Vi polysaccharide–protein conjugate vaccine for the prevention of typhoid fever in children: hope or hype?

Salmonella typhi causes an estimated 21 million new cases of typhoid fever and 216,000 deaths every year [1]. Since typhoid fever is an enteric infection that is spread feco-orally through contaminated water and food, it is found most commonly in developing countries where infrastructural facilities, including the provision of clean drinking water and sewage disposal facilities, are inadequate. Most parts of South Asia, Southeast Asia, Africa, Central Asia and South America are considered endemic for this disease, with an annual incidence of more than 100 cases per 100,000 population [2]. Conventionally thought to be a disease of schoolchildren and young adults, since the mid-1990s there has been increasing evidence of high rates in young children in these endemic countries [3–5].

The inactivated, whole-cell vaccine is immunogenic in all age groups but the high rates of adverse effects necessitated its replacement for wider use [6]. The live oral vaccine contains attenuated bacteria, which generate local immunity in the gut and cell-mediated immunity after oral administration. Limitations of the oral vaccine include the requirement for three or four doses for adequate immunogenicity, the need for cold storage and a theoretical possibility of reversion to pathogenicity. Furthermore, this vaccine has not been licensed in children below 6 years of age owing to safety and practical concerns that have limited its wide use in developing countries.

The virulence factor (Vi) capsular polysaccharide (ViPS) vaccine was developed in the 1980s, offering the possibility of population control [7]. However, similar to other T-cell-independent polysaccharide vaccines, the ViPS does not generate immunological memory and is not boostable by repeated vaccination [8,9]. A significant disadvantage of the ViPS is the fact that, similar to other polysaccharide vaccines, it is not immunogenic in children below 2 years of age, presumably owing to the absence of a splenic marginal zone in early childhood, which is necessary for the generation of antipolysaccharide immunity.

A ViPS-recombinant Pseudomonas aeruginosa exotoxin A protein conjugate vaccine was developed by John Robbins and colleagues at the NIH in 1994 [10]. Such glycoconjugate vaccines are T-cell-dependent antigens and produce immunological memory that can be boosted and might be expected to provide protection in young children. Indeed, two doses of the ViPS–protein conjugate vaccine administered 6 weeks apart was highly immunogenic and had a protective efficacy of 91.1% in children aged 2–5 years in Vietnam over 27 months’ follow-up [11]. This is greater than the protective efficacy of 72, 64 and 69% obtained from field trials with the parenteral ViPS vaccine in Nepal, South Africa and China, respectively [7,12,13]. Protective efficacy for the live oral vaccine, as obtained from a trial in 109,000 schoolchildren in Santiago, Chile, is reported to be 67% following three doses of enteric-coated preparation given on alternate days and 53 and 42% following three doses of either liquid or enteric-coated formulations, respectively, in 20,543 participants in Jakarta, Indonesia [3,14]. These efficacy data fit in with the theory that the conjugate vaccine may offer greater protection than the older vaccines, especially in young children. Moreover, no serious adverse effects were reported from the 5525 children who received two doses of the ViPS-conjugate vaccine. Minor adverse reactions included temperature of 37.5°C or over in 1.35% of participants after the first dose and swelling of at least 5 cm, which resolved within 48 h, in 0.36% of participants following the second dose [11].

With the identification of isolates of Salmonella typhi that do not express the ViPS capsule [15,16], doubts have been raised regarding the usefulness of a ViPS–protein conjugate vaccine [17]. Initially dismissed as artefacts associated with poor microbiological techniques of isolation or repeated subculturing on artificial media, there is now undeniable evidence from PCR studies of clinical isolates of Salmonella typhi of the existence of Vi-negative strains [18]. The genome of these strains demonstrates an absence of one or more loci in the ten genes that constitute the viaB operon which codes for the production and transportation of the ViPS. However, studies that determined the incidence of Vi-deficient Salmonella typhi, using molecular genetic methods, have demonstrated an extremely low incidence of this strain among clinical isolates. Only one out of 2222 (0.045%) clinical isolates of Salmonella typhi tested negative for the ViPS by PCR and immunofluorescence [18]. This is much lower than the numbers suggested by agglutination tests with Vi antisera. In view of the very low incidence of the Vi-deficient strains, it is not unreasonable to believe that these strains would not pose a significant problem to the introduction of the conjugate vaccine. Indeed, the rare occurrence of invasive disease caused by acapsulate meningococci has not prevented the successful implementation of conjugate meningococcal vaccines [19–21].

