Lot-to-lot Consistency, Safety, Tolerability and Immunogenicity of an Investigational Hexavalent Vaccine in US Infants

Abstract

Background: 

This multicenter phase III study (NCT01340937) evaluated the consistency of immune responses to 3 separate lots of diphtheria–tetanus toxoids–acellular pertussis 5, inactivated poliovirus vaccine, Haemophilus influenzae type b, and hepatitis B (DTaP5-IPV-Hib-HepB), an investigational hexavalent vaccine (HV).

Methods: 

Healthy infants were randomized (2:2:2:1) to receive HV or Pentacel (Control). Groups 1, 2 and 3 received HV at 2, 4 and 6 months, and Control at 15 months. Group 4 received Control at 2, 4, 6 and 15 months, plus Recombivax HB (HepB) at 2 and 6 months. Concomitant Prevnar 13 was given to all groups at 2, 4, 6 and 15 months; pentavalent rotavirus vaccine (RV5) was given to all groups at 2, 4 and 6 months. Blood specimens (3–5 mL) were collected immediately before administration of dose 1, postdose 3, immediately before toddler dose, and after toddler dose. Adverse events were recorded after each vaccination.

Results: 

The 3 manufacturing lots of HV induced consistent antibody responses to all antigens. Immunogenicity of HV was noninferior to Control for all antibodies, except for pertussis filamentous hemagglutinin geometric mean concentration postdose 3, and pertussis pertactin (PRN) geometric mean concentration after toddler dose. Postdose 3 immunogenicity of concomitantly administered Prevnar 13 was generally similar (except for serotype 6B) when given with HV or Control. Adverse events of HV were similar to Control, except for a higher rate of fever ≥38.0°C [49.2% vs. 35.4%, estimated difference 13.7% (8.4, 18.8)].

Conclusions: 

HV demonstrated lot-to-lot manufacturing consistency; safety and immunogenicity were comparable with the licensed vaccines. HV provides a new combination vaccine option within the US 2-month, 4-month and 6-month vaccine series.

The recommended immunization schedule for children in the United States protects infants and toddlers against 14 different diseases.¹ The current combination vaccines such as Pediarix [diphtheria–tetanus toxoids–acellular pertussis 3, hepatitis B and inactivated poliovirus vaccine (DTaP3–HepB–IPV), GlaxoSmithKline, Rixensart, BE] and Pentacel [diphtheria–tetanus toxoids–acellular pertussis 5, inactivated poliovirus vaccine and Haemophilus influenzae type b (DTaP5–IPV/Hib), Sanofi Pasteur, Swiftwater, PA] reduce the vaccine schedule from 25 separate injections [27 or 28 separate vaccinations, including Rotarix (RV1, GlaxoSmithKline, Rixensart, BE) or RotaTeq (RV5, Merck & Co., Inc., Kenilworth, NJ)] in the first 2 years of life to a possible 18 injections (20 separate vaccinations).² The high rate of injections during these early years contributes to vaccine hesitancy and the perception that infants receive “too many shots too soon.”³ In response, many parents request alternative vaccination schedules, which lead to deferred and delayed doses and decreased coverage rates.^4–6 Moreover, the crowded vaccination schedule makes it difficult to add new vaccines with the potential for incremental health benefits to young children. For all of these reasons, the Advisory Committee on Immunization Practices,⁷ the American Academy of Pediatrics⁸ and the American Academy of Family Physicians⁹ each endorse the use of combination vaccines.

Higher valency combination vaccines have the potential to mitigate the number of injections, and have been shown to improve coverage and timeliness.^10–12 The investigational hexavalent vaccine (HV; DTaP5–IPV–Hib–HepB) is a fully liquid combination vaccine directed against 6 diseases. These are the results from a pivotal study (NCT01340937) in the US phase III program, assessing the lot-to-lot consistency, safety, tolerability and immunogenicity of HV compared with Pentacel (DTaP5–IPV/Hib; Control vaccine; Sanofi Pasteur, Swiftwater, PA) plus Recombivax HB (HepB; Merck & Co., Inc., Kenilworth, NJ), when administered concomitantly with Prevnar 13 (PCV13) and RV5.

