Typhoid fever causes more than 21 million cases of disease and 200,000 deaths yearly worldwide, with more than 90% of the disease burden being reported from Asia. Epidemiological data show high disease incidence in young children and suggest that immunization programs should target children below two years of age: this is not possible with available vaccines. The Novartis Vaccines Institute for Global Health developed a conjugate vaccine (Vi-CRM197) for infant vaccination concomitantly with EPI vaccines, either starting at 6 weeks with DTP or at 9 months with measles vaccine. We report the results from a Phase 1 and a Phase 2 dose ranging trial with Vi-CRM197 in European adults.
Following randomized blinded comparison of single vaccination with either Vi-CRM197 or licensed polysaccharide vaccines (both containing 25·0 µg of Vi antigen), a randomised observer blinded dose ranging trial was performed in the same center to compare three concentrations of Vi-CRM197 (1·25 µg, 5·0 µg and 12·5 µg of Vi antigen) with the polysaccharide vaccine.
All vaccines were well tolerated. Compared to the polysaccharide vaccine, Vi-CRM197 induced a higher incidence of mild to moderate short lasting local pain. All Vi-CRM197 formulations induced higher Vi antibody levels compared to licensed control, with clear dose response relationship.
Vi-CRM197 did not elicit safety concerns, was highly immunogenic and is therefore suitable for further clinical testing in endemic populations of South Asia.
Citation: van Damme P, Kafeja F, Anemona A, Basile V, Hilbert AK, De Coster I, et al. (2011) Safety, Immunogenicity and Dose Ranging of a New Vi-CRM197 Conjugate Vaccine against Typhoid Fever: Randomized Clinical Testing in Healthy Adults. PLoS ONE 6(9): e25398. https://doi.org/10.1371/journal.pone.0025398
Editor: Ray Borrow, Health Protection Agency, United Kingdom
Received: July 13, 2011; Accepted: September 2, 2011; Published: September 30, 2011
Copyright: © 2011 van Damme et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was conducted in collaboration with Sclavo Vaccines Association with grants from Regione Toscana and Fondazione Monte Dei Paschi di Siena which partially funded the trials. The funders had no role in study design, data collection, and analysis, decision to publish, or preparation of the manuscript.
Competing interests: AA, VB, AKH, SR, FM, RMQK, EM, VDC, AS, LBM, AP are all Novartis employees. AS and AP receive stock options from Novartis. This does not alter the authors' adherence to all the PLOS ONE policies on sharing data and materials. PvD, FK and IDC are all employees of Univeristy of Antwerp, Belgium.
Typhoid fever, a disease caused by Salmonella enterica serovar Typhi (S. Typhi), represents a global public health issue causing over 21 million cases of disease and over 200,000 deaths per year worldwide . Although endemic in many developing countries, typhoid fever has the highest incidence in the Indian subcontinent where 90% of the S. Typhi related morbidity and mortality are reported , . Typhoid fever was previously considered a typical disease of school children and adolescents ; however, most recent epidemiologic data from India, Pakistan, Bangladesh and Vietnam clearly show that younger children are also highly affected by typhoid fever and are at risk of severe disease –. These data have prompted a re-consideration of prevention strategies and suggest that immunization of subjects below two years could be a better approach to protect young children, to increase vaccination coverage and to reduce costs by concomitant delivery with routine vaccines of the Expanded Program on Immunization (EPI) –.
Currently available typhoid vaccines, although moderately effective and well tolerated, cannot be administered to children younger than two years. Additionally, the Vi polysaccharide induces a short lasting protection and requires repeated boosters which might cause hypo-responsiveness . Since S. Typhi is an obligate human pathogen, an effective vaccination program could lead to substantial herd protection . Therefore, new highly efficacious vaccines are needed to have the greatest public health impact in high risk areas and to allow their use in existing EPI vaccination programs. As shown by clinical results with the Vi-rEPA vaccine in Vietnam –, a conjugate Vi vaccine appears to meet these criteria and, in addition, clinical data generated in Vietnam have allowed the establishment of a serological threshold for clinical protection .
The Novartis Vaccines Institute for Global Health (NVGH) was founded in 2007 with the not for profit mission of developing effective and affordable vaccines for neglected infectious diseases of impoverished communities of developing countries . By conjugation of the Vi polysaccharide with the CRM197 protein, NVGH developed a vaccine against typhoid fever (Vi-CRM197) that has the potential to induce a T-cell dependent immunological memory and to be effectively used for infant immunization in developing countries –.
