A phase 2 randomized controlled dose-ranging trial of recombinant pertussis booster vaccines containing genetically inactivated pertussis toxin in pregnant women

Highlights • Recombinant acellular pertussis vaccine is safe and immunogenic in pregnant women.• Vaccine formulations containing PTgen were non-inferior to comparator vaccine.• Low-dose ap1gen is good alternative for pregnant women when DT and TT are not needed.


Introduction
Pertussis, a respiratory disease caused by the bacterium Bordetella pertussis, continues to be a significant worldwide health problem [1]. Although pertussis infection can cause disease at any age, the disease is most severe in infants and newborns [2,3] with the highest case fatality rates between birth and 6 to 8 weeks of age [4]. Maternal immunization boosts the concentration of maternal antibodies that can be transferred across the placenta to directly protect infants too young to be immunized against tetanus, diphtheria and pertussis [5].
Several studies had already demonstrated the immunogenicity of the ''pertussis vaccination during pregnancy" vaccination strategy besides the safety and effectiveness of vaccinating pregnant women with tetanus, diphtheria, acellular pertussis (Tdap) vaccine [6][7][8][9]. In addition, Moro and colleagues (2016) had observed no new or unexpected vaccine-related adverse events (AE) in pregnant women who received Tdap vaccines following routine recommendations for maternal Tdap vaccination [10].
Vaccination of pregnant women with Tdap vaccine in the second or third trimester of pregnancy up to 15 days before delivery has been recommended by the World Health Organization [2].
A vaccine containing recombinant genetically inactivated pertussis PT (PT gen ) and filamentous hemagglutinin (FHA) alone (aP5 gen ) or combined with tetanus and reduced-dose diphtheria toxoids (TdaP5 gen ) was developed and licensed for immunization of individuals aged 11 years and older in Thailand and in Singapore. Considering the high immunogenicity of PT gen , we postulated that lower-dose and therefore cost-saving formulations of the licensed recombinant pertussis vaccines may be suitable candidates for maternal pertussis immunization in low-to-middle income countries if shown to induce sufficiently high PT antibody levels that may be assumed to protect newborns against severe pertussis in the first months of life. We obtained initial evidence of acceptable safety and immunogenicity in women of childbearing age in a randomized controlled phase 2 trial of lower dose investigational formulations of these new generation recombinant pertussis vaccines [11], and through a post-licensure safety study of aP5 gen and TdaP5 gen vaccination in 1,778 pregnant women [12]. In the present phase 2 trial, we extended the investigation of these novel PT gen formulations by comparing serological responses and safety in pregnant women following vaccination across several low-dose formulations versus the comparator vaccine.

Study design and participants
The design of this phase 2, dual-site trial was based on a previous randomized observer-blind trial involving 250 women of childbearing age (50 participants per vaccine group) [11] in which the same investigational and comparator vaccines, as described below, were evaluated. The study vaccines were given as single dose to pregnant women (400 participants: 80 per vaccine group) during the second or third trimester of pregnancy at 20-33 weeks of gestation. The vaccination, frequency of study visits, and the safety and immunogenicity assessments are shown in Fig. 1.
Healthy pregnant women, 18 to 40 years old with no underlying chronic medical conditions, and an uncomplicated singleton pregnancy at !20 weeks and <33 weeks gestational age as confirmed by ultrasound were invited to participate. Women who had previously received diphtheria or tetanus or pertussis-containing vaccine within 1 year prior to enrolment were excluded (complete inclusion/exclusion criteria in Appendix Table 1).
The trial was conducted according to the guidelines of the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) and Good Clinical Practice (GCP), the Declaration of Helsinki, and local ethical guidelines. Ethical approval was obtained from the Institutional Review Boards of the Faculty of Medicine Siriraj Hospital at Mahidol University, Faculty of Medicine at Chulalongkorn University, Bangkok, Thailand and Western Institutional Review Board (now known as WIRB-Copernicus Group), Washington, USA. Written informed consent was obtained from all participants before recruitment into the study.

