Infant immunization programs using combination vaccines have successfully eradicated or reduced morbidity and mortality associated with childhood infections such as diphtheria, tetanus, polio, pertussis, and Haemophilus influenzae type b (Hib). In some federal countries, such as Canada and Germany, provinces and territories can determine which nationally recommended immunization programs recommended are implemented. As a result, there may be significant diversity in childhood immunization schedules across the country.1 As well, lack of vaccine registries across most of the country means that information on previous vaccine receipt may not be available when a child presents for a vaccination.
Two pentavalent infant vaccines are authorized in Canada to provide protection against diphtheria, pertussis, tetanus, polio, and Hib beginning at 2 months of age. Pediacel (Sanofi Pasteur) contains 5 acellular pertussis antigens (pertussis toxin [PT], filamentous hemagglutinin [FHA], pertactin [PRN], and fimbriae 2 and 3 [FIM]), and Infanrix-IPV-Hib (GlaxoSmithKline Biologicals) contains 3 pertussis components (PT, FHA, PRN) as well as differing amounts of alum adjuvant (Table 1). Although only one of these products is currently marketed and widely used, use of both approved vaccines in future infant immunization programs could result in infants receiving both vaccines during the infant series if they move between provinces. Interchangeability of these vaccines for the toddler booster dose has been shown,2 but it is not known if mixed primary infant schedules are acceptable. In this randomized, controlled trial, the serologic correlates of immunity and short-term adverse events (AEs) after 2 mixed schedules of these infant vaccines given as part of the primary immunization series were compared.
MATERIALS AND METHODS
This was a phase 3, prospective, randomized (1:1), double-blinded trial evaluating the safety and immunogenicity of 2 mixed primary schedules. It was conducted at 2 Canadian sites (Halifax, Nova Scotia and Montreal, Quebec) between May 2010 and January 2011.
Eligible children were aged 6 weeks (42 days) to 2 months inclusive (up to 1 day before 3 months of age) on the day of enrolment and born at full term (defined as ≥37 weeks, 0 days). Parents/guardians indicated availability to attend all scheduled visits and comply with study procedures, were able to read and write in English or French, and had access to a telephone.
Exclusionary criteria were participation in another clinical trial in the 4 weeks preceding the first trial vaccination or during the current trial period, personal history of immunodeficiency or immunosuppressive therapy, systemic hypersensitivity to any of the vaccine components, chronic illness that could interfere with trial conduct or completion, receipt of blood or blood-derived products since birth, previous vaccination with vaccines other than those recommended as part of the infant schedule, previous vaccination with any DTaP or whole-cell pertussis-based combination vaccine, Hib conjugate, or poliovirus vaccines, coagulation disorder contraindicating intramuscular vaccination, clinically significant findings on review of systems, developmental delay, or neurologic disorder or history of receipt of H. influenzae, diphtheria, tetanus, pertussis, or poliovirus disease.
The protocol was approved by Research Ethics Boards at each participating institution, and by Health Canada. The study was conducted according to Good Clinical Practice. A parent/guardian provided informed written consent for each subject.
Participants were randomized 1:1 to 1 of 2 primary infant immunization schedules with the 2 DTaP-IPV-Hib vaccines (Table 1). Participants in group 1 received 2 doses of Pediacel at 2 and 4 months of age followed by 1 dose of Infanrix-IPV/Hib at 6 months (Pediacel, Pediacel, Infanrix [PPI]). Participants in group 2 received one dose of Infanrix-IPV/Hib at 2 months of age followed by 2 doses of Pediacel at 4 and 6 months (Infanrix, Pediacel, Pediacel [IPP]). Because Pediacel is currently the vaccine used in public programs, mixed schedules in which this vaccine comprised most of the doses were chosen. Resource limitation prevented study of all possible schedules (eg, IIP, PII, PPP, III).
On Day 0, before vaccine administration, a brief history-directed physical examination was performed, and a venous blood sample (3–5 mL) was obtained. Vaccines were administered intramuscularly in the leg (left preferred). Concomitant pneumococcal conjugate vaccine (Prevnar 7, Pfizer, New York, NY) was given in the other limb. Children were monitored at the study site for 30 minutes after each of the study vaccine injections. Parent(s)/guardian(s) were provided with and instructed on the use of a digital thermometer, ruler, and paper memory aide to record AEs at each vaccination visit.
