing vaccine in pregnancy were also recruited.

Pregnant women, aged 16–45 years at enrolment, receiving antenatal care at St George’s University Hospitals NHS Foundation Trust, or in primary care sites in Gloucestershire and Hertfordshire, were eligible to participate. Exclusion criteria included a bleeding disorder, receipt of a pertussis-containing vaccine in the previous 12 months, receipt of a blood product within the preceding 3 months and any contraindication to vaccination specified in the “Green Book” Immunisation against Infectious Disease [9]. Infants born to women who had not received a pertussis-containing vaccine in the previous 12 months were excluded if there was any contraindication to vaccination specified in the “Green Book” Immunisation against Infectious Disease [9]. Infants were included irrespective of gestational age at birth.

Following written, informed consent, pregnant women were randomized 1:1 to receive either REPEVAX-IPV (2 International Units [IU] diphtheria toxoid [DT], 20 IU tetanus toxoid [TT], 2.5 μg pertussis toxoid, 5 μg filamentous haemagglutinin [FHA], 3 μg pertactin [PRN], 5 μg fimbriae [FIM] types 2 and 3 and inactivated poliovirus [IPV, 40 D-antigen unit Type 1, 8 D-antigen unit Type 2, 32 D-antigen unit Type 3]; TdaP₅-IPV; Sanofi Pasteur) or BOOSTRIX-IPV (2 IU DT, 20 IU TT, 8 μg pertussis toxoid, 8 μg FHA, 2.5 μg PRN and IPV [40 D-antigen unit Type 1, 8 D-antigen unit Type 2, 32D-antigen unit Type 3]; TdaP₃-IPV; GlaxoSmithKline (GSK)).

A computerized block randomization list was generated by the study statistician, with sites allocated blocks of sequential numbers (block size 8).

Visits for pregnant women occurred at 28–32 weeks of gestation, up to 7 days post-partum, and at 13 months following delivery. Cord blood or, if not obtained, peripheral blood up to 7 days of age, was collected from infants born to vaccinated mothers. All infants had peripheral blood collected at 2 months (49–84 days of age), 5 months (21–42 days after third primary vaccinations) and 13 months of age (21–42 days after routine booster vaccines). The study was approved by the MHRA, NHS Health Research Authority and City & East Research Ethics Committee (14/LO/0141). The study was registered with ClinicalTrials.​gov (NCT02145624) prior to study commencement.

Pregnant women received either TdaP₅-IPV or TdaP₃-IPV as a 0.5-ml intramuscular injection into the left upper arm at 28–32 weeks of gestation. Infants received routine vaccines according to the nationally recommended schedule at the time of the study (Table 1).

Women were observed for 20 min post-vaccination for any immediate reaction. Adverse events and serious adverse events (SAEs) were collected for women and infants at each study visit.

Serum IgG to PT, FHA, PRN and FIM 2&3 were quantified using enzyme-linked immunosorbent assays (ELISAs), developed in-house and performed by staff blinded to group allocation. All assays have been validated in accordance with International Conference on Harmonisation guidelines and use the 1st World Health Organization (WHO) International Standard Pertussis Antiserum (human) 06/140 (NIBSC, Item No. 06/140). The lower limit of detection (LLOD) of the assays are 2.128 IU/ml (PT), 0.715 IU/ml (FHA), 0.806 IU/ml (PRN) and 0.636 U/ml (FIM 2&3) with results less than the LLOD, assigned a value half of the LLOD.

Sample size calculation was based on the standard deviation of the post-primary vaccination anti-PT IgG GMC of 0.28 IU/ml from a previous study, generated using the same validated PT ELISA [10]. A sample size of 65 per study arm enabled detection of 1.38-fold differences or greater between study arms with 80% power at a 5% significance level. To allow for loss to follow-up, the target sample size was 75–80 in each vaccinated group, with 50 mother-infant pairs in the non-randomized control group.

