Safety and pathovariant-independent susceptibility in a Salmonella Typhimurium controlled human infection model: a phase 1, randomised, double-blind, dose-escalation study
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background
Invasive non-typhoidal Salmonella (iNTS) disease causes an estimated 605,000 cases and 76,000 deaths each year, concentrated in sub-Saharan Africa, where the African Salmonella Typhimurium sequence type 313 (ST313) lineage predominates. Vaccine development is hampered by an absence of efficacy data and undefined correlates of protection.
Methods
We conducted a phase 1, randomised, double-blind, dose-escalation controlled human infection model (CHIM) in healthy UK-resident adults, who were randomly assigned 1:1 to oral challenge with S. Typhimurium 4/74 (ST19, associated with gastrointestinal disease) or D23580 (ST313, associated with invasive disease). Dose-escalation was guided by a Bayesian continual reassessment method (CRM). The primary endpoint was Salmonella diagnosis, defined as sustained fever ≥38°C on ≥2 occasions ≥12 hours apart and/or bacteraemia. Trial registration ClinicalTrials.gov ( NCT05870150 ).
Findings
Between August 2023 and December 2024, 50 participants were enrolled (25 per strain). 10 5 CFU was the maximum feasible dose, with CRM-estimated attack rates of 57·9% (95% credible interval 37·3–73·8) for D23580 and 47·4% (26·7–65·7) for 4/74. There were no serious adverse events. We found no clinical, microbiological, or immunological difference between the two pathovariants. Higher baseline serum anti-O-antigen IgG was associated with reduced disease (adjusted OR 0·42, 95% CI 0·17–0·90) and higher baseline faecal anti-lipopolysaccharide IgA with reduced colonisation (OR 0·12, 95% CI 0·01–0·59).
Interpretation
This S . Typhimurium CHIM is safe, reproducible, and provides a platform to generate early efficacy signals and candidate correlates of susceptibility, thereby de-risking future iNTS vaccine trials. The absence of a phenotypic difference between the invasive and gastrointestinal pathovariants in immunocompetent adults suggests that host factors, rather than pathogen adaptation alone, shape the invasive phenotype seen in endemic settings.
Funding
Wellcome Trust.