We describe results from a Phase 1 trial and a subsequent Phase 2 dose ranging trial aimed to evaluate the clinical profile of Vi-CRM197 in European adults and to select the dose for further clinical testing in endemic countries, where the vaccine will be ultimately used.
Materials and Methods
Vi-CRM197 contained purified Vi polysaccharide, derived from Citrobacter, chemically conjugated to mutant diphtheria toxin carrier protein, CRM197, widely used as carrier protein for several conjugate vaccines –.
The vaccine clinical lot (VI-10-001), available in single dose vials, contained 25 µg of Vi antigen conjugated to CRM197 in 0.5 mL of injectable solution and did not contain any adjuvant, stabilizer or preservative.
Phase 1 and 2 studies were designed as randomised controlled trials and were conducted sequentially at the Center for the Evaluation of Vaccination of the University of Antwerp (Antwerp, Belgium) from May to November 2010.
Healthy adults aged 18 to 40 years were enrolled in both studies. Subjects were excluded from the trial if they presented with serious chronic or progressive disease, with immune dysfunction, under immunosuppressive therapy, with bleeding diathesis, with hypersensitivity to vaccine components, with progressive or severe neurological disorder, with seizures or Guillain-Barré syndrome, with malignancy or lymphoproliferative disorder, that previously received vaccines against typhoid fever, with previous history of infection with S. Typhi, that received blood products, with body temperature ≥38.0°C within 3 days of study vaccination. Women who were pregnant or breast-feeding and women of childbearing age not willing to use acceptable birth control measures were also excluded. Ethical approval was obtained from the Ethical Committee of the University Hospital of Antwerp. The studies were conducted according with the International Conference on Harmonization, Good Clinical Practice guidelines (ICH-GCP). All participants provided written informed consent.
In Phase 1, Vi-CRM197 vaccine was administered as single 0.5 mL dose containing 25 µg of Vi antigen. Subsequent to the Phase 1 study, with the objective to define the optimal Vi-CRM197 concentration for further development, the Phase 2 dose ranging trial was implemented, using three different concentrations administered as single 0.5 mL doses containing 1.25 µg, 5.0 µg and 12.5 µg of Vi respectively; the formulations were obtained by bedside dilution of the 25 µg vaccine with appropriate amount of saline solution. In both studies the control vaccine was a licensed Vi polysaccharide (Vi-PS) vaccine (Typherix®, GlaxoSmithKline) containing 25 µg of Vi.
In the Phase 1 study, 50 subjects were randomized, with a 1∶1 ratio using a randomization block size of four, to receive either Vi-CRM197 25 µg or Vi-PS as a single intramuscular (IM) injection in the deltoid of the non-dominant arm. In the Phase 2 study, 88 subjects were randomized, with a 1∶1∶1∶1 ratio using randomization block size of four, to receive a single IM dose of one of three Vi-CRM197 concentrations (1.25 µg, 5.0 µg and 12.5 µg of Vi respectively) or the Vi-PS. In both studies, randomization sequence was computer generated by the sponsor. Assignment of subject to treatment group was performed by designated unblinded clinical site personnel by means of individual envelopes provided by the sponsor who also established ad hoc monitoring visits during both trials to ensure appropriateness of all protocol procedures. The protocols for the two trials and supporting CONSORT checklist are available as supporting information; See Checklist S1, Protocol S1 and Protocol S2.
Evaluation of immunogenicity
Blood samples of 20 mL were obtained for hematological and hematochemical analyses (Phase 1 only) prior to vaccination (day 1) and 28 days post vaccination. For both studies, urine samples were obtained prior to vaccination to evaluate potential drug abuse; additionally, urine pregnancy tests were performed in females before vaccination and at the study end. Blood samples of 10 mL were obtained from both Phase 1 and Phase 2 subjects on days 1, 28, and on day 180 in Phase 1 only. Blood samples were centrifuged within 24 hours and serum was maintained below −20°C.