METHODS

Population

Healthy infants 46–89 days old who had previously received 1 dose of hepatitis B vaccine (outside of the study) by 1 month of age were eligible for the study. Participants were excluded if they had (1) received or were expected to receive immunosuppressive agents; (2) received or expected to receive systemic steroids (greater than the equivalent of 2 mg/kg total daily dose of prednisone) since birth; (3) a history of cancer; (4) known or suspected hypersensitivity to any of the vaccine components; (5) received more than 1 dose of hepatitis B vaccine or a combination vaccine containing hepatitis B vaccine; (6) received any vaccines other than hepatitis B vaccine; (7) a febrile illness, or a rectal temperature ≥38.0°C (≥100.4°F), within 24 hours before enrollment; (8) a coagulation disorder contraindicating intramuscular (IM) vaccination; (9) a maternal or personal history of HBsAg seropositivity (e.g., chronic hepatitis B); (10) a history of invasive Hib disease, hepatitis B, diphtheria, tetanus, pertussis, poliomyelitis, rotavirus gastroenteritis or pneumococcal disease; or (11) any contraindication to the concomitant study vaccines. In accordance with principles of Good Clinical Practice, we obtained written informed consent from each participant’s parent or legal guardian before study entry. The protocol was approved by the human studies committees at each research site.

Vaccines

Table 1 shows characteristics of the vaccines used in this study. All products were prepared, packaged, labeled, stored and distributed in accordance with Good Manufacturing Practice, guidelines for Good Clinical Practice from The International Conference on Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use, as well as applicable local laws and regulations. HV was produced as a sterile, single-dose, liquid, preservative-free formulation. All vaccine doses were given intramuscularly, except for RV5, which was given orally.

TABLE 1: Composition of Investigational and Licensed Comparator Vaccines

Design

This was a randomized, partially double-blind, active comparator-controlled study conducted at 73 sites in the United States from May 2011 through July 2013. A total of 2808 healthy infants were randomized using a computer-generated, site-balanced allocation schedule to receive 1 of 3 lots of HV (lot A, lot B and lot C) or a Control vaccine (DTaP5–IPV/Hib) in a 2:2:2:1 ratio. Participants in the HV groups received HV, PCV13 and RV5 at 2, 4 and 6 months of age followed by Control and PCV13 at 15 months of age. Participants in the Control group received Control vaccine, PCV13 and RV5 at 2, 4 and 6 months of age, as well as HepB at 2 and 6 months of age, followed by Control vaccine and PCV13 at 15 months of age. Parents/legal guardians and study personnel were blinded to which of the 3 HV lots were administered to a given participant, but aware of assignment to HV versus Control vaccine.

It was planned that approximately 2400 subjects would be enrolled in the HV group (800 subjects in each lot group) and 400 subjects in the Control group. Assuming the evaluability was about 85% at postdose 3 and 80% after the toddler dose, the power for the primary lot-consistency hypothesis based on geometric mean titers (GMTs) was ~92% and the overall power for the secondary consistency hypothesis based on response rates was ~90%, both at the 1-sided 2.5% alpha level if the 3 lots of HV were equivalent. The study had >99% power for the secondary hypothesis of noninferiority regarding pertussis components at 1 month after toddler dose, 97.5% power for the secondary hypothesis of noninferiority for all HV endpoints at postdose 3, and ~97.9% power for the secondary concomitant hypothesis tests, all with a 1-sided 2.5% alpha level if the HV group was equivalent to the Control group.

Blood specimens (3–5 mL) used to assess immunogenicity were collected immediately before administration of dose 1, 1 month after completion of the infant series (dose 3), immediately before the toddler dose, and 1 month after the toddler dose.

