Viewed from a worldwide pediatrie perspective, the problem of enteric infections, including diarrheal diseases, dysenteries, and enteric fever, mainly resides in developing countries where morbidity and mortality remain high. In developing countries, as many as 7 to 10 separate episodes of diarrheal disease occur per child per year during the first 2 years of life. Whereas many etiologic agents have been incriminated as causes of pediatrie diarrheal disease in developing countries, a relative few account for most disease of public health importance, and these have been prioritized for accelerated vaccine development.1 Thus, if saie and effective vaccines were available to prevent disease due to totavirus, enterotoxigenic Escherichia coli (ETEC), enteropathogenic E coli (EPEC)1 Shigeüa, and Vibrio choleroe 01, and if these vaccines could be reliably delivered to the population at risk, the burden of diarrheal diseases might be markedly lightened in developing countries.1
In developing countries, typhoid fever is the main enteric disease threat faced by school-aged children who have successfully run the gauntlet of diarrheal diseases and dysentery encountered during the first 5 years of life. Prevention of this infection is a priority, particularly for areas where Salmonella typhi strains exhibit multiple resistance to the main antibiotics.2
Within industrialized countries, S/ugeua infections persist as a problem in pediatrie day-care centers and in other situations where one encounters populations with compromised personal hygiene. These populations are targets for ShigeUa vaccines.1 Children from industrialized countries are at risk of a wider range of bacterial enteric infections if they travel to developing countries.
Characteristics of Live Oral Vaccine Ty21a and Parenteral Vl Polysaccharide vaccine
This review of Shigella, V cholerae 01, ETEC1 EPEC, and S typhi vaccines will focus on products that have been licensed already (and are therefore available to practicing pediatricians), that have been evaluated in controlled field trials of efficacy, or that have entered clinical trials to assess their safety and immunogenicity.
TYPHOID FEVER VACCINES
Ty 2 Ia, Live Oral Typhoid Vaccine
TyZIa, an attenuated strain of S typhi that is safe and protective as a live oral vaccine, was developed in the early 1970s by chemical mutagenesis of a pathogenic S typhi strain. TyZIa has proven to be remarkably well tolerated in placebo-controlled clinical trials.2
Controlled field trials of TyZIa emphasize that the formulation of the vaccine, the number of doses administered, and the spacing of the doses markedly influence the level of protection that can be achieved.2 In the first field trial of TyZIa in Alexandria, Egypt, ß?? 7-year-old children received three doses of vaccine (suspended in a diluent) on Monday, Wednesday, and Friday of 1 week; to neutralize gastric acid, the children chewed a 1-g tablet of NaHCO3 before ingesting the vaccine or placebo. During 3 years of surveillance, 96% protective efficacy against confirmed typhoid fever was observed. A later formulation consists of lyophilized vaccine in enteric -coated, acid-resistant capsules. In a randomized, placebocontrolled field trial in Santiago, Chile, three doses of this enteric-coated formulation given within 1 week provided 67% efficacy during the first 3 years of follow-up3 and 63% protection over 7 years of followup
Four doses of TyZIa in enteric-coated capsules given within 8 days (one dose every other day for four doses) are significantly more protective than two or three doses.4 When Ty21a was licensed in the United States by the Food and Drug Administration in late 1989, it was with a recommended schedule of four doses given at every-other-day intervals; other countries use a three-dose immunization schedule.
To determine whether TyZIa administered as a liquid suspension (somewhat similar to that used in Alexandria) is superior to vaccine in enteric-coated capsules, two field trials were initiated in Chile5 and Indonesia.6 In both trials, the liquid formulation was superior but the difference in efficacy between the two formulations was particularly marked in the Chilean trial.5 It is expected that a liquid formulation of TyZIa will become widely available in 1994. In summary, orally administered, well tolerated TyZIa offers a moderate level of protection against typhoid fever if one can assure that the required 3 to 4 doses (every-other-day schedule) are taken. The salient characteristics of TyZIa vaccine are summarized in the Table.
Purified Vi Polysaccharide Parenteral Vaccine
The Vi antigen is a capsule polysaccharide found on the surface of S typhi that has been associated with its virulence properties.7 Purified Vi used as a parenteral polysaccharide vaccine was evaluated in randomized, placebo-controlled, double-blind field trials in Nepal8 and South Africa? where a well-tolerated dose of 25 µg about 65% protection against typhoid fever for at least Z years. Thus, the Vi vaccine provides a moderate level of protection with just a single dose. The characteristics of purified Vi vaccine also are summarized in the Table.