Antibody levels were measured by an enzyme-linked immunossorbent assay (ELISA) in the Novartis Vaccines & Diagnostics human serology laboratory (Marburg, Germany), using Vi polysaccharide from Citrobacter. Round bottom 96-well microtiter plates were coated overnight with 1.0 µg/mL Vi in PBS pH 7.0, plates were blocked with 5% skim milk powder in PBS, then duplicate samples of human sera, serially diluted 2-fold, were added and incubated for 2 hours at room temperature. Goat anti-human IgG (γ-chain specific) conjugated to alkaline phosphatase was added and incubated 2 hours at room temperature. After 30 min incubation with p-nitrophenyl-phosphate solution, the reaction was stopped. The absorbance at 405 nm and 650 nm was read. The A405–A650 were used as the test parameter.
Each plate contained a serially diluted pool of anti-Vi human sera. This pool was assigned a value of 115.4 ELISA U/mL. A standard curve was generated from the corresponding values using a 4 parameter fit. ELISA units in test sera were calculated from this curve. Low and high human serum controls were included on each plate. The ELISA lower detection limit ranged from 3.0 to 7.8 U/mL.
Evaluation of safety
Safety of study vaccines was assessed by evaluation of solicited indicators of reactogenicity within 7 days after vaccination. Following vaccination, subjects were observed at the site for at least 30 minutes. Subjects maintained a daily diary card for 7 days post-vaccination recording local reactions (pain, induration, erythema), systemic reactions (headache, malaise, chills, myalgia, arthralgia, fatigue), body temperature (fever defined as axillary temperature ≥38.0°C) and use of analgesic and antipyretic drugs. Local pain and systemic reactions were scored by severity (mild, moderate and severe) and erythema and induration by the maximum diameter per day. All adverse events (AEs) were recorded for 28 days post-vaccination, while all serious adverse events (SAEs) were assessed and recorded for the entire study duration. For consistent measurement of reactions, subjects were provided with digital thermometers for measurement of body temperature and with a plastic ruler for measurement of size of local reactions.
No statistical null hypothesis was associated with immunogenicity or safety objectives. In the Phase 1 study with 25 subjects per group, there was a 90% probability of observing at least 1 subject with an adverse event if the true rate of such an event was 8.7%. In the Phase 2 study with 20 subjects per vaccine group, there was a 90% probability of observing at least 1 subject with an adverse event if the true rate of such an event was 10.9%.
All analyses were descriptive in nature. The pre-specified primary population for immunogenicity analysis was the Modified Intention to Treat Population (MITT). The MITT Population included all subjects who received the vaccination, had at least one post-vaccination blood sample collected, and at least one ELISA result available.
Seroconversion was defined as a post vaccination increase of at least four times the pre-vaccination antibody concentration.
Geometric mean concentrations (GMCs) and associated two-sided 95% confidence intervals (CIs) were calculated by exponentiating (base 10) the means of log-transformed (base 10) titers and their 95% confidence intervals. Antibody concentrations below assay limit of detection were assigned a value of one half of the cut-off. At each sampling time GMCs of antibodies (with two-sided 95% CIs) and reverse cumulative curves were calculated. The percentages of subjects with seroconversion were calculated for each vaccine group in each study. Analyses were performed using SAS version 9.1 (SAS Institute Inc, Cary, North Carolina).
A total of 138 subjects were enrolled across both studies (50 Phase 1; 88 Phase 2), of whom 133 (96%) completed the studies [47 (94%) Phase 1; 86 (98%) Phase 2]. There were no withdrawals from either study due to AEs (figure 1). Demographic characteristics were similar among vaccine groups and between the two studies (table 1). None of the subjects showed clinically significant variations of haematological and hematochemical parameters after vaccination (day 28) compared to pre-vaccination values. The most frequently reported local reaction was mild to moderate local pain, which occurred with a similar frequency among all four Vi-CRM197 groups and was consistently higher in Vi-CRM197 groups compared with Vi-PS control groups. Percentages of subjects reporting pain did not show a dose dependent relation in the Vi-CRM197 groups. One subject in each of the four Vi-CRM197 groups reported severe pain lasting <1 day; no severe cases of erythema and induration were reported (table 2). Percentages of subjects reporting systemic reactions were similar among all vaccine groups. Headache and fatigue were the most commonly reported systemic reactions in all vaccine groups (overall range was 27% to 48% and 24% to 50% for headache and fatigue respectively). Severe systemic reactions were uncommon and were reported by seven subjects overall (one case in the Vi-CRM197 25 µg group and six in Vi-PS group) (table 1). No differences were observed in the use of analgesic or antipyretic drugs among study groups (table 2). Of all study subjects, 59% of the Vi-CRM197 and 53% of the Vi-PS recipients reported at least one AE during the 28 days after vaccination and 38% and 26% of these were considered possibly or probably related to the study vaccine, in the Vi-CRM197 and Vi-PS groups respectively. In Phase 1, no possibly or probably related AEs were reported from day 29 to day 180. Most of the AEs reported in both studies were mild or moderate in nature. The percentage of subjects reporting any AEs or AEs possibly or probably vaccine related was lower in Phase 1 than in Phase 2 (table 3). Two unrelated SAEs were reported in Phase 1 study in the Vi-PS group, one hospitalization for pneumothorax and one hospitalization for ankle fracture. No SAEs were reported in the Phase 2 study. No AEs leading to premature withdrawal from the study and no deaths were reported in both studies.
