This study has shown that both the MenA and MenC conjugate antigens present in the novel DTPw-HBV/Hib-MenAC combination vaccine candidate are immunogenic and induce bactericidal antibody titres ≥1
8 in close to 90% of infants compared to a prevalence of less than 10% prevalence in unvaccinated control infants. High levels of immunity to serogroup A exist in populations living in the “meningitis belt” 
. Carriage of meningococci of serogroup A and C was reported to be low in Burkina Faso, 
although a study reported 17% carriers in Nigeria 
. The high level of bactericidal antibodies in our study may be due to either cross-reacting antibodies or vaccination since a large segment of the population is routinely vaccinated with meningococcal polysaccharide vaccine. Thus, the presence of bactericidal antibodies (SBA titre ≥1
8) in a substantial proportion of infants before vaccination (44–48% of infants had SBA-MenC, 25–28% had SBA-MenA titres at this level) is not surprising, probably reflecting the transfer of maternally derived antibody. The proportion of infants in the control vaccine group with titres above this level fell during the following months reflecting a natural decline in maternally acquired antibody. In contrast, bactericidal antibody titres persisted in infants who had received the study vaccine and at 12 months of age, approximately half of them still had SBA titres ≥1
8 for each serogroup. The higher prevalence of anti-PSA (78%) and anti-PSC (75%) maternal antibodies compared to bactericidal antibodies prior to vaccination suggest that not all meningococcal antibodies transferred by the mother to her infant are bactericidal. When mothers were asked whether they had been vaccinated against meningitis in the year preceding the start of the study, approximately one quarter of them remembered being vaccinated; the protective immune response to meningococcal polysaccharide vaccines in adults is known to last for at least 3 years 
. Vaccination against meningitis is undertaken repeatedly in Northern Ghana and this could explain the high background prevalence of anti-PSA and anti-PSC seropositivity in our study.
MenA and Men C bactericidal antibody titres were lower in Ghanaian infants than those obtained in a similar study undertaken in the Philippines where 97.7 and 99% of vaccinated infants had SBA titres ≥1
8 for Men A and Men C respectively; GMTs were 316.7 (95%CI [251.4–398.9]) and 3132.6 (95%CI [2496.9–3930.1]) for serogroups A and C respectively 
. The present study was conducted in a tropical country but the cold-chain was closely monitored and there were no major deviations of the temperature at which the vaccines were kept (2–8°C). The immune response to the other antigens in the combined vaccine was as expected, so it is highly improbable that an undetected break in the cold-chain occurred. Another possibility that was considered was that the immune response of Ghanaian infants to the meningococcal conjugates might have been suppressed by high antibody concentrations in their mothers induced by natural exposure to meningococci or by vaccination. Therefore, the influence of infants' pre-vaccination antibody status (SBA-MenA, SBA-MenC, anti-PSA, anti-PSC) on the immune response to the new vaccine was studied. The response to the MenC component of the study vaccine was lower in infants seropositive for SBA-MenC or anti-PSC prior to vaccination, as has been found in the United Kingdom with the meningococcal serogroup C conjugate vaccine 
. In contrast to MenC, pre-existing anti-MenA antibodies did not influence the immune response to the MenA component of the vaccine. Another possible reason for the lower immunogenicity seen in the present trial compared with the one undertaken the Philippines might be the very high level of pre-existing anti-TT antibodies found in the prevaccination samples of Ghanaian infants in the study vaccine and control groups (93% and 97% respectively) resulting from the fact that most women had been vaccinated against tetanus during pregnancy following the recommendation of the WHO 
. However, a negative impact of pre-existing anti-TT antibodies on the MenAC response (the MenAC conjugate uses TT as a carrier) is unlikely as the anti-PRP response was higher in infants who received the study vaccine than in those who received the control vaccine, even though the Hib conjugate also uses a tetanus carrier. This effect was seen even though the quantity of PRP used in the study vaccine was 4 times less than in the control, probably as a result of a TT carrier effect as reported previously 
. Other possible explanations for the lower immunogenicity of the MenA and MenC components of the candidate vaccine in Ghanaian infants are malnutrition and malaria. Cases of malnutrition were excluded at study entry and no serious episodes of malnutrition were reported during the primary vaccination phase of the study. Malaria can suppress the response to many vaccines including Hib conjugate vaccines and meningococcal polysaccharide vaccines 
but malaria is relatively uncommon in the first six months of life, even in areas of high transmission. Malaria was reported in about half of the study infants during the primary vaccination phase of this study but, in most cases, diagnosis was not confirmed by microscopy. Further research may be needed to determine if malaria influences the immune response to the meningococcal conjugate antigens.