From day 1 (day of vaccination) to day 5 following each vaccination, the following safety assessments were obtained by diary cards: temperature; solicited injection-site adverse events (AEs), including pain, tenderness, redness and swelling; and solicited systemic AEs, including pyrexia (fever), vomiting, abnormal crying, drowsiness, appetite loss and irritability. Reports of unsolicited injection-site and systemic AEs were collected through day 15 after each vaccination. Serious AEs (SAEs), defined as leading to hospitalization or death, were recorded from study entry up to 6 months after the last HV vaccination by diary, office visit or phone call.

Varicella, measles, mumps and rubella vaccines were permitted as long as they were given more than 30 days before or 30 days after any dose of study vaccine. Inactivated influenza vaccine was permitted more than 14 days before or 14 days after any dose of study vaccine.

Objectives

The primary objective was to evaluate the consistency of the postdose 3 HV immune responses to 3 manufactured lots of HV when given at 2, 4 and 6 months of age with respect to geometric mean concentrations (GMCs; concentrations for all antibodies except for polio) or GMTs (polio antibodies measured via titer).

The secondary objectives were to (1) evaluate the consistency of the postdose 3 antibody response rates to 3 manufactured lots of HV when given at 2, 4 and 6 months of age; (2) compare the immunogenicity of pertussis responses at 1 month after the toddler dose; (3) compare the postdose 3 immunogenicity of HV with Control vaccine; (4) evaluate the immunogenicity of PCV13 postdose 3 when administered concomitantly with HV; (5) describe the rates of fever, injection-site AEs, systemic AEs and overall safety profile of HV and the Control vaccines when given concomitantly with PCV13 and RV5; and (6) describe the incidence of SAEs up to 6 months after the last dose of HV or Control vaccine.

Additional objectives were to describe (1) the GMCs against all antigens in HV and Control vaccine after 3 doses; and (2) the response rates to all antigens except pertussis 1 month after the toddler dose.

Analyses

Immunogenicity

Antibody response rates were defined based on accepted immune correlates of protection, or previously accepted definitions of vaccine response for licensed vaccines (Table, Supplemental Digital Content 1, https://links.lww.com/INF/C591).¹³ The primary and key secondary endpoints, analysis populations and statistical methods for immunogenicity analyses are provided in Table, Supplemental Digital Content 2, https://links.lww.com/INF/C592. The per-protocol analyses included all participants who met the inclusion criteria, did not deviate from the protocol, and had serology results within the specified day ranges. The per-protocol, Original Windows population consisted of those participants in the per-protocol population who had vaccination windows of 46–74 days after the previous vaccination for dose 2 and dose 3, and blood draw sample windows of 28–44 days after the infant series (dose 3) or the toddler dose. The per-protocol, Revised Windows population consisted of those children in the per-protocol population who had vaccination windows of 42–84 days after the previous vaccination for dose 2 and dose 3, and blood draw sample windows of 28–51 days after the infant series (dose 3) or the toddler dose. The revised windows were prespecified in an amendment to the protocol and a statistical analysis plan before database lock and before knowledge of immunogenicity results. They were based on earlier phase II studies of HV and allowed for inclusion of immunogenicity data from more vaccinated participants in the per-protocol analysis. Key immunogenicity summaries and analyses were also provided for all endpoints associated with hypotheses using the full analysis set, which included all randomized participants with available serology data at postvaccination regardless of protocol deviation.

Safety

All randomized participants who received at least 1 dose of study vaccine and had safety follow-up were included in the safety analysis. The AEs and fever profile were described for the study vaccine after each vaccination and for the entire vaccination period. Incidence rates of each solicited AE occurring from days 1 through 5 after any infant dose, as well as incidence rates of each unsolicited AE from days 1 through 15 after any infant dose that occurred in more than 1% of the participants in either vaccination group, were compared using point estimates and 95% confidence intervals (CIs).¹⁴ Incidence rates of elevated temperatures (≥38°C) occurring from days 1 through 5 after any infant dose were compared using point estimates and 95% CIs.¹⁴ Other safety endpoints were summarized using frequency counts and percentages.

