Microbial biomarkers of intestinal MDR colonization after solid organ transplantation

ABSTRACT Novel approaches are needed to tackle the increasing global public health threat of antimicrobial resistance (AMR). Multi-drug resistant organisms (MDRO), such as Enterobacterales resistant to carbapenems (CRE), extended spectrum beta-lactamases (ESBL) and vancomycin-resistant Enterococci (VRE), are associated with high morbidity and mortality as effective treatment regimens are lacking. Patients undergoing advanced medical treatments, such as solid organ transplant recipients (SOTR) are disproportionally affected by MDRO infections with poor outcomes. Intestinal colonization with MDRO has been identified as a risk factor for infection and a source of onward transmission. However, while a risk factor, it is not sufficiently predictive of infection. Better predictors are needed to detect those at high risk of MDRO infection and treat them appropriately while avoiding unnecessary empiric broad-spectrum antibiotic exposure in those at low risk. Reduction in misuse of antibiotics is an important pilar in the global effort to reduce AMR. Moreover, novel non-antibiotic therapeutic strategies to eliminate MDR gastrointestinal carriage are needed to decrease individual infection risk and MDRO spread. The gut microbiome has a strong potential to address these challenges as it serves as a major ecosystem for bacterial colonization and can promote or suppress the growth of MDROs. We have shown that microbial diversity is strongly associated with MDRO colonization, and our preliminary results indicate that metagenomic analyses can predict future MDRO colonization and clearance. Here, we propose to study solid organ transplant recipients (liver, kidney, heart) who are at very high risk for MDRO infections and adverse outcomes. The long- term objective of our study is to provide a predictive framework for clinical management of MDR colonization that is informed by the gut microbiome and to identify potential therapeutic targets for future study and validation. We will first leverage and expand our existing collection of >2,200 samples with known MDR status from an existing cohort and biobank of liver transplant patients. We will generate a combination of whole-genome sequencing of clinical infectious isolates and both short- and long-read metagenomics of gut microbial communities. Among other tools, we will use copangraph, a new framework for characterizing genomic variability in the accessory metagenome via joint high-resolution analysis of short and long read sequencing. We will devise rigorous machine learning models that use the microbiome to predict MDRO colonization, infection, and clearance. To test the applicability of our observations we will validate our models in contemporaneous liver, kidney and heart transplant recipients. Contrasting different SOTR cohorts will facilitate the identification of both generalizable and diseases-specific properties of the microbiome. Overall, this study will form the basis for a microbiome-based toolset and risk management algorithms for critical clinical decision-making, including the identification of high- risk patients for whom prophylactic treatment strategies for MDR colonization could then be developed.