Safety, Immunogenicity, and Induction of Immunologic Memory by a Serogroup C Meningococcal Conjugate Vaccine in Infants: Title and subTitle BreakA Randomized Controlled Trial FREE

Neisseria meningitidis has become the most common cause of childhood bacterial meningitis in the United Kingdom since the successful introduction, in 1992, of routine infant vaccination against Haemophilus influenzae type b (Hib) disease.¹ Three serogroups (A, B, and C) account for the majority of meningococcal disease worldwide. In developed countries serogroups B and C predominate, but in recent years in a number of countries, including England and Wales, the proportion of cases due to serogroup C has increased.² Around half of all cases of meningococcal disease occur in children younger than 5 years with the highest attack rates in children younger than 2 years.³ Despite advances in intensive care and heightened public and professional awareness about the disease, the case fatality rate remains greater than 10%.² It is likely that a significant impact on disease incidence and mortality will be achieved only through the development of effective vaccines.

Purified serogroup C polysaccharide vaccines have been available for many years⁴ but have limitations for routine use. They induce a T cell–independent antibody response that does not result in immunologic memory. The antibody response to purified polysaccharide vaccination is short lived in children younger than 2 years,^{5 - 6} and multiple doses in infancy may result in tolerance.^{7 - 8} Hyporesponsiveness has also been reported after multiple doses of meningococcal C polysaccharide in children between the ages of 1 and 3 years ⁹ and in adults.¹⁰ Recruitment of T-cell help by conjugation of polysaccharide to a protein carrier is thought to enhance immunogenicity and induce immunologic memory in infants.¹¹ This is the rationale behind the highly successful Hib conjugate vaccines, and the same technology is now being applied to several serogroups of N meningitidis. The oligosaccharide size and protein-to-polysaccharide ratio were shown to be important considerations in the immunogenicity of Hib conjugate vaccines.^{12 - 13} First-generation meningococcal A and C conjugate vaccines consisting of unsized oligosaccharides were shown to be immunogenic in Gambian¹⁴ and British¹⁵ infants and immunologic memory to the meningococcal C component was demonstrated in the Gambian cohort.⁸

We performed a single-center, double-blind, randomized controlled trial to determine whether the second-generation Chiron serogroup C meningococcal conjugate vaccine containing sized oligosaccharides is safe, immunogenic, and induces immunologic memory in infants. In addition, we wanted to determine whether the vaccine had any effect on the antibody responses to the simultaneously administered routine infant vaccines. Because of this and other studies, the UK government has embarked on a national meningococcal C vaccine program in which the meningococcal C conjugate vaccine will be part of the routine infant vaccines. In addition, a meningococcal vaccine will be offered to all others younger than 20 years.

Parents of infants born at the John Radcliffe Hospital, Oxford, England, between April and November 1995 were invited to participate in the study. Infants born before 37 weeks' gestation, weighing less than 2.5 kg at birth, with congenital abnormalities, or having had any vaccine previously were excluded. Written informed consent was obtained at the first visit. The study was approved by the Central Oxford Research Ethics Committee.

Infants were randomized, using a computer-generated random numbers list, to receive a 0.5-mL dose of 1 of 2 lots of meningococcal C conjugate vaccine (groups 1 and 2, n=60 in each group) or hepatitis B virus (HBV) vaccine (group 3, n=62) as a control vaccine (10 µg, Engerix B, SmithKline Beecham, Hertfordshire, England). The list was held in the pharmacy where the study vaccines were prepared as sterile injections and presented in identical syringes labeled with an identification number for each child. The study vaccine was given by intramuscular injection into the right leg within 8 hours of preparation. Routine vaccines coadministered in the left leg were diphtheria and tetanus toxoids and whole-cell pertussis (DTP) vaccines (DTP, Evans/Medeva, Surrey, England) reconstituted with Hib–tetanus conjugate (Act Hib, Pasteur Merieux, Berkshire, England). Polio vaccine was given orally (SmithKline Beecham). The UK primary immunization schedule of 2, 3, and 4 months of age was used with a window of 28 to 42 days between vaccinations. At 12 months, children within these groups received an intramuscular injection of either meningococcal C conjugate vaccine or meningococcal A and C polysaccharide vaccine (Mengivac [A and C], Pasteur Merieux) containing 50 µg of each meningococcal polysaccharide (MPS). Children were randomized to both the primary vaccine and the booster vaccine at the beginning of the study. The investigators and parents were blinded until completion of the study.

