Predicting treatment-emergent antimicrobial resistance at the patient level in LMICs
Year: 2026
Type: article
Study design: Study Protocol
Abstract: Resistance that emerges within a patient during treatment is clinically devastating but poorly predicted, and the evidence that it is foreseeable from routine infection histories comes almost entirely from high-income settings, leaving open whether such prediction transports to low- and middle-income countries where resistance prevalence, circulating pathogens, and prescribing practices all differ. This pilot proposes a prospective, standardised, multi-site LMIC cohort, framed as an extension of the ACORN II surveillance network, that systematically re-cultures patients enrolled with a baseline susceptible isolate of a priority organism–drug pair at fixed timepoints, with multi-colony whole-genome sequencing of paired within-patient isolates alongside structured clinical, exposure, and outcome data. This would quantify emergence incidence per site, adjudicate genomically whether emergence reflects reinfection, within-host mutation, or expansion of a pre-existing minority lineage, and train interpretable patient-level prediction models benchmarked against prior antibiotic exposure alone and externally validated across countries to test geographic transportability. This proposal is a product from “Roadmapping the Future of AI-Driven Antimicrobial Resistance Research”, a community alignment initiative led by The Align Foundation and supported by Google DeepMind, with the goal of prioritizing key challenges and opportunities for the use of AI/ML in helping combat AMR.
Language: English
Subfield: Applied Microbiology and Biotechnology
Field: Immunology and Microbiology
Domain: Life Sciences
Open Access status: green
OpenAlex ID: W7202208743