New Attenuated S typhi Strains as Live Oral Vaccines
New strains of S typhi with precise attenuating mutations are being engineered that may successfully immunize following administration of a single oral dose. The strategies being pursued are based on mutations that attenuate Salmonella typhimurium for mice and the assumption (which is not always true) that analogous mutations in S typhi may accomplish the same for humans. One candidate strain, CVD 908, has been shown to be well tolerated yet highly immunogenic following ingestion of a single oral dose.9 CVD 908 harbors mutations in the pathway for biosynthesis of aromatic amino acids, thereby rendering it nutritionally dependent on substrates that are not present in human tissues. Consequently, CVD 908 undergoes only limited proliferation in the human body.
An additional potential use of the new generation of engineered attenuated strains of S typhi is as so-called live vector vaccines. In this context, genes encoding protective foreign antigens of other bacteria, viruses, or parasites are expressed in the attenuated S typhi, which serves to deliver those antigens to the human immune system where a protective immune response is induced. The new generation of attenuated S typhi vaccines is particularly attractive for this application as live vectors because they are administered orally and elicit a broad immune response in humans that includes intestinal secretory immunoglobulin A (SIgA) and serum antibody as well as cell-mediated and antibody-dependent cellular cytotoxic responses. Moreover, techniques are available to integrate foreign genes into the Salmonella chromosome or to stabilize plasmids carrying foreign genes.
Vi Polysaccharide-Carrier Protein Conjugate Vaccines
Booster doses of purified Vi do not raise antibody titers over those elicited by a single dose of vaccine; ie, immunologie memory does not appear to occur. To increase its immunogenicity by conferring T celldependent properties on the antigen, including the induction of immunologie memory, Szu et al10 have conjugated Vi polysaccharide to carrier proteins, such as tetanus toxoid. In animals, booster doses of conjugate vaccine clearly increase the titers of antibody over those elicited by a priming dose. A Vi conjugate vaccine currently is undergoing clinical trials.
In the 1960s and early 1970s, attenuated strains of ShigeUa, including streptomycin-dependent mutants and the T32 colonial mutant, were shown to be safe and protective live oral vaccines. Nevertheless, those pioneering vaccines had shortcomings: a requirement for multiple doses, the necessity to administer an annual booster to maintain protection, occasional genetic reversions, lack of knowledge of the molecular basis of attenuation, and dose-related vomiting in a few percent of vaccinées. For these reasons, research has continued to develop improved vaccines.
In the early 1980s, Formal et al11 prepared a Shigella sonnei vaccine by inserting into attenuated S typhi Ty21a the plasmid of S sonnei that contains genes for synthesis of the sonnei O antigen. The resultant hybrid strain, 5076- 1C, expresses both S typhi and S sonnei O antigens. Strain 5076-1C was well tolerated, and some lots significantly protected adult volunteers against sonnei shigellosis in experimental challenge studies. However, because of lot-to-lot variation in protective efficacy in adult volunteer studies, controlled field trials were not undertaken.
Clinical studies are underway with several live oral Shigelìa vaccine candidates. One vaccine consists of an E coli strain modified so that it expresses Shigeila flexneri 2a O antigen and invades epithelial cells yet does not cause keratoconjunctivitis in guinea pigs (a test of virulence of Shigella strains); this strain also harbors a mutation that limits proliferation in human tissues. Another S flexneri candidate has deletions in genes necessary for the organisms to proliferate in vivo after invasion. A third attenuated Shigelìa candidate has mutations that limit growth in human tissues and decrease intercellular spread from enterocyte to enterocyte
The inactivated whole cell parenteral cholera vaccines that have been available for 90 years can protect older children and adults in endemic areas for a few months but provide little protection to children younger than 5 years of age. Because V cholerae Ol does not invade the intestinal mucosa and specific SIgA is deemed important, modern approaches have focused on oral vaccines to stimulate SIgA intestinal antibody more efficiently.