Blood was obtained at ages 2, 5, 12, and 13 months, before and 1 month (21-42 days) after the primary vaccination series and booster vaccination, respectively. Blood was also drawn before the second or third vaccination in children born on odd or even numbered days of the month, respectively.

Meningococcal C capsular oligosaccharides conjugated to cross-reacting materials₁₉₇ (CRM₁₉₇), a nontoxic mutant of diphtheria toxin, were manufactured in 2 lots (Chiron, SPA Siena, Italy). Each 0.5-mL dose contained 10 µg of meningococcal C oligosaccharide. They were designated meningococcal C-1 and C-2 conjugate vaccine and contained 20 µg and 13 µg of CRM₁₉₇, respectively.

Each child was examined by a pediatrician prior to vaccination at 2 months of age. Before each vaccination information about recent illness and the child's temperature were recorded. After each vaccination the study nurse observed infants for 30 minutes and recorded local and systemic reactions. Parents recorded their child's axillary temperature 6 hours after vaccination and then daily for 6 days, and they documented local and systemic reactions on a standardized diary card. The local reactions documented were tenderness, erythema, and induration. The systemic reactions sought were rash, change in eating habits, sleepiness, unusual cry, persistent cry, vomiting, diarrhea, irritability, and temperature of 38°C or higher. Any use of antipyretics was noted. Parents were given a 24-hour telephone number to call if they were concerned. They were also telephoned at 48 to 72 hours after each immunization to ensure that the infant was well and to remind parents to complete the diary card.

Serogroup C anticapsular antibody immunoglobulin (IgG) was measured by enzyme-linked immunosorbent assay (ELISA) at Chiron Corp, Emeryville, Calif, using a modified technique that selectively measures high avidity antibodies.¹⁶ Results are expressed in U/mL for which 1 unit is approximately equivalent to 1 µg of the Centers for Disease Control and Prevention's human reference standard (24.1 µg/mL). Serum bactericidal activity (SBA) was measured at Chiron Corp using a human complement source. Dilutions of test serum samples were started at 1:8. Antibodies to diphtheria and tetanus toxoids and Hib polyribosylribitol phosphate vaccine were measured by ELISA in the department of immunology, Churchill Hospital, Oxford, England, using methods described previously.^{17 - 18} Antibodies to pertussis (Filamentous haemaglutinin, Fimbriae, and 69 K) were analyzed at the Centre for Applied Microbiology and Research, Porton Down, England.¹⁹ Antibodies to polio were analyzed at the British National Institute for Biological Standards, Potters Bar, England, using standard methods and calibrated against the Second International Standard for antipoliovirus antibodies.²⁰

Sample size calculations were based on results obtained in Gambian infants with the first-generation meningococcal A and C conjugate vaccine.¹⁴ In that study healthy Gambian infants achieved an antimeningococcal C geometric mean concentration (GMC) of 2761 U/mL 1 month after their third immunization with a mean (SD) log₁₀ concentration of 3.441 (0.289) (80% confidence interval [CI], 0.258-0.330). Assuming an SD log₁₀ concentration of 0.330 with 60 evaluable postvaccination samples for each group, a 2-tailed t test with a significance of .05 would have 83% power to detect a 50% increase in GMC and a 33% decrease in GMC between the 2 conjugate groups. Using the same assumptions for the antibody responses to a booster vaccination at 12 months of age, with 30 evaluable postvaccination samples for each group, a 1-tailed t test with significance of .05 would have a greater than 95% power to detect a doubling in GMC between groups that have previously received the HBV vaccine and groups that have received the meningococcal C conjugate vaccines.