There is no evidence that use of ViPS vaccine produces a positive selection pressure for Vi-negative bacteria to appear. There has been no uniform implementation of the ViPS vaccine with a wide enough coverage and for a sufficient period of time to substantiate this theory. Additionally, while Vi-deficient strains of Salmonella typhi have been proven to be hyperinvasive in ex vivo conditions, via an increase in secreted bacterial proteins [22], the negligible incidence of typhoid fever caused by this strain suggests a decreased ability to survive once in the blood stream.

The ViPS–protein conjugate vaccine has not yet been evaluated in children under 2 years of age and these data are necessary to support incorporation into the Expanded Program of Immunization in developing countries. In such an evaluation, it is worth considering whether the administration of the vaccine should be delayed until late infancy, when immunogenicity of conjugate vaccines is known to improve and age-specific rates of disease begin to rise. Furthermore, establishment of laboratory correlates of immunity is a priority in order to evaluate the potential impact of the vaccine in different populations without further efficacy studies.

Other causes of enteric fever are an additional challenge to the use of ViPS–protein conjugate vaccine for control of this disease. Once thought to contribute to only a small proportion of enteric fevers, paratyphoid fever caused by Salmonella paratyphi A has been recently increasing in incidence in South Asia and Southeast Asia [23,24]. A prospective study comparing the rates of typhoid and paratyphoid fevers in four countries in Asia reported Salmonella paratyphi to contribute to 15, 24, 14 and 64% of enteric fevers in Karachi, Calcutta, North Jakarta and Heichi City, respectively [23]. A retrospective study in New Delhi revealed an increase in the proportion of Salmonella paratyphi from 6.5% in 1994 to 44.9% in 1998 [25]. Also, the assumption that paratyphoid fever is less severe than typhoid fever has been proven untrue in areas with high incidence of enteric fevers [26]. The fact that there are no licensed vaccines against paratyphoid fever is of concern and merits address from vaccine developers.

New-generation live oral vaccines designed to be more immunogenic than the Ty21a vaccine and in various stages of clinical trials include the ΔaroC, ΔaroD, ΔhtrA strain CVD908-htrA; strain χ4073 with mutations in cya, crp, cdt; strain Ty800 with a mutation in phoP/phoQ; ΔaroC, ΔssaV strain ZH9; Strain CVD909 and ΔguaBA strain CVD915 and CVD916 [27]. Successful results with these recombinant vaccines could provide an alternate option for the eradication of typhoid fever.

Despite the future promise of these new live oral vaccines, the ViPS–protein conjugate vaccine has an immense potential for the prevention of typhoid fever in endemic countries today. Clearly, investment in improved water and sanitation must be the primary goal for prevention of enteric fever but, in the meantime, the ViPS–protein conjugate vaccine could be manufactured with relative ease using the technology that we have taken for granted since the launch of the Haemophilus influenzae type b conjugate vaccine in the 1980s. With this in mind, it is quite shameful that children continue to die from typhoid fever today.

Acknowledgements

Anoop S Pulickal is a Rhodes Scholar and Andrew J Pollard is a Jenner Institute Investigator.

Conflicts of interest

Andrew J Pollard acts as chief investigator for clinical trials conducted on behalf of the University of Oxford (UK), sponsored by vaccine manufacturers (Novartis Vaccines, GlaxoSmithKline, Sanofi–Aventis, Sanofi–Pasteur MSD and Wyeth Vaccines), and has received assistance from vaccine manufacturers to attend scientific meetings. Industry-sourced honoraria for lecturing or writing are paid directly to an independent charity or an educational/administrative fund held by the Department of Paediatrics, the University of Oxford (UK). Anoop S Pulickal declares that he has no conflicts of interest.