RESULTS

Results discussed in this section refer to the per-protocol, Revised Windows population; results based on the per-protocol, Original Windows and Full Analysis Set populations were consistent with the results for per-protocol, Revised Windows for all immunogenicity endpoints.

Participant Accounting and Demographics

As shown in Figure 1, 2285 (81.3%) randomized participants completed the study. Participants in both groups were comparable with respect to baseline characteristics (Table 2). The most common previous ancillary diagnoses were neonatal jaundice (37.4% of HV group, 35.2% of Control group) and gastroesophageal reflux disease (14.6% of HV group, 13.5% of Control group). The most frequently reported concomitant medications administered during the study were analgesics (63.6% of HV group, 60.6% of Control group).

TABLE 2: Participant Demographics and Baseline Characteristics

FIGURE 1: Participant disposition. Reasons for discontinuation: ^aDeath (2); lost to follow-up (20); physician decision (2); protocol violation (1); withdrawal of consent (33). ^bAE (1); lost to follow-up (14); physician decision (3); protocol violation (4); withdrawal of consent (32); other (1). ^cAE (3); death (3); lost to follow-up (12); physician decision (1); protocol violation (4); withdrawal of consent (30); other (1). ^dLost to follow-up (13); protocol violation (3); withdrawal of consent (15). ^eAE (1); lost to follow-up (29); protocol violation (5); withdrawal of consent (28); other (4). ^fOne subject discontinued postdose 3, but received the toddler dose at a nonstudy visit and completed study follow-up. ^gLost to follow-up (48); physician decision (1); protocol violation (3); withdrawal of consent (23); other (4). ^hAE (1); lost to follow-up (53); physician decision (1); protocol violation (6); withdrawal of consent (17); other (6). ⁱTwo subjects received the toddler dose at a nonstudy visit, and completed study follow-up. ^jLost to follow-up (23); protocol violation (3); withdrawal of consent (12); other (3). ^kLost to follow-up (6); withdrawal of consent (2); other (1). ^lLost to follow-up (14); withdrawal of consent (2); other (1). ^mLost to follow-up (7); withdrawal of consent (1); other (1). ⁿLost to follow-up (8); protocol violation (1); withdrawal of consent (1).

Immunogenicity

Lot consistency analysis regarding GMC and GMT (concentrations for all antibodies except for polio, which was measured via titer) based on an ANCOVA model at 1-month postdose 3 is provided in Figure 2 and Table, Supplemental Digital Content 3, https://links.lww.com/INF/C593. The 3 manufacturing lots (lots A, B and C) of HV induced similar GMCs and GMTs to all antigens contained in HV 1 month after the third dose of HV, when given concomitantly with PCV13 and RV5. The lower and upper limits of the 2-sided 95% CI of the GMC and GMT ratios between any 2 lots were within the equivalence margin (0.67–1.5), indicating lot consistency among the 3 manufacturing lots.

FIGURE 2: GMC or GMT ratio and 95% CI at 1-month postdose 3 (per-protocol, revised windows population).

GMCs of antibody responses to pertussis antigens, PRP and pneumococcal antigens after the infant series and for pertussis antigens after the toddler dose are given in Table 3. The GMC ratio (HV group/Control group) was above the prespecified noninferiority margin for each prespecified endpoint for all antibodies after the infant series except for filamentous hemagglutinin and pneumococcal (PN) 6B. The lower bound of the 2-sided 95% CI for the GMC ratio was above the noninferiority margin for all pertussis antibodies after the toddler dose except the GMC for the pertussis pertactin (PRN) antibody. The noninferiority margins for GMC ratio of the pneumococcal antibody concentrations were met for 12 out of the 13 serotypes. The percentage values of participants with pneumococcal antibody concentrations ≥0.35 μg/mL are presented in Table, Supplemental Digital Content 4, https://links.lww.com/INF/C594. The results show that the 95% CI for percentage of participants who achieved protective titers against pneumococcus overlaps between the PR5I group and Control group for all 13 serotypes.