Comparisons of adverse effect rates between groups were analyzed using χ² or Fisher exact test. The χ² test for trend was used for comparisons between groups, and McNemar test for comparisons between legs of the routine (DTP/Hib) or study vaccines (meningococcal C conjugate vaccine or HBV vaccine).

Two-tailed probabilities were used for all significance tests. Geometric mean concentrations and their 95% CIs were calculated to describe the IgG ELISA and SBA responses for each group at each time point. Bactericidal titers lower than 1:8 were assigned a titer of 4, and ELISA antibody concentrations lower than 0.4 µg/mL were assigned a titer of 0.2 µg/mL for the analyses. Responses to vaccination were analyzed using 1-way analysis of variance (a second analysis adjusting for prevaccination titers from paired serum samples was also undertaken, which gave similar results). Where comparisons were made between only 2 of the groups, the Scheffe adjustment for multiple comparisons was used. Analysis was undertaken using STATA software.²¹

During the study, 1320 infants were eligible for inclusion, of which 182 were enrolled. Those whose parents did not return their letter of invitation were deemed to have refused participation. An informal analysis of reasons for nonparticipation was performed on a subset of families. Reasons included fear of pain or venipuncture, uncertainty about the long-term effects of a new vaccine, and unhappiness about having to make a decision soon after the birth of their infant. The majority of infants (97%) were white, 2 were Asian, and 3 were of mixed race. The demographic characteristics are shown in Table 1. There was an imbalance of sex between groups, but the groups were comparable in other respects. A secondary analysis controlling for sex did not substantially alter the results.

The study cohort is summarized in the patient flow diagram (Figure 1). One hundred seventy-nine children (98%) completed primary vaccination and evaluation at 5 months. One was withdrawn due to pyrexia, and 2 moved out of the area. Five randomization errors occurred. Four errors occurred at the first dose. These children completed the course with the same vaccine and were included in the final analysis according to the first dose given. One error occurred at the second dose and this child was excluded from the analysis. Other exclusions are detailed in Figure 1.

Safety data were analyzed for all doses administered. The vaccines were generally well tolerated. Local adverse effects are shown in Table 2. The respective rates of local reactions to the meningococcal C conjugate and HBV vaccines were very similar. Both study vaccines resulted in less tenderness (P<.001) and induration (P<.001) than the routine vaccines given in the opposite leg. One child in the control group given HBV vaccine was withdrawn from the study with a fever of higher than 40°C after the first injection. There was no significant difference in systemic reactions between any of the vaccine groups (data not shown).

Local and systemic reactions after booster vaccination are shown in Table 3. Parents of children who received MPS reported in the infants significantly more local tenderness, general irritability, and change in eating habits than those whose children received meningococcal C conjugate vaccine, regardless of the primary vaccination course. There was also an increased use of antipyretic medication in children who received MPS. There were no significant differences with respect to rash, sleepiness, unusual or persistent cry, vomiting, or diarrhea.

Response to Primary Vaccination. At 2 months of age, there was no significant difference in the IgG GMC between the 3 study groups (Table 4). The antibody levels in infants receiving either lot of meningococcal C conjugate vaccine increased progressively with each vaccination and after a single injection were already significantly higher than those in the control group (P<.001). There were no differences in antibody responses between the 2 meningococcal C conjugate vaccine lots at any time. Antibody levels had diminished by 12 months of age but remained significantly higher in the meningococcal C-1 and C-2 vaccine groups compared with those in the control groups (1.8 and 1.3 vs 0.22 U/mL, respectively; P<.001).

At baseline, SBA was determined only in a selected subgroup of serum samples with low ELISA titers. All had absent bactericidal activity (n=15). The SBA GMC levels increased progressively in children receiving either meningococcal C-1 or C-2 conjugate vaccines and after a single dose were significantly higher than the control group (P<.001). All subjects in the meningococcal C conjugate vaccine groups achieved a bactericidal titer of 1:8 or higher after primary immunization with 3 doses of vaccine, 98% after 2 doses, and 56% after 1 dose. At 12 months of age, the GMC had dropped in both study groups, but remained significantly higher than that in the control group (24 and 16, respectively, vs 4.2; P<.001), and more than 75% of children given meningococcal C conjugate vaccines in infancy had a titer of 1:8 or higher compared with 5% of those in the control group.