A total of 130 subjects had evaluable day 28 anti-Vi ELISA results and were included in the immunogenicity analyses at day 28 post-vaccination. At baseline (day 1), a high percentage of the sera (84%) had antibody levels below limit of detection. These sera were assigned values half the detection level and thus the calculated GMCs, which were similar in all the six vaccine groups, reflect the assigned values (table 4). The GMCs at day 28 of the four Vi-CRM197 groups were higher (range from 63 to 304) than the licensed Vi-PS vaccine groups (range from 37 to 52) (Table 3). Day 28 GMCs in the Vi-CRM197 groups were dose dependent; the highest Vi-CRM197 dose (25 µg) elicited the highest antibody level (304) and the lowest level (63) was observed for the Vi-CRM197 1.25 µg dose. These findings are further illustrated by reverse cumulative distribution curves for anti-Vi ELISA concentrations (figure 2). Four weeks after vaccination, at least 95% of the subjects vaccinated with Vi-CRM197 achieved seroconversion, whereas 88% and 95% of subjects achieved seroconversion in the Vi-PS control groups. In the Phase 1 study, six months after the vaccination (day 180), seroconversion rates were still very high in both vaccine groups (95% and 92% in the Vi-CRM197 25 µg and Vi-PS groups, respectively). However, although the GMC of Vi-CRM197 25 µg group (69) was still higher than that of Vi-PS (51), a faster decrease in antibody levels compared to the control group was noticed (table 4).
Antibody concentrations are given in log scale. Values below the limit of detection of the assay were set to 1.0 U/mL.
Epidemiological studies provide compelling evidence that typhoid fever induces significant disease burden in Asia and that, particularly in the urban slums of the Indian subcontinent, the disease incidence is huge –. Additionally, the most recent data do not support the previous belief that typhoid is a disease affecting almost exclusively school children and adolescents ; in fact, children aged less than two years are highly affected in India, Pakistan and Bangladesh by clinically severe disease –. In industrialized countries, typhoid control has been largely achieved by provision of clean water, adequate sanitation and overall improved hygiene conditions. Unfortunately these conditions are only a long term solution for most of the developing countries where the disease is endemic , . This public health issue is further complicated by the increasing emergence of multi-drug resistant strains of S.Typhi , –. Resistance to antimicrobials renders the control of the disease even more difficult and expensive, as the easily available and cheaper first-line antibiotics are not effective and their use leads to increased hospitalization and associated costs , , . Therefore, vaccine prevention represents the only concrete option for efficient disease control.
After discontinuation of the highly reactogenic whole-cell S. Typhi vaccine , effective and well tolerated vaccines, such as Vi polysaccharide and live attenuated Ty21a vaccine, have been developed. However, these vaccines have limitations; the Vi polysaccharide vaccine cannot be administered to subjects younger than 2 years and, even in pre-school children, its efficacy is controversial . Additionally, polysaccharide vaccines do not induce immunological memory, require frequent boosters, to maintain adequate protection, and may be associated with hyporesponsiveness . The commercially available formulation of Ty21a vaccine can only be administered to children older than 6 years and requires three doses to induce protection. Consequently the use of these vaccines and their impact on the control of the disease in endemic countries has generally been very limited. The development of new vaccines has been advocated to overcome the limitations of the current ones and allow the use of alternative vaccination strategies for younger children and infants , –. Although the National Institutes of Health (NIH) pioneering work has shown that a prototype Vi conjugate vaccine can be developed and is highly efficacious, there has not been a sufficient commercial incentive for vaccine manufacturers to invest in the development of improved typhoid vaccines for developing countries.