TABLE 3: GMCs and Noninferiority Evaluation for Selected Antigens

Adverse Events

Safety follow-up was obtained for ≥99% of participants in each group. As seen in Table 4, 94.5% of the HV group and 92.4% of the Control group reported at least 1 AE after any dose in the infant vaccination series. In general, the proportion of participants reporting injection-site AEs (days 1–15), solicited injection-site AEs (days 1–5), systemic AEs (days 1–15) and solicited systemic AEs (days 1–5) was similar between both groups. For the infant series, the most notable exception was overall solicited reports of pyrexia (defined similarly to fever, but not counted in cases of intercurrent illness) on days 1–5, which were more common in the HV group (47.1%) than the Control group (33.2%). Most of these reports were mild to moderate in intensity and did not lead to medical intervention. During the entire study, 1 participant in the HV group and 1 participant in the Control group discontinued because of vaccine-related non-SAEs.

Table 4 also shows that at least 1 SAE after any infant series vaccination was reported by 3.8% and 3.5% of participants in HV group and Control group, respectively. There were 5 SAEs assessed as vaccine-related (all HV group): 1 case of ileocolic intussusception, considered related to third dose of RV5; 3 cases of fever, considered related to all study vaccines; and 1 case of diarrhea, considered related to first dose of RV5. No participants discontinued due to a vaccine-related SAE. During the entire study duration, 5 participants (0.2%) of 2308 participants in the HV group died: obstructive hydrocephalus (1 participant) 1 day postdose 2; death of unknown cause (1 participant) 44 days postdose 1; group A streptococcal sepsis (1 participant) 2 days postdose 1; and sudden infant death syndrome (2 participants) 10 days postdose 2 and 49 days postdose 1, respectively. No death was assessed to be related to the study vaccinations. None of the participants in the Control group (N = 402) died.

TABLE 4: Adverse Events After Any Infant Series Dose in All Participants Who Received at Least 1 Dose of Vaccine

Compared with the Control group, more participants in the HV group reported moderate fever (8.7% higher difference; statistically significant) during day 1 to day 5 after any infant vaccination (Table 5). Fever was of brief duration (lasting ≤2 days after vaccination for the vast majority). Fever was reported for 49.2% of participants in the HV group and for 35.4% of participants in the Control group. Convulsion or febrile convulsion was not reported in either group within 30 days postvaccination. A total of 3 participants (0.1%) in the HV group had vaccine-related SAEs of fever: 2 were participants hospitalized for diagnostic work-up of fever after the first dose of vaccine, and 1 had fever concurrent with group A streptococcal sepsis (2 days postvaccination; sepsis was not related to vaccination).

TABLE 5: Fever Summary (Any Infant Dose)

DISCUSSION

The DTaP5–IPV–Hib–HepB hexavalent vaccine, when administered as a 3-dose infant series to healthy infants at 2, 4 and 6 months of age, demonstrated consistent immunogenicity for all 6 antigens across 3 lots of vaccine. The secondary endpoints for HV noninferiority comparison to Control were similar to the primary endpoints for a completed US phase III study of HV.¹⁵ When compared with the Control group, postdose 3 HV responses met noninferiority criteria for each antibody, except for the filamentous hemagglutinin GMC comparison. Post-toddler dose, the pertussis responses for participants who had received an infant series of HV were noninferior to the responses for the Control group, except for the PRN GMC comparison [HV/Control GMC ratio 0.74 (0.66, 0.83)], which narrowly missed the noninferiority criteria for the lower bound to exceed 0.67. Although GMC for filamentous hemagglutinin after the infant series and for PRN after the toddler dose did not meet noninferiority criteria, the seroconversion rates were comparable.^16,17 The narrow noninferiority miss for PRN GMC after the toddler dose is also unlikely to be important, given the synergistic effect of the multiple pertussis antibodies.¹³

