Abstract
Infection with antimicrobial-resistant (AR) pathogens is a leading cause of morbidity and mortality among patients with hematological malignancies; however, little is known about the gut microbiome dynamics in acute myeloid leukemia patients and its impact on AR infections (ARI) and/or colonization with AR pathogens (ARC). Longitudinal stool samples collected from 154 patients undergoing induction chemotherapy were analyzed using 16S rRNA sequencing, selective and differential media culturing, MALDI-TOF, and VITEK2 to identify patients with ARC or ARI and to isolate AR infectious and colonizing bacterial strains. Shotgun metagenomic sequencing of baseline stool samples revealed taxa abundances, resistome features, and KEGG pathways associated with AR-events. Baseline observed species were lower in patients with AR-events (p = 0.01). Although several baseline taxa were more abundant in AR-event patients, they were not statistically significant when they were corrected for false discovery. Functional analysis revealed that penicillin and cephalosporin biosynthesis pathways were significantly enriched in patients with ARC. In summary, identifying the baseline microbiome, resistome, and functional pathway biomarkers may forecast an increased risk of ARI and/or ARC, thereby informing antimicrobial treatment strategies in AML patients.