NVGH, an institute created with the specific objective of developing new and affordable vaccines against neglected diseases for the impoverished populations of developing countries , has developed a Vi conjugate vaccine which, in preclinical studies, has shown to be safe and highly immunogenic –.
The clinical studies described in this paper have shown that Vi-CRM197 is clinically well tolerated. The only evidence of reactogenicity was given by the higher incidence of pain at the injection site compared to Vi-PS. Interestingly, the reaction rates did not show a dose dependent relationship and were similar across the four Vi-CRM197 concentrations tested. However, in most of the cases, reported pain was mild to moderate and of short duration. Additionally, pain was never associated with any evidence of inflammation at the injection site, as shown by the very low rates of erythema and induration, which were not different between the two study vaccines. The use of analgesic medications in the conjugate vaccine group was low and similar to that of the Vi-PS control group; finally, no subject withdrew from the trial due to adverse reactions. Systemic post immunization reactions were consistent with an injectable vaccine; their rates were generally low and did not significantly differ among the various vaccine groups. Importantly, no case of fever was reported, nor was any case of vaccine related serious adverse event. Larger Phase 2–3 studies will help define more precisely the safety profile of the vaccine, but the initial data show that Vi-CRM197 is a safe vaccine.
The immunogenicity data show that the Vi-CRM197, even at very low antigen concentrations (i.e., 1.25 µg/dose = 1/20th of the licensed control concentration), is similar or better than Vi-PS (25 µg/dose) and the overall postimmunization Vi-CRM197 to Vi-PS GMC ratio varies between 1.7 and 5.8. As shown by field trials with the Vi-PS – and with the Vi-rEPA conjugate vaccine , antibodies to the Vi antigen confer protection against typhoid fever. Therefore, inferring the efficacy of the conjugate vaccine based on immunogenicity data is scientifically justified and demonstration that antibody levels induced by Vi-CRM197 in infants are not less than those induced by the Vi-PS vaccine in older children, a population where vaccine efficacy has been shown, could be an appropriate basis for regulatory approval. However, establishing a correlation with the immunogenicity of the Vi-rEPA vaccine and, more importantly, with the serological threshold of protection defined in the Vi-rEPA Vietnamese efficacy trial  would be extremely valuable and is currently underway.
The current data are very promising and suggest that an effective conjugate vaccine against typhoid fever can be rapidly developed. This vaccine has been primarily developed for immunization of the impoverished populations of South Asia, which have the highest risk of infection and severe disease. Therefore, it is essential to evaluate its clinical profile in the populations that will ultimately use the vaccine. To this purpose, age de-escalation Phase 2 clinical trials with Vi-CRM197 are now ongoing in South Asia.
We thank all study participants for their time and support. We thank all NVGH staff who contributed with their work to make the clinical trials possible, the entire staff at the Center for the Evaluation of Vaccination – Vaccine & Infectious Disease Institute in Antwerp for their support and dedication, the staff at GenIbet Biopharmaceuticals (Portugal), Areta International (Italy) and B&C Group (Belgium) for the preparation, release and shipment of the clinical lot.
Conceived and designed the experiments: PvD AP AA. Performed the experiments: PvD FK IDC AP AA EM VB RMQK. Analyzed the data: AA. Wrote the paper: PvD AP AA VB AS LBM AKH. Development of the vaccine and production and release of the clinical lot: FM SR VDC LBM AS.
- 1. Crump JA, Luby SP, Mintz ED (2004) The global burden of typhoid fever. Bull World Health Organ 82: 346–53.
- 2. Ochiai RL, Acosta CJ, Danovaro-Holliday MC, Baiqing D, Bhattacharya SK, et al. (2008) A study of typhoid fever in five Asian countries: Disease burden and implications for controls. Bull World Health Organ 86: 260–8.
- 3. Sinha A, Sazawal S, Kumar R, Sood S, Reddaiah VP, et al. (1999) Typhoid fever in children aged less than 5 years. Lancet 354: 734–7.
- 4. Brooks WA, Hossain A, Goswami D, Sharmeen AT, Nahar K, et al. (2005) Bacteremic typhoid fever in children in an urban slum, Bangladesh. Emerg Infect Dis 11: 326–9.