Nonliving Oral Vaccines
Two related nonliving orai vaccines were tested concomitantly in a placebo-controlled, randomized, double-blind field trial in Bangladesh: one vaccine consisted of killed whole V chokrae Ol a mixture of both serotypes and biotypes) and the second vaccine consisted of whole vibrios in combination with the B subunit of cholera toxin (the immunogenic nontoxic portion of cholera toxin). During the first 6 months of follow-up after vaccination, the combination vaccine gave significantly better protection (85% versus 56% efficacy). However, this difference disappeared during the next 6 months of surveillance. Over the full 36 months of follow-up, the two vaccines provided the same moderate level of protection (approximately 50% to 53% vaccine efficacy).13
The otal inactivated vaccines constitute an improvement over the inactivated patenterai vaccines. They confirm that oral immunization can elicit protection and that immunity can endure for a rather extended period (3 years). Nevertheless, these vaccines leave much room for improvement: they afford only short-lived protection to young children (who suffer the highest incidence of cholera) and multiple spaced doses must be administered. Moreover, they protect less well against cholera caused by the El Tor biotype, particularly in persons of blood group O (an important host risk factor for cholera gravis). Live oral vaccines may overcome these drawbacks.
Live Oral Cholera Vaccines
The main approach to develop live oral vaccine candidates has been to attempt to attenuate V chokrae Ol by making deletion mutations in genes encoding critical virulence properties. Using recombinant DNA technology to pursue this approach, a highly promising vaccine candidate has emerged: strain CVD 103-HgR.14 CVD 103-HgR has been given to more than 4000 children (as young as 24 months of age) and adults in randomized placebo-controlled trials both in developing countries and in industrialized countries. CVD 103-HgR has proven to be well tolerated and highly immunogenic in all settings.14"16
A single dose of CVD 103-HgR confers on adult volunteers significant protection against experimental challenge with pathogenic V choleras Ol of either biotype or serotype.14 In these experimental challenge studies, protection is evident as early as 8 days after vaccination and lasts for at least 6 months the shortest and longest intervals tested). A large-scale, randomized, placebo-controlled, double-blind field trial involving 66 000 pediatrie and adult subjects is currently underway in Indonesia to evaluate the efficacy of CVD 103-HgR in preventing cholera under natural challenge conditions in an endemic are.
New Antigenic Strain of V choleras
Beginning in late 1992, a new antigenic strain of V chokroe (serogroup 0139 Bengal) emerged that proved to be capable of causing epidemic cholera of notable severity. Regrettably, immunity against V choleroe 01, such as that conferred by either natural infection or CVD 103-HgR, appears not to provide crossprotection against cholera caused by the new 0139 serogroup; 0139 vaccines are under accelerated deveU opment.
VACCINES AGAINST ETEC
Enterotoxigenic E coli comprise many different O;H serotypes, multiple antigenic types of fimbrial colonization factors, and three different toxin phenotypes (heat-labile toxin only, heat-stable toxin only, or both toxins). To confer broad-spectrum protection in the race of such antigenic heterogeneity is a dilemma that must be overcome. Fortunately, epidemiologie evidence and results of experimental challenge studies in volunteers clearly demonstrate that immunity follows ETEC infection. It is believed that broad-spectrum immunity in endemic areas derives from serial infections with strains bearing different antigens. Protection appears to be mediated by secretory immunoglobulin A (SIgA) antibody directed against fimbriae, other surface antigens, and heatlabile enterotoxin (LT); heat-stable toxin (ST), a small peptide, does not elicit neutralizing antitoxin following natural infection.
Given the complexity of vaccine development in this particular area, candidate vaccines of many different strategies are under construction. l>1 ? These include nonliving antigen vaccines (toxoids, mixtures of inactivated whole bacteria, and purified surface antigens) and several live oral vaccines. Phenotypes of some of these vaccines have been tested in clinical trials.
Based on the strong antigenic similarity between LT and cholera toxin, the killed whole vibrio/B subunit combination oral cholera vaccine tested in Bangladesh also conferred a moderate level (about 65%) of short-lived cross-protection against diarrhea due to LT-producing E coli in the initial 3 months following vaccination. This combination cholera vaccine provided Finnish travelers with a comparable level of protection against travelers' diarrhea due to LT-producing E coli.18
When conjugated to a carrier protein as a hapten, ST peptide can elicit binding and (sometimes) neutralizing antitoxin. Several groups have attempted to construct ST toxoids as recombinant fusion proteins or as synthetic peptides. One oral toxoid, consisting of synthetic ST cross-linked to a synthetic immunodominant epitope of LT, was well tolerated by adult volunteers and elicited neutralizing intestinal SIgA anti-LT and anti-ST. No ST toxoid has ever been tested for efficacy.