Non-inferiority of the DTPw-HBV/Hib-MenAC vaccine compared to the control vaccine was demonstrated for the five antigens common to each vaccine. Although seroprotection for hepatitis B was in the lower range of that found in published studies, it is comparable to that induced by the control vaccine and remains at an acceptable level given that no hepatitis B vaccine was administered at birth and that similar levels of protection have been seen previously following immunization with a hepatitis B monovalent vaccine in other African countries using the same immunization schedule 
The reactogenicity of DTPw-HBV/Hib-MenAC was similar to that of the control vaccine. The trend to more marked swelling at the injection site was without consequence and did not lead to any drop-outs. Three deaths reported in the study group within one month of primary vaccination were unrelated to the vaccine. These deaths, which occurred during the rainy season, when mortality is highest, are fewer than expected in a cohort of 280 children followed to the age of 7 months in a community with an infant mortality rate of 95.5/1000 
. One infant abandoned the study after a malaria episode followed by epilepsy; there is now good evidence that malaria increases the risk of epilepsy 
. Although most SAEs during the primary vaccination phase of the trial were in the study vaccine group (11 vs. 2), the pattern of SAEs was that of common infections in this community, such as malaria and acute respiratory infections, and none of the events were considered to be vaccine related. The occurrence of a higher incidence of SAEs in the study than in the control group was probably a result of chance as evidenced by a more extended safety follow-up that was done up to the end of the challenge phase which did not reveal a difference in the overall incidence of SAEs between the groups. Furthermore, in a large safety study including 1780 infants vaccinated with either the DTPw-HBV/Hib-MenAC or the DTPw-HBV/Hib vaccines in Asia, the incidence of SAEs was similar in both groups 
. Nevertheless, the safety of the new vaccine will require further evaluation in Africa, especially as a similar increase in SAEs was seen in a previous study that evaluated another meningococcal conjugate vaccine in Africa, although this was also considered to be most likely to be a result of chance 
Because the age distribution of cases of meningococcal disease in Africa covers a broader range than Hib or pneumococcal infections in children, introduction of the new vaccine into the EPI program would, alone, have only a limited impact on the overall burden of meningococcal disease until vaccination had been continued for many years. Recent experience in the UK with a MenC conjugate suggests that immunization in infancy gives only short-term direct protection and that a booster dose is required to provide sustained immunity 
. Additional complementary vaccination strategies are needed in addition to infant immunization if a maximum effect is to be achieved. Vaccination of all children and young adults with a monovalent group A conjugate, or even better with a tetravalent ACWY conjugate, is a possible way of achieving this using infant immunization to maintain the pool of protected individuals.
In conclusion, the DTPw-HBV/Hib-MenAC vaccine is non-inferior to the control vaccine in inducing antibody titres that are associated with protection against 5 important childhood diseases in a single series of injections given during the routine vaccination EPI schedule. The vaccine induces bactericidal antibody titres which should provide protection against infection with group A and group C meningococci in infants in the “meningitis belt” during the first year of life. However, induction of more sustained protection will require booster immunization given routinely or perhaps through mass vaccination campaigns. Introduction of this combined conjugate vaccine into the routine immunization programme of countries within the meningitis belt could be an important option towards lowering the burden due to recurrent N. meningitidis epidemics in Africa.