Response to Polysaccharide Booster Vaccination. Children in the meningococcal C conjugate vaccine groups achieved an IgG geometric mean bactericidal titer of 25 U/mL after receiving the MPS booster vaccination compared with 0.80 U/mL among those in the control group and a geometric mean bactericidal titer of 789 compared with 4.5 in controls (P<.001 for both comparisons) (Table 5). There was no significant increase in SBA titer among those in the control group receiving MPS as the primary meningococcal vaccination. After MPS booster vaccination, 97% of children receiving meningococcal C conjugate vaccine in infancy achieved a bactericidal titer of 1:8 or higher, compared with 11% of controls receiving HBV vaccine in infancy (P<.001).

Children aged 12 months receiving meningococcal C conjugate vaccine for the first time achieved an IgG GMC of 2.7 U/mL and a bactericidal titer of 15, a 3- to 4-fold higher response than those in the control group who received MPS (P<.001). Children both primed and boosted with either meningococcal C conjugate vaccine achieved a GMC of 60 U/mL and an SBA titer of 2400 or higher.

A single dose of meningococcal C conjugate vaccine at 12 months of age resulted in an SBA titer of 1:8 or higher in 66% of children compared with 100% of children both primed and boosted with meningococcal C conjugate vaccine.

All children achieved protective serum antibody concentrations after 3 doses of Hib (polyribosylribitol phosphate) (0.15 µg/mL), tetanus (0.01 IU/mL) and diphtheria (0.1 IU/mL) vaccines. There were no significant differences between groups with respect to the GMC for Hib, tetanus, pertussis, or polio serotypes 2 and 3 (data not shown). Those who received HBV vaccine had a higher GMC for polio serotype 1 than those who received either of the meningococcal C conjugates (29,299 vs 10,880, and 10,503 mIU/mL, respectively; P=.05), although all achieved titers 1:8 or higher.

At 5 months of age, infants in the meningococcal C conjugate vaccine groups had achieved significantly higher antidiphtheria antibodies than those who received HBV vaccine (7.7 and 11, respectively, vs 4.3; P<.001). At 12 months the titer had fallen, but remained significantly higher for children vaccinated with either lot of meningococcal C conjugate vaccine (0.62 and 0.76, respectively, vs 0.33; P=.001). There was no significant difference in diphtheria titers between meningococcal vaccine lots despite different concentrations of diphtheria CRM₁₉₇ toxoid. The antidiphtheria antibody titer increased significantly in all groups after booster vaccination at 12 months with meningococcal C conjugate vaccine. There was no difference in the response between those who had received either meningococcal C-1 or C-2 conjugate vaccines or between those who received either meningococcal C conjugate vaccine or HBV vaccine. As expected a booster dose of MPS did not affect diphtheria antibody concentrations.

This meningococcal C-CRM₁₉₇ conjugate vaccine was well tolerated in infants as young as 2 months of age when given with their other routine vaccines. There were no significant differences in local or systemic reactions between the children given either meningococcal C conjugate vaccine or the control vaccine during the primary series. The general reaction rates observed are similar to those reported when routine vaccines are given alone.¹⁸ A fourth dose of meningococcal C conjugate vaccine given as a booster at 12 months was also well tolerated, and there was significantly less local tenderness, general irritability, and change in eating habits compared with children who received MPS booster immunization.

The antibody results indicate that these meningococcal C conjugate vaccines are highly immunogenic and able to induce both a primary response in young infants and immunologic memory. Antimeningococcal C ELISA antibody levels and SBA activity increased progressively with each dose in the primary vaccination series. The results are consistent with the antibody titers achieved in studies using the first-generation unsized meningococcal A and C conjugate vaccine.^{14 - 15} At 12 months of age, the antibody titer had fallen markedly, and only 75% of children who received meningococcal C conjugate vaccine in infancy had a titer of 1:8 or higher. This was significantly higher than the controls, but raised the question of whether immunologic memory had been induced. Polysaccharide challenge is generally accepted to imitate the response to natural infection. The impressive increase in serum ELISA and SBA responses in all children following MPS booster vaccination suggested that immunologic memory had indeed been induced.