The administration of HV with other licensed infant vaccines showed a lack of immune interference for the majority of the PCV13 serotypes. The postdose 3 immunogenicity of concomitantly administered PCV13 was noninferior for 12 out of 13 serotypes. Only the 6B serotype did not meet our prespecified GMC noninferiority criteria (lower bound of the 95% CI around the GMC HV/Control ratio >0.67). However, these 6B responses (GMC ratio of 0.79; 95% CI 0.64, 0.96) would have satisfied GMC noninferiority criteria used in earlier pivotal study of 13-valent pneumococcal vaccine, which used a lower bound of GMC ratio >0.50 for each serotype.¹⁸ Additionally, the proportion of participants who achieved seroprotective concentrations (≥0.35 μg/mL) was not statistically different whether PCV13 was given with HV or Control vaccine for all 13 serotypes.

The safety profile of HV group was comparable with the known safety profile of its licensed component vaccines given to the Control group. HV was safe and well-tolerated, but did demonstrate a small but statistically significant higher incidence of moderate but brief fever compared with Control group. Nonetheless, a low rate of fever-related hospitalization (0.1%) and no seizures or febrile seizures within 15 days of vaccination were observed. Based on the results of this study, the higher rate of self-limited mild to moderate fever observed after HV vaccination as compared with Control vaccination is not considered clinically significant. As with other licensed vaccines, the safety profile of HV will be monitored on an ongoing basis after approval.

The approximately 6:1 ratio of HV:Control participants represents a limitation of the study as it influences the interpretation of case counts for rare events in the respective study groups. Although 5 deaths occurred in the HV group (0.2%) and 0 in the Control group (0%), the difference was not statistically significant (difference 0.2%; 95% CI: −0.7, 0.50). None of the deaths in the study were considered vaccine-related. In addition, the death rate in the HV cohort (N = 2232) is lower than the background rate of infant mortality in the United States which is approximately 6 per 1000 live births.¹⁹ Another limitation was the partially double-blind study design, which was open-label between HV and Control regimens.

The use of a fully liquid hexavalent combination vaccine should further improve timely adherence to the US vaccination schedule for infants by reducing injections and the corresponding pain, simplifying billing and record-keeping for clinicians, reducing administration errors, and easing storage and inventory issues.

ACKNOWLEDGMENTS

The authors would like to thank all of the study participants, their parents, the study investigators and their staff.

Following are the members of V419 Protocol 006 Study Group:

AL: W.H.Johnson; AR: T.D. Stewart; AZ: K. Concannon, K.G. Rouse, B.J. Silvey; CA: O.S. Basta, J.A. Hunter, J. Jaramillo, S. Khamis, N. Klein, J. Singh, P.J. Walsh; FL: M.F. Drusano, W.P. Lorentz, S. Padron; GA: W.P. Andrews, R.G. Tull; ID: R. Aguilar; KS: R.H. Egelhof, T.R. Naccarato; KY: S.L. Block; MI: R.L. Hines, L.G. Lello; MN: Al London; NE: A. Chatterjee, S. Russell; NC: G.L. Adams, E.R. Franklin, R.B. Scott; ND: J.K. Tillisch; NY: L.B. Weiner; LA: F.B. Hughes, T.G. Latiolas, J.A. Vanchiere; OH: S.J. Alter, B.E. Blue, J.S. Shepard; OK: S.E. Grogg; OR: J. Calcagno; PA: P. Cognetti, R.T. Kratz, K.R. McLelland, S. Shapiro; SC: M.L. Leonardi; TN: L.A. Harris-Ford; TX: O. DeValle, J.A. Fling, B.W. Nauert, K. Palanpurwala, K.W. Sarpong, R. Yetman; UT: M.J. Cornish, D.S. Davenport, M.P. Husseman, A.A. Gabrielsen, G.W. Schlichter; VA: L.D. Meloy, M.K. Saunders; WA: G. Bader, T.E. Crum, S.R. Luber.