Randomization and masking
Participants were block-randomized in a 1:1:1:1:1 ratio per study site (80 participants per vaccine group; 40 per study site) to receive one dose (0.5 mL intramuscular injection) of either ap1 gen , Tdap1 gen , Tdap2 gen , TdaP5 gen (Boostagen Ò ), or a comparator, Tdap8 chem (Boostrix TM ) according to a computer-generated (PROC PLAN, Statistical Analysis System (SAS Ò ) version 9.4) randomization scheme. A randomization list in a sealed envelope was provided to non-blinded clinical research staff who were responsible for participant randomization for study vaccine assignment, vaccine preparation, administration, and accountability. All other research staff, study participants, and all laboratory personnel performing immunologic testing were blinded to the vaccine administered.

Study vaccines
The composition of the study vaccines is shown in Appendix Table 2. The recombinant acellular pertussis vaccines contain two pertussis components (PT gen and filamentous hemagglutinin, FHA), whereas the chemically inactivated acellular pertussis comparator is a three-component vaccine containing PT chem , FHA, and pertactin (PRN). The study evaluated the stand-alone recombinant acellular pertussis vaccine (ap gen , pertussis-only) and trivalent (combination) vaccines containing tetanus, diphtheria, in addition to recombinant acellular pertussis vaccines (Tdap or TdaP). Uppercase letters 'T' and 'P' in these abbreviations mean the vaccine has full-strength doses of tetanus and pertussis. The lowercase letters 'd' and 'p' refer to reduced-dose diphtheria and pertussis.

Immunogenicity assessments
Venous blood samples (5 mL) were obtained before vaccine administration (Day 0; baseline), and 28 days after vaccination (Day 28). It was then processed for serum separation on the same day. The blood sample was centrifuged at 1300g for 15 minutes at room temperature. The serum sample was collected and stored at À20 ⁰C and below prior to analysis. Concentrations of serum IgG antibody against PT gen and FHA were measured using an inhouse validated indirect enzyme-linked immunosorbent assay (ELISA), and concentrations were expressed in IU/mL, calibrated to the WHO International Standard Pertussis Antiserum (Human) 06/140. The WHO Reference Reagent Pertussis Antiserum (Human) 06/142 (NIBSC, UK) was used as the positive control [13]. Serum TT-and DT-specific IgG concentrations were measured using validated commercially available ELISA kits (Serion ELISA classic Diphtheria IgG kit, ESR130G; and Serion ELISA Classic Tetanus IgG kit, ESR108G, Virion/Serion, Germany), and concentrations were expressed in IU/mL. Functional anti-PT antibody titers were assessed in a randomly selected 30% subset of study participants using a validated Chinese hamster ovary (CHO) cell pertussis toxin neutralization assay and JNIH-5 (NIBSC) as the source of pertussis toxin, as described previously [14,15], with titer reported as IU/mL on the basis of the relative activity of the WHO International Standard Pertussis Antiserum (Human) 06/140 [13]. The lower limits of quantification (LLOQ) (assay cut-off) were 5 IU/mL for PT-IgG, 1 IU/ mL for FHA-IgG and 0.05 IU/mL for DT-IgG and TT-IgG. Antibody concentrations below the assay cut-offs (or LLOQ) were given arbitrary values of half the assay cut-offs.

Safety assessment
The schedule of safety and immunogenicity assessments is shown in Fig. 1. Participants were monitored for 30 minutes post-immunization for any immediate reactions. Diary cards were distributed on the day of vaccination to record solicited adverse events (AEs) at the injection site (pain, redness, swelling, induration), or systemic AEs (fever, headache, fatigue, malaise, arthralgia, chills, myalgia, nausea, vomiting) for 7 days after vaccination. Unsolicited AEs, medically attended adverse events (MAAEs) and serious adverse events (SAEs) were collected until study end. However, in this paper, MAAEs and SAEs were presented up to 28 days after vaccination. AEs were assessed by the investigator for severity using the Division of AIDS (DAIDS) Table for grading the severity of adult and pediatric adverse events of the US National Institute of Health, 2017 [16], and causality by the study vaccine according to ICH guidelines [17] and protocol-specified safety considerations.