Are these children now protected against meningococcal group C disease? Serological correlates of long-term protection against meningococcal disease following vaccination are not well defined. Studies in adults linked natural protection against invasive disease due to serogroup C with a serum bactericidal titer of 1:4 or higher,²² but serological correlates following conjugate vaccination have not been defined. Vaccination of Brazilian infants who received a serogroup C MPS vaccine resulted in detectable antibodies by ELISA but not by SBA, and they were shown to be unprotected against disease.²³ In our study, all infants achieved SBA titers of 1:8 or higher after 3 doses of conjugate vaccine and, in addition, were shown to have immunologic memory. This raises the question of whether the absolute bactericidal titer is really the best measure of protection against invasive disease after conjugate vaccination or whether low titers may still be protective in the presence of immunologic memory. It is anticipated that long-term protection is more dependent on the induction of immunologic memory than on serum antibody titer but this remains an issue for debate. The experience with the Hib conjugate vaccine in the United Kingdom lends some support to this argument. A 3-dose schedule given at 2, 3, and 4 months of age with no booster dose has resulted in excellent control of Hib disease despite relatively low anti-polyribosylribitol phosphate antibody concentrations in vaccinated children (P.T.H., unpublished data, 2000). However, as with the UK Hib vaccine program, careful postimplementation surveillance will be necessary to document the progress of the meningococcal C conjugate vaccine program and the possible need for further doses in vaccinated children.

The response to a single dose of conjugate vaccine at 12 months of age is relevant to a catch-up program when the vaccine is introduced routinely for all children. Children vaccinated in this study with a single dose of conjugate vaccine at 12 months had a significantly higher antibody response than children vaccinated with a single dose of MPS at the same age or a single dose of conjugate vaccine at 2 months of age. This study does not provide information about the long-term response to a single dose at 12 months or whether immunologic memory has been induced in these children. This is particularly relevant since only 66% of these children achieved a titer of 1:8 or higher. Further studies will need to be performed to answer these questions.

We found no difference in reactogenicity or immunogenicity between the 2 lots of meningococcal conjugate. At 5 and 12 months, the diphtheria titers were higher in the children who received either meningococcal C-1 or C-2 conjugate vaccines than controls. Given that the protein incorporated in the meningococcal C conjugate vaccine is a mutant diphtheria toxin, this result is not surprising. No interference was seen in the response to the meningococcal component of the conjugate vaccine in either the primary series or after booster vaccination, and there was no increase in either local or general reactions to suggest that this is currently a clinical problem. However the total dose of protein given in the primary series when new protein-conjugate vaccines are added to the routine schedule needs further evaluation. Currently, insufficient data are available to rely on the CRM₁₉₇ component of the conjugate alone as an immunizing agent against diphtheria. Neither formulation resulted in a reduction in the response to other antigens. The difference in GMC for poliovirus 1 just reached statistical significance; however, 100% of children in the meningococcal C conjugate vaccine groups achieved a titer of 1:8 or higher. Since no differences were detected for polioviruses 2 and 3, this was thought to be a chance finding of no clinical significance.

The UK schedule requires 3 doses to be given 1 month apart with completion by 4 months of age. Despite this demanding accelerated schedule, all infants achieved potentially protective titers of 1:8 or higher, with 98% achieving this threshold after just 2 doses of the vaccine. A larger study assessing disease occurrence would be required to demonstrate protective efficacy but would require a sample size of several hundred thousand. Given that meningococcal C conjugate vaccines appear to be protective, as judged by bactericidal titers, the UK government has decided to implement a vaccine program in which 3 doses are given to infants at 2, 3, and 4 months of age, 2 doses to infants between 4 and 12 months of age, and 1 dose to those between 13 months and 18 years of age. This commenced at the end of 1999. As the first nation in the world to embark on such a program, the safety and efficacy of this approach will be watched with great interest.