Purified Colonization Factor Fimbriae as Oral Vaccines
Heretofore, purified colonization factor fimbriae used as oral vaccines have been poorly immunogenic in nlimans because of the harmful effect of gastric juice on the fimbrial protein. The novel delivery system of biodegradable polymers has rekindled interest in this type of vaccine. These polymers may protect protein antigens such as ETEC fimbriae from gastric juice, thereby allowing successful oral immunization. Clinical trials with polymer-microencapsulated fimbrial antigens are currently in progress.
Inactivated Fimbriated Whole E coli With or Without Toxoid
Inactivated fimbriated whole E coli bacteria, given alone or with toxoids, are under investigation as oral vaccines with variable preliminary results. A colicintreated whole-cell oral vaccine protected volunteers from experimental challenge. Such treatment inactivates the bacteria without adversely affecting surface antigens such as fimbriae.
Swedish investigators have prepared an oral vaccine consisting of inactivated ETEC strains bearing different fimbrial colonization factor antigens, in combination with the B subunit of cholera toxin. This vaccine is undergoing clinical trial to establish the safety and immunogenicity of a two-dose regimen. Another formalin-treated whole cell oral ETEC vaccine consisting of inactivated nontoxigenic E coli bearing colonization factor antigen II fimbriae was only modestly immunogenic and did not protect volunteers from experimental challenge.
Live Oral Vaccines Against ETEC
A nonenterotoxigenic E coli strain expressing colonization factor fimbriae has served as a prototype live oral vaccine. A single dose elicited prominent intestinal SIgA antifimbrial antibody in volunteers who were significantly protected against experimental challenge. Attenuated E coli and S typhi strains are being engineered to express various ETEC colonization factor antigens, in addition to LT (and ST toxoid) antigens.
VACCINES AGAINST EPEC
Enteropathogenic E coí of "classical" O:H serotypes were originally identified in the 1940s and 1950s as important causes of community outbreaks of infant diarrhea and of nosocomial outbreaks of diarrhea in newborn nurseries.19 Since the late 1970s, the pathogenesis of EPEC infection has been steadily unraveled,20 paving the way for vaccine development activities. A plasmid (the so-called EPEC adherence factor plasmid) that is found exclusively in EPEC encodes critical virulence properties that include a novel attachment factor (bundle-forming piii). This plasmid also regulates expression of a protein, "intimin," that is required for intimate attachment to enterocytes. Humans who experience EPEC diarrhea mount an immune response to this protein. Vaccine development research is following two broad tracks. One involves attenuation of EPEC by modifying critical virulence properties, eg, by altering the sequence of intimin to render it biologically inactive without abolishing its immunogenicity. The other approach entails expressing immunogenic EPEC proteins, such as bundle-forming pili or a modified intimin protein, in nonpathogenic E coil or attenuated S typhi live vectors.
Considerable progress has been made in the last decade in developing vaccines against the most important bacterial enteric infections. Two new vaccines against typhoid fever (oral Ty21a and parenteral Vi polysaccharide) have been licensed in many countries. Newer generations of more sophisticated typhoid vaccines are undergoing clinical testing including recombinant attenuated S typhi strains and Vi polysaccharide-carrier protein conjugate vaccines. Two inactivated oral cholera vaccines, consisting of inactivated V chokrae Ol bacteria alone or in combination with the B subunit of cholera toxin, each conferred 50% to 53% protection over 3 years in a field trial in Bangladesh where subjects were immunized with a three-dose regimen. An engineered live oral cholera vaccine, strain CVD 103-HgR, has been shown in extensive clinical trials to be well tolerated by children and adults in developing countries and highly immunogenic following administration of just a single oral dose; a large-scale field trial of the efficacy of this vaccine is underway. Several candidate vaccines against Shigella and enterotoxigenic E coli are in clinical trials.