Outcomes
The primary outcome was the assessment of non-inferior immunogenicity of each study group (ap1 gen , Tdap1 gen , Tdap2 gen , or TdaP5 gen ) to the comparator group (Tdap8 chem ) based on the geometric mean of anti-PT antibody concentration (GMC) on Day 28 after vaccination in the pooled per-protocol population of pregnant women and women of childbearing age. The secondary outcomes for immunogenicity included the following comparisons of each study group (ap1 gen , Tdap1 gen , Tdap2 gen , or TdaP5 gen ) to the comparator group, Tdap8 chem in pregnant women only: (i) comparison of GMC of anti-PT and anti-FHA antibodies and geometric mean titer (GMT) of PT neutralizing antibody; (ii) comparison of seroconversion rate of anti-PT, anti-FHA antibodies and PT neutralizing antibody; (iii) comparison of seroprotection rate and GMC of anti-DT and anti-TT antibody. For anti-PT, anti-FHA and PT neutralizing antibodies, seroconversion was defined as an antibody concentration/titer increase of four-fold or more from baseline to Day 28. Seroprotection for anti-DT and anti-TT antibody was defined as an antibody concentration ! 0.1 IU/mL at any time point. The secondary outcomes for safety were (i) percentage of participants with solicited local and systemic AEs reported within Pregnant women were given one dose of study vaccine between !20 weeks and <33 weeks gestational age. Local and systemic reactions were collected within 7 days after vaccination. The safety assessment was performed by active follow-up until Visit 4 (2 months after delivery) and by passive follow-up until Visit 10 (13 months after delivery). The blood sample collection for immunogenicity assessment was performed at Visit 1 (before vaccination), Visit 2 (28 days after vaccination) and Visit 3 (delivery). The growth and development assessment and safety assessment for infants were performed at Visit 3 (delivery) until Visit 10 (13 months of infant age). Infants received DTwP-Hib-HB vaccine at 2, 4 and 6 months of age and Pneumococcal 13-valent conjugate vaccine (PCV) at 2, 4 and 12 months of age as primary infant immunization. The cord blood sample at delivery and infant blood sample at Visit 4 (2 months of age), Visit 6 (5 months of age), Visit 8 (7 months of age) and Visit 10 (13 months of age) were collected for immunogenicity assessment. seven days after vaccination (ii) percentage of participants with unsolicited AEs, MAAEs, and SAEs reported through 28 days following vaccination.

