Early warning of outbreaks
Dr Patrick Kimu (Kilimanjaro Clinical Research Institute, Tanzania) described the development of a machine learning-based early warning system for diarrhoeal disease outbreaks. He drew together clinical data on diarrhoeal diseases from a range of sites across Tanzania, as well as environmental and other data, then used machine learning to pull out the factors most strongly associated with outbreaks. Those showing the strongest correlations were integrated into a predictive model that was able to reproduce the observed trends in case numbers.
The tool could be used to support targeting of factors linked to the risk of outbreaks, and to provide programme staff with an early warning so that preventive measures can be intensified or response mechanisms activated.
Preventing diarrhoeal disease
Dr Jahangir Hossain (Medical Research Council Unit The Gambia at the London School of Hygiene and Tropical Medicine) summarised the results of a phase IIb trial of the ETVAX vaccine, designed to protect against enterotoxigenic E. coli (ETEC), responsible for more than 10% of deaths from diarrhoeal disease.
Conducted at three sites in The Gambia and involving 5000 young children, the trial missed its primary vaccine efficacy endpoint, protection specifically against ETEC. However, coinfections were common, and protection against any non-parasitic causes of diarrhoeal disease was significantly higher – particularly among the youngest children. The vaccine therefore appears to have cross-protective properties, potentially important when coinfections are so common, and may be most beneficial when given before 9 months. Findings have informed the design of a phase III trial of ETVAX.
Continuing the vaccine theme, Dr Ashwani Kumar Arora (GSK Vaccines for Global Health (GVGH), Italy) described preliminary safety data from a trial of a vaccine against invasive non-typhoidal Salmonella (iNTS), responsible for at least 62,000 deaths a year, most of them in sub-Saharan Africa. The vaccine is based on the novel GMMA platform, which is based on the budding of antigen-presenting vesicles from engineered cells in culture.
Following successful phase I studies in Europe, a phase IIa trial is being conducted in Ghana. The trial is an age-descending, dose escalation trial, which ultimately will assess vaccine efficacy in children down to 6 weeks, those in most need of protection. Blinded safety data have raised no issues of concern, and the study is now beginning to vaccinate the youngest age group.
The rising AMR challenge
Shifting gear, Dr Tizazu Zenebe Zelelie (Debre Berhan University, Ethiopia) described work on the presence of beta-lactamase AMR genes in E. coli at two sites in Ethiopia. AMR in pathogenic strains of E. coli is a growing concern globally but very few data are available from Ethiopia.
Dr Zelelie analysed samples from nearly 500 children, examining antibiotic susceptibility and the presence of AMR genes. He identified multiple AMR genes, including beta-lactamase genes, suggesting that AMR could be a significant challenge in Ethiopia and arguing for greater AMR surveillance.
Dr Mwelwa Chibuye (Center for Infectious Diseases Research in Zambia (CIDRZ), Zambia) has focused on Shigella, a very common and highly contagious cause of diarrhoeal disease. Drawing on a cohort of children in a peri-urban area of Lusaka, she has analysed hospital samples from diarrhoeal disease cases to identify Shigella species and strains. These then underwent antibiotic susceptibility testing and whole genome sequencing to identify AMR genes.
The 22 positive isolates comprised a mix of Shigella species and strains. More than half (64%) were found to be multidrug-resistant (resistant to at least one drug from at least three classes of antibiotic). A particular issue was a fluoroquinolone resistance gene found on a plasmid, which also carries several other antibiotic resistance genes. This plasmid has also been seen in Malawi, suggesting it may be circulating regionally.