To increase power, the data for infants born to unvaccinated mothers was supplemented with data from 19 infants from another study conducted at the same sites, at a similar time for which laboratory analysis was performed by the same laboratory using the same assays (NCT01896596). Infants in this study received Infanrix hexa (GSK) instead of Infanrix-IPV-Hib at 2, 3 and 4 months and had been randomized to receive one of three different Men C vaccines at 3 months of age. Blood samples were collected at 5 and 13 months.

A modified intention-to-treat analysis was performed; a per-protocol analysis was not performed as there were fewer than 10% of individuals with data that differed between these populations. Pertussis IgG GMCs and geometric mean fold ratios (GMR) between groups, with 95% confidence intervals (CI), were calculated (Additional materials 2: Statistical Analysis Plan). Normal errors regression on log-transformed data was used to investigate the effect of pre-primary antibody on post-primary GMCs. Two-sided 5% significance was shown when the 95% CI for the GMR or fold effect of pre-primary antibody did not contain unity. The placental transfer ratio (PTR) was calculated as the geometric mean ratio of infant-to-mother-specific IgG at delivery; normal errors regression on logged ratios with adjustment for the interval between vaccination and birth were calculated. The effect of interval from birth to blood test, sex, ethnicity, birth weight and previous maternal pertussis vaccination was also examined by multiple regression when assessing transfer ratios. The proportion of mothers and infants experiencing SAEs was calculated for each group. Missing data were random and excluded from analyses and analyses were performed using Stata version 13.

The PTR of IgG was greater than 1 for all pertussis antigens, with no difference observed according to maternal vaccine received. There was a strong positive correlation between infant and maternal GMCs of IgG for PT, FHA, FIM 2&3 and PRN. The placental transfer ratio increased with increasing time from vaccination to birth of the infant.

Multivariable analysis was performed to identify the factors associated with PTR. Time from vaccination to birth was significantly associated with PTR, with a fold change of 1.08 (95% CI 1.05–1.10), 1.10 (1.08–1.13), 1.06 (1.04–1.09) and 1.11 (1.08–1.14) for PT, FHA, FIM 2&3 and PRN respectively per week. If a cord sample was not collected, the time at which the infant peripheral blood sample was obtained was significantly associated with PTR with a fold change of 0.97 (0.96–0.99), 0.97 (0.95–0.99), 0.98 (0.96–1.00) and 0.99 (0.96–1.01) for PT, FHA, FIM and PRN respectively per day from birth to blood sampling. Other factors assessed were not associated with the PTR.

Prior to primary vaccination, infants born to mothers vaccinated with TdaP₅-IPV had lower GMC of anti-PT and FHA IgG compared to infants born to TdaP₃-IPV-vaccinated mothers (GMR 0.64 [95% CI 0.43–0.94] and 0.48 [0.35–0.65] respectively). Infants born to mothers vaccinated with TdaP₅-IPV had higher GMCs of anti-FIM IgG (GMR 8.71 [5.2–14.58]). Anti-PRN IgG were similar for infants born to TdaP₅-IPV- and TdaP₃-IPV-vaccinated mothers.

Compared to infants in the control group, infants born to vaccinated women had higher GMCs of anti-PT, FHA, FIM 2&3 and PRN IgG (GMR 4.05 [2.45–6.68], 5.32 [3.55–7.96], 16.43 [8.47–31.88] and 31.52 [15.41–64.46] respectively for the DTaP₅-IPV group; GMR 6.36 [3.82–10.61], 11.12 [7.36–16.79], 1.89 [0.96–3.71] and 23.73 [11.46–49.15] respectively for the TdaP₃-IPV group).

Following infant primary vaccination at 2, 3 and 4 months, infants born to women vaccinated with TdaP₅-IPV had similar GMCs of anti-PT, FHA and PRN as infants born to women vaccinated with TdaP₃-IPV, but higher anti-FIM 2&3 IgG concentrations.

The GMC of anti-PT IgG in infants born to TdaP₅-IPV- or TdaP₃-IPV-vaccinated mothers was lower than that in infants born to unvaccinated mothers (GMR 0.71 [0.56–0.90] and 0.78 [0.61–0.98], respectively). For anti-FHA IgG, this effect was only seen in infants born to TdaP₅-IPV-vaccinated mothers (GMR 0.75 [0.6–0.94]).

