Utilization of machine learning to predict antibiotic resistant event outcomes in acute myeloid leukemia patients
Stephanie McMahon1, Samantha Franklin1, Jessica Galloway-Peña1
1Laboratory of Jessica Galloway-Peña, Texas A&M University, Department of Veterinary Pathobiology, Interdisciplinary Graduate Program in Genetics and Genomics, College Station, TX, United States.
Introduction:
Acute myeloid leukemia (AML) patients are highly susceptible to infection. Moreover, prophylactic and empirical antibiotic treatment during chemotherapy disrupts the gut microbiome, raising the risk for antibiotic-resistant (AR) opportunistic pathogens. There is limited data on risk factors for AR infections or colonization events in treated cancer patients, and no predictive models exist. This study aims to combine metagenomic and antibiotic administration data to develop a model predicting AR event outcomes.
Methods:
Baseline stool microbiome, antibiotic administration, resistome, and clinical metadata from 95 patients were utilized to build a Random Forest model to predict AR infection and colonization events by serious AR threats. Additionally, sparse canonical correlation analysis assessed correlations between microbiome and resistome data, while Spearman correlation networks identified direct associations with AR event outcomes and secondary variables.
Results:
AR-events were identified in 14 of the 95 included patients, with 8 developing AR infections and 9 identified as AR colonized. A Random Forest model predicted AR event outcomes (AUC = 0.73), identifying bacterial taxa and antibiotic resistance gene (ARG) classes as key variables of importance. Methanobrevibacter smithii, Clostridium leptum, and Bacteroides dorei were identified as key taxa associated with reduced risk of AR events, suggesting the potential roles of commensals in maintaining gut microbial resilience during chemotherapy. ARG classes, particularly those conferring resistance to lincosamides, macrolides, and streptogramins, were negatively associated with AR events.
Conclusion:
These results underscore the value of integrating microbiome and resistome features to reveal potential protective mechanisms and improve risk prediction for AR outcomes in vulnerable patients.
Insights
This study predicts antibiotic-resistant (AR) events in acute myeloid leukemia (AML) patients by analyzing gut microbiome and antibiotic resistance genes. Certain gut bacteria and resistance gene classes can reduce the risk of AR events during chemotherapy.
Area of Science:
- Microbiome research
- Infectious disease epidemiology
- Computational biology
Background:
- Acute myeloid leukemia (AML) patients face high infection risks.
- Chemotherapy-induced gut microbiome disruption increases antibiotic-resistant (AR) pathogen risk.
- Limited data exists on AR risk factors and predictive models in cancer patients.
Purpose of the Study:
- To develop a predictive model for AR events in AML patients.
- To integrate metagenomic and antibiotic administration data for risk prediction.
- To identify key microbial and genetic factors associated with AR events.
Main Methods:
- Utilized baseline stool microbiome, antibiotic administration, resistome, and clinical data from 95 patients.
- Developed a Random Forest model to predict AR infection and colonization events.
- Employed sparse canonical correlation analysis and Spearman correlation networks to analyze data.
Main Results:
- A Random Forest model achieved an AUC of 0.73 in predicting AR events.
- Key predictors included bacterial taxa (e.g., *Methanobrevibacter smithii*, *Clostridium leptum*, *Bacteroides dorei*) and antibiotic resistance gene (ARG) classes.
- Specific commensal bacteria were associated with reduced AR event risk, while certain ARG classes showed negative associations.
Conclusions:
- Integrating microbiome and resistome data improves AR event risk prediction in vulnerable patients.
- Identified potential protective roles of commensal bacteria in maintaining gut resilience.
- Highlights the value of these features for understanding and mitigating AR outcomes.