Statistical analysis
The sample size was calculated based on non-inferiority test with a significance level of 0.00625 for the primary objective after a Bonferroni adjustment. In order to have sufficient sample size to conduct the primary analysis, we pooled both pregnant women (80 enrolled per group; 400 in total) recruited in the current study and non-pregnant women of childbearing age from the previous study (50 enrolled per group; 250 in total) [11] for a total of 130 women per group in the pooled population.
The power to show comparability of immune response in terms of anti-PT GMCs 28 days after a single dose with study vaccine versus comparator was calculated using a one-sided two-sample ttest with a significance level of 0.00625. Assuming the standard deviation of log 10 anti-PT ELISA antibody concentration was 0.6 in both study vaccine and comparator groups and the true anti-PT GMC ratio between the two groups was 1.0, with 117 evaluable participants per group in the pooled population (assuming an attrition rate of 10%), and a non-inferiority margin of 0.5, the study had 91% power to detect non-inferiority of ap1 gen /Tdap1 gen /Tdap2 gen / TdaP5 gen versus the Tdap8 chem group.
With 80 pregnant women vaccinated per group, the present study had a 95% probability to detect at least one SAE if the true incidence is 3.7%. Conversely, if no SAEs were observed in 80 vaccine recipients, the study would be able to rule out SAEs occurring at a rate of approximately 4.6% or above, based on the upper bounds of the 95% CI.
To assess the primary outcome, GMC of anti-PT antibody at 28 days following immunization in pregnant women and women of childbearing age [11] were computed for each vaccine group along with its two-sided 95% CI, by exponentiating the corresponding log-transformed mean and its 95% CI limit. The ratio of the GMC in each of the ap1 gen /Tdap1 gen /Tdap2 gen /TdaP5 gen groups to that in the Tdap8 chem group and a two-sided 98.75% CI for the ratio were provided. The log 10 -transformed concentrations were used to construct a mean difference between the two vaccine groups and its two-sided 98.75% CI using analysis of covariance (ANCOVA). The log 10 -transformed baseline concentrations and an indicator of study populations were included as covariates. Other variables such as age, study site, and gestational age were evaluated for inclusion in the ANCOVA model as covariates. Interaction terms among selected covariates were evaluated for inclusion as well. The mean difference and corresponding 98.75% CI limits were exponentiated to obtain the GMC ratio and the corresponding 98.75% CI. Non-inferiority was demonstrated if the lower confidence limit of the 98.75% CI of the GMC ratio (GMC S /GMC C ) was larger than 0.5 (non-inferiority margin) (where s refers to each study vaccine group and c refers to the comparator group).
The secondary endpoints, GMC of anti-PT, anti-FHA, antitetanus and anti-diphtheria, and GMT of PT neutralizing antibody at baseline and 28 days following immunization in pregnant women, were calculated for each vaccine group along with twosided 95% CI. The ratio of the GMC or GMT in each of the ap1 gen / Tdap1 gen /Tdap2 gen /TdaP5 gen groups to that in the Tdap8 chem group and a two-sided 95% CI for the ratio were obtained using methods similar to those outlined above. Seroconversion rates for anti-PT and anti-FHA antibody and PT neutralizing antibody, and seroprotection rates for anti-DT and anti-TT antibody were computed along with the corresponding exact two-sided 95% CI based on the Clopper-Pearson method for each vaccine group. The differences in the seroconversion rates between each of ap1 gen /Tdap1 gen / Tdap2 gen /TdaP5 gen groups and the Tdap8 chem group were calculated along with the two-sided 95% CI obtained by the Miettinen and Nurminen method. All immunogenicity analyses were done in both full analysis (FA) and per protocol (PP) populations with the PP population as the primary analysis population.
For safety analysis, the proportion of participants with safety endpoints at different study time points along with an exact two-sided 95% CI were calculated using the Clopper-Pearson method. A pair-wise comparison in the percentage of participants with safety endpoints between each of the ap1 gen /Tdap1 gen / Tdap2 gen /TdaP5 gen groups and the Tdap8 chem group was performed using Fisher's exact test. All safety analyses were done in the safety population.
All statistical analyses were performed using SAS Ò software version 9.4.

Results
This paper reports the findings from the initial phase of the study up to Day 28. The immunogenicity results from the PP population are presented. The disposition of the pregnant women meeting the protocol requirements is shown in Fig. 2a. Between January 4, 2019, and October 10, 2019, a total of 412 pregnant women were screened, of whom 400 (200 from each study site) were enrolled and randomized into five vaccine groups (80 pregnant women per vaccine group). All enrolled pregnant women were included in the safety analysis (Safety population, N=400). The PP population for immunogenicity analysis included 394 (six pregnant women were excluded as one subject did not have blood sample collection and five had at least one major protocol deviation that might affect the immunogenicity results).
To assess the primary objective of the present study, we added the PP population from women of childbearing age (250 women consisting of 50 in each vaccine group) to the PP population of pregnant women (394 pregnant women). Hence, the pooled population included 644 participants in total (Fig. 2b). For other analyses outside the primary analysis (non-inferiority test), only data from pregnant women were used.
Demographic characteristics (age, ethnicity, height, and body weight) at baseline were similar for participants in all vaccine groups ( Table 2). Adjusted GMC ratios of the anti-PT antibody of each group with the comparator, Tdap8 chem at Day 28 (Table 2) are as follows: 1.7 (98.75% CI 1.3-2.3), 0.9 (98.75% CI 0.7-1.3), 1.2 (98.75% CI 0.9-1.5), and 2.6 (98.75% CI 2.0-3.5) for ap1 gen , Tdap1 gen , Tdap2 gen and TdaP5 gen , respectively. For the analysis of pooled data, the adjusted GMC was obtained by ANOVA adjusted for concentrations at baseline, study, and interaction between vaccine group and study. In this analysis, which included pregnant women and women of childbearing age, the primary outcome of noninferiority was met. Moreover, the lower bound of the 98.75% CI of the adjusted GMC ratio was greater than one for ap1 gen and TdaP5 gen , indicating that both vaccines could be considered superior to Tdap8 chem in inducing anti-PT antibody response.

