1832. Insights into Typhoid Fever Pathogenesis and Immunobiology Provided By a Controlled Human Infection Model
Session: Oral Abstract Session: Featured Oral Abstract
Saturday, October 5, 2013: 4:45 PM
Room: The Moscone Center: Esplanade Ballroom 301-310
Background: Typhoid infection is a major cause of disease worldwide. A controlled human infection model was established to investigate disease pathogenesis, immunobiology and as a mechanism for future vaccine appraisal, and provided data relevant to disease modelling and diagnostics.

Methods: A dose escalation, oral challenge study was conducted. 40 healthy adults swallowed 2.1g sodium bicarbonate to neutralise gastric acid, prior to ingesting 103 or 104 CFU of Salmonella Typhi. Disease endpoints were 12 hours of fever >38.0oC and S. Typhi bacteraemia in the 14 days post challenge. Ciprofloxacin 500mg twice daily for 14 day was initiated in those with typhoid or at 14 days post challenge. Daily blood and stool culture and alternate day biochemistry and haematology assays were performed. Symptom reporting was by diary card. Antibody responses to the H, LPS and Vi antigens were measured by enzyme-linked immunosorbent assay (ELISA).

Results: Challenge was safely performed with 103 CFU (mean actual dose, 1.28x103 CFU; n=20) and, subsequently, 104 CFU (mean, 2.07x104 CFU; n=20). Compared to challenge with 104 CFU, 103 CFU gave a lower attack rate (55% vs. 65%), longer incubation period (mean IP, 9.7 vs. 7.7 days), fewer symptoms (mean, 31 vs. 37 solicited symptoms), fewere symptomatic days (mean, 7.1 vs. 9.1 days), and shorter bacteraemia duration(mean, 1.9 vs. 3.1 days). Similarly, C-reactive protein (mean 55.4 vs. 77.0 mg/L) and erythrocyte sedimentation rate (5.7 vs. 10.3 mm/hr) at diagnosis was lower, as were Anti-H and LPS responses following challenge with 103 CFU compared to 104 CFU. Anti-Vi responses were minimal in all participants, as were anti- H and LPS responses in those without typhoid. Arthralgia was the most discriminatory symptom (reported by 19/24 (79%) with typhoid vs. 1/16 (6%) without typhoid). Platelet count descrease in all those with typhoid, with counts below 150x103 µL in 11/24 participants. Eosinophil count declined from 6 days prior to diagnosis of typhoid in 23/24 participants, and was the earliest blood parameters to change.

Conclusion: This model of typhoid suggested a dose-response relationship in disease severity and immune response, not previously been described. Arthralgia, decreased platelet and eosinophil counts were useful markers of disease. This model can now be used for evaluation of vaccine protective efficacy.

Claire Waddington, MRCP(UK), DPhil1, Tom Darton, MRCP(UK), DTM&H1, Claire Jones, PhD1, Kathryn Haworth2, Simon Kerridge2, Derrick Crook, MB BCh3, Jeremy Farrar, MD, PhD4, Gordon Dougan, FMedSci, FRS5, Myron Levine, MD, FIDSA6, Brian Angus3 and Andrew Pollard, FRCPCH, PhD1, (1)Oxford Vaccine Group, University of Oxford, OXFORD, United Kingdom, (2)Paediatrics, Oxford Vaccine Group, University of Oxford, Oxford, United Kingdom, (3)Nuffield Department of Medicine, University of Oxford, OXFORD, United Kingdom, (4)Oxford Univ. Clinical Res. Unit, Ho Chi Minh City, Viet Nam, (5)Wellcome Trust Sanger Institute, CAMBRIDGE, United Kingdom, (6)University of Maryland School of Medicine, Baltimore, MD


C. Waddington, None

T. Darton, None

C. Jones, None

K. Haworth, None

S. Kerridge, None

D. Crook, Optimer Pharmaceuticals: Investigator and Scientific Advisor, Consulting fee and Research support

J. Farrar, None

G. Dougan, None

M. Levine, None

B. Angus, None

A. Pollard, None

See more of: Featured Oral Abstract
See more of: Oral Abstract Session

Findings in the abstracts are embargoed until 12:01 a.m. PST, Oct. 2nd with the exception of research findings presented at the IDWeek press conferences.