On days 2 and 3 and day 8 after each vaccination, study personnel contacted the family by telephone to review information collected on the memory aide regarding AEs and use of concomitant medications. The second and third vaccinations occurred at 4 and 6 months of age, respectively. At the study visit at 7 months of age (28–42 days after the third dose of the primary series for both groups), a venous blood sample (3–5 mL) was obtained, and a brief history-directed physical examination was performed.
Serological outcome measures were assessed on day 0 and at 7 months of age. The outcomes were seroprotection rates for antibodies against Hib PRP—defined as percentage of subjects with antibody concentrations ≥0.15 and ≥1.0 μg/mL, and geometric mean concentration (GMC) for antibodies against PT, FHA, PRN, and FIM. The proportions of participants achieving ≥4-fold rise in antipertussis antibody concentrations (4FAR; post-dose 3/pre-dose 1) were also calculated.
Safety and Reactogenicity
Safety analyses included the frequency of solicited injection-site reactions (tenderness, erythema, swelling) and solicited systemic AEs (fever, vomiting, abnormal crying, drowsiness, decreased appetite, irritability) within 8 days (Day 0–Day 7) after each vaccination. The occurrence, nature, time to onset, duration, severity, and relationship to vaccination of unsolicited AE occurring within 31 days of each vaccination and of any serious AE during the entire study period for all groups were also determined.
Statistical Considerations and Analysis
The primary exploratory null hypothesis was that the proportion of subjects with anti-PRP (Hib) and antipertussis immune responses in group 2 (IPP) was nonequivalent to the proportion of subjects with an immune response in group 1 (PPI). With 125 subjects per group, there was power >70% to declare equivalence between the 2 schedules if the common proportion of subjects with an immune response in groups 1 and 2 was greater than 0.7, and power >80% to declare equivalence if the common proportion was greater than 0.8. The exploratory hypothesis set α = 0.05 as the level of significance, with a nonequivalence margin δ = .15, assuming a common value P1 = P2 = p. The proportions of subjects achieving anti-PRP titers ≥0.15 and ≥1.0 μg/mL were calculated by group at pre-dose 1 and post-dose 3, and compared between groups. GMCs of pertussis antibodies (PT, FHA, PRN, and FIM) at prevaccination and at 1 month after third dose were calculated by group, and compared using confidence intervals on the difference of means. The proportion of vaccine responders at 1 month after third dose was also calculated.
For the within-group analysis of proportions, binomial point estimates and exact binomial confidence intervals were calculated. Two-sided 95% confidence intervals for the difference of proportions were used to compare groups. For the within-group analysis of continuous variables, point and interval estimates of means were calculated. For the difference of means, 95% confidence intervals were constructed. The exploratory hypotheses were tested separately for each outcome measure, with no adjustment for multiple comparisons.
Of 253 infants enrolled, 127 infants were randomized to IPP and 126 to PPI. All infants received a first dose of vaccine and were included in the intention-to-treat analysis (safety). Two hundred forty-four infants completed the study (96.4%) and were considered the per-protocol cohort for immunogenicity. Three children were withdrawn because of protocol noncompliance (inability to obtain blood; PPI = 1, IPP = 2), 4 voluntarily withdrew (all from IPP), and 2 children were lost to follow-up (1 from each group). Males comprised 54.96% of the sample, and 85.8% were Caucasian.
Anti-PRP responses: The percentage of children with PRP antibody ≥0.15 μg/mL after completion of the 3-dose infant immunization schedule was higher in the IPP group than in the PPI group (98.3%, 95% CI: 94.1, 99.8 vs. 86.1%, 95% CI: 78.6, 91.7, P < 0.001; Table 2). The percentage of children with anti-PRP antibody ≥1.0 μg/mL, 4FAR, and GMC were also statistically significantly higher in the IPP group (Table 2).
Anti-PT GMC and anti-FIM 4FAR and GMC were statistically significantly higher in the IPP than in the PPI group (Table 3); anti-PT 4FAR was not different between schedules. By contrast, higher anti-FHA 4FAR and GMC occurred after PPI than IPP. No difference between groups was observed in anti-PRN antibody responses.
Reactogenicity and Safety
No withdrawals occurred because of AEs or severe AEs from either group. There was no difference in the frequency of unsolicited AEs related to vaccination (PPI 20.4 vs. 22.8, ns, 95% CI: 15.1, 26.5 vs. 16.3, 30.4), and no serious unsolicited AEs occurred. The time to onset of any overall solicited reaction was not different between groups after each vaccination.