1. Levine MM. Vaccines against enteric infections. Lancei. 1990;335:958-961.
2. Levine MM, Taylor DN, Ferraccio C. Typhoid vaccines come of age. Pedieur infeci Dis i. 1989;8: 374-381.
3. Levine MM. Ferreccio C, Black RE, Chilean Typhoid Committee, Germanier R. Large-scale field iría! of Ty2!a live oral typhoid vaccine in enteric-coated capsule formulation. Lonctt. 1987 ;l·. 1049- 1 05 Z.
4. Ferraccio C, Levine MA, Rodríguez H, Contreras R, Chilean Typhoid Committee. Comparative efficacy of two, three, or four doses of Ty2 U live oral typhoid vaccine in enteric coated capsules. 7 Infect DIJ. 1989:159:766-769.
5. Levine MM, Ferreccio C, Cryi S, Ortii E. Comparison of enteric-coated capsules and liquid formulation of Ty21a typhoid vaccine in a randomized controlled field trial. Lancet. 1990;336:891-S94.
6. Simanjuntak C, Paleólogo F, Punjabi N. et al. Oral immuniïatUm against typhoid fever In Indonesia with TyZIa vaccine. Lancer. 1991 338:105 5-1 059.
7. Klugman K, Gilbertson IT, Koomhof HJ, et al. Protective activity of Vi capsular polysaccharide vaccine against typhoid fever. Lancet. 1987:2:1 165-1 169,
8. Acharya IL, Lcwe C, Thapa R, et al. Prevention of typhoid fever in Nepal with the Vt capsular polysaccharide of Salmonella typhi. A preliminary report. N Eng! ] Med, 1987;317:1101-1104.
9. Tacket CQ Hone DM, Losonsky G, Guers L, Edelman R, Levine MM. Clinical acceptability and immunogenicity of CVD 908 SabnoriiUa typhi vaccine strain. Vaccine. 1992:10:443-446.
10. Siu SC, Li X, Schneerson R, et al. Comparative immunogen (cities of Vi polysaccharideprateín conjugates composed of cholera toxin or its B subunit as a carrier bound to high- or lower-molecular weight Vi. Infecí Immun. 1 989 i57:3823 -382 7.
11. Formal SB, Baton LS, KopeckoD], et al. Construction of a potential bivalent vaccine strain: introduction of Shigeäa ionnei form 1 antigen into the GaJE Salmonera tyfhì TyZIa ryphoid vaccine strain. Infect Immun. 1981 ;34:746- 760.
12. Lindbetg AA, Pal T. Strategies fur development of potential candidate Shigeua vaccines. Vacane. 1993;11;168-179.
13. Clemens JD, Sack DA, Harris J, et al. Field trial of oral cholera vaccines in Bangladesh: results from 3 -year follow-up. Lancet. 1990; 1:2 70-273.
14. LevineMM, Kapei JB. Live vaccinesagainsteholera: an update. Vaccin«. 1993;11'207212.
15. Suharyono, Simanjuntak C, Witham N, et al. Safety and immunogenicity of single-dose live oral cholera vaccine CVD 103-HgR in 5- io 9-year-old Indonesian children. Lancei. 1992 i340:689-694.
16. Simanjuntak C, O'Hanley P, Punjabi NH, et al. The safety, immunogenicity and transmissibility of single-dose live oral cholera vaccine CVD 103-HgR in 24- to 59-month-old Indonesian children../ Infect Du. I993;168:1169-1176.
17. Levine MM. Vaccines against enterotoxigenic Eicfwrtclua coli infections. Vaccines based predominantly on antibacterial immunity. In: Woodrow GC, Levine MM, eds. New Generation Vacaras, New York, NY: Marcel Dekker Ine; 1990:649-660.
18. Peltola H, Sitonen A, Kyronseppa H, et al, Prévention of travelers' diarrhea by oral B-subunlt/whole-cell cholera vaccine. Lancet. 1991 ;338: 1285- 1289.
19. Levine MM, Edelman R. Entempathogenic Escherichia coti of classical seroiype« associated with mfent dacthea - eptdecmtitogy and pattwgeneîis. Epidemial Rev. 1984;6:31-51
20. DonnenbergM, Kaper JB. Enteropathogenic Eschenc/uocoíi'. ínfea Immun. 1992:60:39533961.
Characteristics of Live Oral Vaccine Ty21a and Parenteral Vl Polysaccharide vaccine