Other endpoint analysis at Day 28 post-vaccination in pregnant women only
At baseline, the GMC of anti-PT antibody was low and similar in all maternal vaccine groups. On Day 28, the anti-PT antibody GMC change from baseline was highest for the TdaP5 gen group followed by the ap1 gen group, and the lowest for the Tdap8 chem group. Day 28 adjusted GMC ratios between the study vaccines and comparator (Tdap8 chem ) ranged from 0.9 (98.75% CI 0.7-1.3, Tdap1 gen ) to 2.9 (98.75% CI 2.0-4.1, TdaP5 gen ). These results indicated that all study vaccines (ap1 gen , Tdap1 gen , Tdap2 gen , and TdaP5 gen ) were at least as effective as the comparator (Tdap8 chem ) in inducing anti-PT antibody responses in pregnant women (Table 2).
Similarly, at baseline, the GMT of PT neutralizing antibody was low and similar in all groups. On Day 28, the GMT change from baseline was highest for the TdaP5 gen group followed by the ap1 gen group, and the lowest for the Tdap8 chem group. The Day 28 adjusted GMT ratio between the study vaccines and comparator (Tdap8 chem ) ranged from 0.9 (95% CI 0.5-1.6, Tdap1 gen ) to 3.0 (95% CI 1.7-5.3, TdaP5 gen ). Thus, all study vaccines containing PT gen  were shown to be at least as effective as the comparator vaccine (Tdap8 chem ) in inducing functionally active PT-neutralizing antibody in pregnant women ( Table 2). The GMC of anti-FHA antibody at baseline was similar in all groups. On Day 28, the change in GMC from baseline was highest in the Tdap8 chem group, followed by TdaP5 gen , and lowest in the Tdap1 gen group. The Day 28 adjusted GMC ratios between the study vaccine and the comparator (Tdap8 chem ) ranged from 0.3 (95% CI 0.3-0.4, Tdap1 gen ) to 0.7 (95% CI 0.5-0.8, TdaP5 gen ) ( Table 2).
The proportion of pregnant women with ! 4-fold increase in antibody concentration to PT and FHA at Day 28 with respect to baseline (seroconversion rates) ranged from 76.9% to 97.4% for PT in Tdap8 chem and TdaP5 gen respectively, and from 73.8% to 94.9% for FHA in Tdap1 gen and TdaP5 gen , respectively compared to 88.5% in Tdap8 chem (Fig. 3).
For PT neutralizing antibody, the seroconversion rate was 100% for TdaP5 gen , and 69.6% for ap1 gen , compared to 70.8% in the Tdap8 chem group (Fig. 3).
Reverse cumulative distribution curves (RCDCs) for anti-PT IgG, anti-FHA IgG, and PT neutralizing antibody at baseline and Day 28 after vaccination are shown in Appendix Fig. 1.
Participants vaccinated with the formulations containing tetanus (TT) and diphtheria (DT) toxoid antigens all showed satisfactory tetanus and diphtheria responses at Day 28, and the anti-TT and anti-DT antibody GMC changes from baseline were similar for all vaccines. The anti-TT and anti-DT responses elicited by the study vaccines were similar to those elicited by the comparator, Tdap8 chem except for the anti-TT antibody GMC in TdaP5 gen and the anti-DT antibody GMC in Tdap1 gen where the 95% CI of the adjusted GMC ratios did not include unity (Appendix Table 3).
The TT antibody concentration was !0.1 IU/mL (equated with seroprotection) in more than 96% of participants at baseline, rising to 100% by Day 28. For DT, the proportion of participants with antibody concentration !0.1 IU/mL (equated with seroprotection) was doubled from 46%, rising to 92% by Day 28 after vaccination (Fig. 3).
The solicited local and systemic reactions are presented in Appendix Table 4. Pain at the injection site was the most common solicited local reaction for all vaccine groups (72.5% for ap1 gen , 80.0% for Tdap1 gen and Tdap2 gen , 82.5% for TdaP5 gen , and 88.8% for the comparator, Tdap8 chem ). Most of these cases were mild and lasted for a few days. Myalgia was the most commonly reported solicited systemic reaction in all vaccine groups. The proportion of pregnant women with myalgia was similar in Tdap2 gen and TdaP5 gen groups (36.3% and 35.0%, respectively) and Tdap8 chem (36.3%). However, fewer pregnant women reported myalgia in the ap1 gen (15.0%) and Tdap1 gen (25.0%) groups. Most systemic reactions were transient and mild or moderate in severity.
A total of 120 unsolicited AEs were reported by 20.3% of participants up to Day 28 following vaccination (Appendix Table 5). The incidence was similar among vaccine groups.
Overall, five SAEs (four cases of preterm labor and one case of vaginal bleeding) were reported up to Day 28 following vaccination. None of these SAEs were considered vaccine-related (Appendix Table 5).