GMCs of anti-PT IgG increased from pre- to post-primary vaccination for all infants (TdaP₅-IPV group: fold change 1.89 [1.33–2.69]; TdaP₃-IPV group 3.25 [2.29–4.62]; unvaccinated group 14.1 [8.67–22.93]). However, there was no significant difference in the fold change in IgG to any pertussis antigens from pre- to post-primary vaccination in infants born to mothers receiving TdaP₅-IPV compared to those infants receiving TdaP₃-IPV.

At 13 months of age, GMCs of anti-PT, FHA and FIM 2&3 were similar in those infants born to TdaP₃-IPV- and TdaP₅-IPV-vaccinated mothers, whereas GMCs of anti-PRN IgG was lower in infants born to TdaP₅-IPV-vaccinated mothers compared to infants born to DTaP₃-IPV-vaccinated mothers (GMR 0.67 [0.47–0.97]). Compared to infants whose mothers received neither vaccine in pregnancy, infants born to TdaP₃-IPV-vaccinated mothers had lower GMCs of anti-FHA (GMR 0.63 0.44–0.91). For anti-PRN IgG, infants born to TdaP₅-IPV-vaccinated mothers had lower concentrations compared to unvaccinated infants (GMR 0.63 [0.42–0.93]).

A higher GMC of anti-PT IgG at 2 months of age was associated with a lower GMC of anti-PT IgG at 5 months of age; the fold change of the post-primary GMC of anti-PT IgG was 0.92 (0.87–0.98) per twofold change in the pre-primary GMC of anti-PT IgG. This effect was not observed at 13 months of age. A similar effect was seen for anti-FHA IgG at 13 months of age (fold change in anti-FHA IgG GMC at 13 months of age 0.87 [95% CI 0.77–0.99] per twofold change in pre-primary GMC), but not at 5 months of age (fold change in post-primary anti-FHA IgG GMC 1.07 [95% CI 0.99–1.14] per 2-fold change in pre-primary GMC). Conversely, GMCs of anti-FIM 2&3 IgG were higher post-primary vaccination with lower concentrations pre-vaccination (fold change in anti-FIM IgG GMC at 5 months of age 1.61 [95% CI 1.55–1.68] per twofold change in pre-primary GMC).

Overall, there were 5 maternal SAEs: 2 in women in the TdaP₅-IPV group, 2 in TdaP₃-IPV and one in an unvaccinated mother. There were 10 SAEs in infants born to women in the TdaP₅-IPV group, 4 in infants born to women in the TdaP₃-IPV group and 3 in infants in the control group. No SAEs were assessed as related to maternal or infant vaccination.

This randomized clinical trial is the first to compare the effect of vaccination with either TdaP₅-IPV or TdaP₃-IPV in pregnancy on transplacental transfer of antibody and the serological response to infant primary immunization and the concentration of vaccine-specific antibody at 13 months of age.

We did not include a randomized control group as this would have been unethical in the context of national recommendations for antenatal pertussis vaccination. We recruited the control group of women and infants in the postnatal period to ensure that women were not discouraged from receiving a pertussis-containing vaccine in pregnancy. We had difficulty recruiting to the control group; therefore, included data from infants included in another study, carried out at the same sites using similar protocols and with samples analysed using identical methods in the same laboratory.

The combination vaccine used in the maternal pertussis vaccination programme in the UK contains IPV; to our knowledge, other countries do not use a combination vaccine that includes IPV. Despite this, we do not consider that this would have significantly influenced our results. The UK primary immunization schedule is administered at 2, 3 and 4 months of age; other countries employ different primary immunization schedules; therefore, this might impact on the generalizability of the results.

The study commenced towards the beginning of the antenatal pertussis programme and most women received a pertussis-containing vaccine for the first time in the current pregnancy; further studies should examine the effects of repeat doses of vaccine in subsequent pregnancies.