The frequency of any erythema, swelling, or tenderness is seen in Table 3; moderate-to-severe reactions were not different between groups at any time point for these individual outcomes. Swelling was more common in the PPI group after the third dose of the primary series (26.4% vs. 10.7%, P = 0.002), as was the frequency of any tenderness (35.2% vs. 20%, P = 0.01). Any local moderate-to-severe reaction at 2 months of age (first dose of the series) was more common in children receiving Pediacel compared with Infanrix (15.9% vs. 7.1%, P = 0.031).
Systemic reactions were similar for the 2 vaccine schedules (Table 4). Irritability was more common after the third dose in the PPI group compared with the IPP group (67.2 vs. 51.6, P = 0.014). Mild crying and overall systemic reactions (any intensity) were also more common in the PPI group after the third dose (35.2% vs. 23.0%, P = 0.037) and (80.0 vs. 68.0, P = 0.042), respectively.
Children in both groups had a higher frequency of fever after the second immunization than the first, whereas rates of other systemic reactions were more consistent across the 3 doses.
In countries where multiple vaccine products are available, the possibility of completing a vaccine series with an alternate product exists. The infant primary immunization with DTaP-IPV-Hib-containing vaccines consists of 4 doses, the first 3 of which are considered priming. In federal countries with mobile populations, an infant could easily move between jurisdictions and be exposed to a mixed vaccine schedule. If a vaccine registry were established, health care providers could easily determine which product a child received previously. In Canada, only a few provinces have the ability to track vaccine product receipt in this way,1 and there is variation in vaccine schedules across provinces, suggesting that a child could have reduced immunogenicity to the priming series compared with a child who received all 3 immunizations with the same product.
Although the 2 vaccines evaluated in this study have been shown to have comparable immunogenicity and reactogenicity when given as an 18-month booster dose after primary immunization with Pentacel,2 the interchangeability of these vaccines during the primary infant immunization series had not been previously reported. Clearly, this study indicates that mixed schedules during the primary infant immunization series may not produce similar immunogenicity and reactogenicity. Product monographs recommend that the primary immunization schedule be completed with the same vaccine where possible, as does the American Academy of Pediatrics.3 In Canada, the National Advisory Committee on Immunization has acknowledged that factors may necessitate giving different product to the same individual over time, and that deferral of vaccine may result in a missed vaccination opportunity.4 The National Advisory Committee on Immunization provides the immunizer with principles to consider when making decisions about the interchangeability of vaccines. These include that the vaccines should be approved with the same indications, specified for the same population, and be equally acceptable in terms of safety, reactogenicity, immunogenicity, and efficacy, and a regularly scheduled primary or booster vaccine should not be deferred because of lack of availability of a particular product.4
Although combination childhood vaccines reduce the number of injections or administrations required at a single health care visit, they may be associated with some disadvantages, including immunologic interference. Differing immune responses may result from the different constituents of the 2 combination vaccines, as well as the particular conformational epitope of each antigen that is presented to the immune system as a result of differing manufacturing processes.5
Rates of AEs observed in this study, although different between groups, are comparable with those observed in other infant schedules.6 Some differences in rates were observed at the point where children were exposed to a new vaccine with different constituents (Table 1); higher reactogenicity in this setting is not entirely surprising. However, other studies have shown that repeated doses of the same product may be associated with more reactogenicity than with introduction of a new product. Previous studies of DTaP-IPV-Hib-containing vaccines using a single product indicate that more than 95% of children have protective.3,7
There are several limitations to this study. Serologic correlates of immunity were evaluated; therefore, it is not certain that clinical protection against these infections would be altered with mixed schedules since surrogate outcomes may not always predict clinical results. Serologic correlates of clinical protection against invasive Hib disease are generally accepted; there is less unity of thought regarding the serologic correlates of protection against pertussis.8 As well, no comparator with a single vaccine schedule was used (eg, PPP or III); therefore, conclusions about the efficacy or reactogenicity of schedules using the individual combination vaccine products cannot be made.
In summary, some AEs were more common (irritability, crying, any local AE) with 1 of 2 mixed primary infant immunization schedules, and reduced immunogenicity to some pertussis antigens and to Hib antigens was observed. We conclude that it is preferable to complete the primary infant 3-dose vaccine series with the same vaccine, rather than considering infant vaccines as interchangeable. To implement this recommendation, the health care provider would need access to complete data on the child's immunization history, such as that provided by a national vaccine registry.