Discussion
Our study focused on the induction by pertussis booster vaccines of anamnestic anti-PT response to obtain antibody levels sufficiently high to likely protect newborns against severe pertussis in the first months of life (data will be presented later in a separate publication). The antibody level of our study formulations has elicited similar or higher levels than the comparator in pregnant women at day 28 post-vaccination which was previously shown to protect infants in a study conducted by Amirthalingam et al (2014) [6].
Findings from this phase 2 trial indicate that recombinant genetically inactivated acellular pertussis vaccines may be suitable for use in immunizing pregnant women at a range of doses. Not only were they non-inferior in terms of immunogenicity to the chemically inactivated pertussis vaccine licensed for this use in the pooled population of pregnant women and women of childbearing age (Table 2), they also showed acceptable safety profiles (Appendix Tables 4 and 5). In addition, the ap1 gen and TdaP5 gen vaccines generated significantly higher antibody responses than the Tdap8 chem vaccine. The high immunogenicity of ap1 gen , a low-dose stand-alone vaccine, is compatible with its use as a booster with an acceptable safety profile in pregnant women where the benefit of the additional tetanus and diphtheria toxoids is not required, analogous to the current use of aP5 gen (Pertagen Ò , a  Adjusted GMC was obtained by ANOVA adjusted for concentrations at baseline, study, and an interaction between vaccine group and study. 2 The ratio of adjusted GMC between each study group (ap1 gen /Tdap1 gen /Tdap2 gen /TdaP5 gen ) and comparator group (Tdap8 chem ) for non-inferiority test. The lower limit of the 98.75% CI of the adjusted GMC ratio was compared to the non-inferiority margin of 0.5.