- 5. Siddiqui FJ, Rabbani F, Hasan R, Nizami SQ, Bhutta ZA (2006) Typhoid fever in children: some epidemiological considerations from Karachi, Pakistan. Int J Infect Dis 10: 215–22.
- 6. Bhutta ZA, Khan MI, Soofi SB, Ochiai RL (2011) New advances in typhoid Fever vaccination strategies. [Review]. Adv Exp Med Biol 697: 17–39.
- 7. Lin FY, Ho VA, Phan VB, Nguyen TT, Bryla D, et al. (2000) The epidemiology of typhoid fever in the Dong Thap Province, Mekong Delta region of Vietnam. Am J Trop Med Hyg 62: 644–8.
- 8. Saha SK, Baqui AH, Hanif M, Darmstadt GL, Ruhulamin M, et al. (2001) Typhoid fever in Bangladesh: Implications for vaccination policy. Pediatr Infect Dis J 20: 521–4.
- 9. Ghosh S, Batabyal P, Rajendran K, Palit A (2010) Typhoid fever in rural communities of West Bengal, India-an age-wise perspective. Jpn J Infect Dis 63: 219–21.
- 10. Podda A, Saul A, Arora R, Bhutta Z, Sinha A, et al. (2010) Conjugate vaccines for enteric fever: Proceedings of a meeting organized in New Delhi, India in 2009. J Infect Dev Ctries 4: 404–11.
- 11. Overbosch D, Peyron F, Picot N, Varichon JP, Dumas R, et al. (2005) Combined typhoid fever and hepatitis A vaccine: Comparison of immunogenicity and safety to concomitant monovalent vaccine over 3 years. J Travel Med 12: 319–26.
- 12. Sur D, Ochiai RL, Bhattacharya SK, Ganguly NK, Ali M, et al. (2009) A cluster-randomized effectiveness trial of Vi typhoid vaccine in India. N Engl J Med 361: 335–44.
- 13. Lin FY, Ho VA, Khiem HB, Trach DD, Bay PV, et al. (2001) The efficacy of a Salmonella typhi Vi conjugate vaccine in two-to-five-year-old children. N Engl J Med 344: 1263–9.
- 14. Canh DG, Lin FY, Thiem VD, Trach DD, Trong ND, et al. (2004) Effect of dosage on immunogenicity of a Vi conjugate vaccine injected twice into 2- to 5-year-old Vietnamese children. Infect Immun 72: 6586–8.
- 15. Thiem VD, Lin FY, Canh DG, Canh do G, Son NH, et al. (2011) The Vi conjugate typhoid vaccine is safe, elicits protective levels of IgG anti-Vi, and is compatible with routine infant vaccines. Clin Vaccine Immunol 18: 730–5.
- 16. Podda A (2010) Aims and role of Novartis Vaccines Institute for Global Health (NVGH). Procedia in Vaccinology 2: 124–127.
- 17. Micoli F, Rondini S, Pisoni I, Proietti D, Berti F, et al. (2011) Vi-CRM197 as a new conjugate vaccine against Salmonella Typhi. Vaccine 29: 712–20.
- 18. Rondini S, Micoli F, Lanzilao L, Hale C, Saul AJ, et al. (2011) Evaluation of the Immunogenicity and Biological Activity of the Citrobacter freundii Vi-CRM197 Conjugate as a Vaccine for Salmonella enterica Serovar Typhi. Clin Vaccine Immunol 18: 460–8.
- 19. Cooper B, Detora L, Menveo StoddardJ (2011) A novel quadrivalent meningococcal CRM197 conjugate vaccine against serogroups A, C, W-135 and Y. Expert Rev Vaccines 10: 21–33.
- 20. Acharya IL, Lowe CU, Thapa R, Gurubacharya VL, Shrestha MB, et al. (1987) Prevention of typhoid fever in Nepal with the Vi capsular polysaccharide of Salmonella typhi. A preliminary report. N Engl J Med 317: 1101–4.
- 21. Klugman KP, Koornhof HJ, Robbins JB, Le Cam NN (1996) Immunogenicity, efficacy and serological correlate of protection of Salmonella typhi Vi capsular polysaccharide vaccine three years after immunization. Vaccine 14: 435–8.