3
PT: The ratio of adjusted GMC between each study group and comparator group with 98.75% CI was calculated based on ANCOVA model, adjusted for concentrations at baseline. FHA: The ratio of adjusted GMC between each study group and comparator group with 95% CI was calculated based on ANCOVA model, adjusted for concentrations at baseline and study site. PT neutralizing: The ratio of adjusted GMT between each study group and comparator group with 95% CI was calculated based on ANCOVA model, adjusted for concentrations at baseline and interaction between baseline and vaccine group. *The adjusted GMC or adjusted GMT between study group and comparator group is significantly different. Abbreviations: GMC, geometric mean concentration for PT and FHA; GMT, geometric mean titer for PT neutralizing; GMFR, geometric mean fold rise from baseline; CI, confidence interval. In this study, we found that 57% of pregnant women were seronegative for PT-IgG antibody at baseline which is in line with another study conducted in Thai pregnant women [18]. The protection of newborn infants via the transfer of maternal antibodies through the placenta and breastmilk would be inadequate without maternal booster immunization [19][20][21]. The substantial enhancement of maternal antibodies by a single booster dose of any of the vaccines studied here was comparable to our previous findings with the same vaccines in women of childbearing age [11] and other recent studies of Tdap8 chem vaccination in pregnancy [22][23][24][25]. Notably, we found no major difference in PT-IgG response and serum PT-neutralizing activity between vaccination during the second versus the third trimester of pregnancy when the responses were measured at 28 days after vaccination (Appendix Fig. 2).
For both seroresponse rates and GMC/GMT for PT antibody and PT neutralizing antibody, our study formulations have elicited similar or higher levels than the comparator. The higher levels were observed with Tdap2 gen and TdaP5 gen but more interestingly also with ap1 gen , which contains only one microgram of PT gen .
For FHA antibody, ap1 gen , Tdap2 gen and TdaP5 gen vaccines have induced similar or higher seroresponse rate than the comparator whereas it was found to be lower with Tdap1 gen . Anti-FHA antibody GMCs are also lower in our study vaccines compared to Tdap8 chem . However, the contribution of anti-FHA to protection against severe pertussis illness likely principally mediated by PT is expected to be limited [26]. This supports the fact that monocomponent vaccines containing only PT have been shown to be sufficient to protect infants from severe pertussis [27][28]. In addition, an infant baboon model had shown that anti-PT alone was sufficient to prevent neonatal pertussis disease but not infection in animals intratracheally challenged with a large dose of pathogenic Bordetella pertussis [29].
One limitation of this study in pregnant women is the small sample size. However, pooling of this study with the study in women of childbearing age provided adequate power for a noninferiority comparison. The pooling was considered appropriate given the similar study methods, populations, immunogenicity testing in the same laboratory with the same validated assay, and adjustment of responses based on baseline titer, cohort, and treatment. Another limitation of this study is the fact that the immune system in pregnant women behave differently compared to non-pregnant women. By pooling the data, the effect of pregnancy on immune response will be decreased.
More information on pertussis immune response at birth and in infants will be needed before generalized conclusions ranking the responses to the different vaccine formulations can be drawn. Additional data from later stages/phases of the same study are forthcoming to assess the impact of the timing of maternal vaccination (second versus third trimester of pregnancy) on IgG placental transfer to the infant and to explore the impact of maternal antibodies on infant responses to primary immunizations, including to pneumococcal conjugate vaccine (PCV).
More extensive studies with the selected study vaccines should be then considered in pregnant women in other populations, in diverse countries, ethnicities, and social contexts. Author contribution BLI, NB, SM, YT and PHT conceived and designed the study. KC, TP and SM implemented the study. TP and KC were the principal site investigators. SC, SAg, CC, SAw, KL, SR and MT conducted the study with support from LF, CK and VY. IAL and RH supported with project management of the funder. All authors contributed to data interpretation. All authors reviewed the drafts and approved the final version of the manuscript for submission.   Data availability

CRediT authorship contribution statement
Data will be made available on request.

Declaration of Competing Interest
The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Librada      - 1 The GMC of anti-TT and anti-DT IgG were calculated with 95% CIs at baseline and 28 days after vaccination. 2 The adjusted Day 28 GMC ratio between each study group (Tdap1 gen /Tdap2 gen /TdaP5 gen ) and comparator group Tdap8 chem with 95% CI were calculated based on ANCOVA model, adjusted for concentrations at baseline and subject age for anti-TT IgG analysis and adjusted for concentrations at baseline only for anti-DT IgG analysis. 3 The 95% CI of the adjusted GMC ratio is 0.63-0.98.

Table A4
Proportion of pregnant women reporting local and systemic reactions within 7 days after vaccination.  Abbreviation: CI, confidence interval.

Table A5
Incidence of unsolicited adverse events in pregnant women within 28 days after vaccination, by vaccine group.