Characterization of Antibiotic Administration Factors Associated with Microbiome Disruption and Subsequent Antibiotic-Resistant Infection and Colonization Events in Acute Myeloid Leukemia Patients Receiving Chemotherapy

Samantha Franklin, C. Ramont, Maliha Batool, Stephanie Hunter McMahon, Pranoti Sahasrabhojane, John C. Blazier, Dimitrios P. Kontoyiannis, Yang Ni, Jessica Galloway-PeñaAntibiotics 2025 · DOI

TL;DR

This study investigates how different antibiotic administration strategies impact the gut microbiome and the emergence of antibiotic-resistant (AR) infections and colonizations in acute myeloid leukemia (AML) patients undergoing chemotherapy. The findings suggest that while some antibiotics can reduce microbial diversity, they don't always lead to an increase in ARGs or AR events, highlighting the complex interplay between antibiotic use, the microbiome, and resistance.

What Problem Does This Paper Address?

Patients with hematological malignancies, like AML, are highly susceptible to severe infections. To combat this, they often receive broad-spectrum antibiotics, which can disrupt their gut microbiome and promote antibiotic resistance. While the general link between antibiotic use and resistance is known, the specific impact of how antibiotics are administered (number of administrations, types of drugs) on the emergence of AR pathogens and resistance genes in the gut microbiome of AML patients undergoing remission induction chemotherapy (RIC) remains poorly understood. This knowledge gap is critical for developing targeted interventions to improve patient outcomes and preserve antibiotic efficacy.

Methods

The researchers employed a multi-faceted approach combining clinical data with advanced molecular techniques:

  • Shotgun Metagenomic Sequencing: This was used to analyze the gut microbiome and its "resistome" (the collection of antibiotic resistance genes) at two time points: baseline (before chemotherapy) and end-of-study (after chemotherapy). This allowed for the assessment of changes in microbial diversity and the abundance of ARGs.
  • 16S rRNA Gene Sequencing: This technique was used to identify bacterial species present in the gut samples, helping to understand shifts in microbial community composition.
  • Selective Culturing and MALDI-TOF Mass Spectrometry: These methods were used to identify specific AR-threat pathogens (like VRE, MRSA, ESBL-producing bacteria, CRE, and MDRP) in patient samples.
  • Antibiotic Susceptibility Testing (VITEK2): This was performed on identified isolates to determine their resistance profiles.
  • Statistical Modeling: A range of statistical models were used, including:
  • Mann-Whitney U test: For comparing non-parametric groups.
  • Spearman's rank correlation: To assess the strength and direction of associations between variables.
  • Time-varying Cox proportional hazards models: To analyze the risk of AR-threat events (infections and colonizations) associated with daily antibiotic exposure.
  • Generalized linear models (GLMs): To evaluate the association between antibiotic exposure and the acquisition of specific ARG classes.

Datasets

The study utilized longitudinal stool samples and clinical data from 119 newly diagnosed AML patients who underwent remission induction chemotherapy at MD Anderson Cancer Center between September 2013 and November 2019. This dataset included:

  • 168 stool samples in total (98 baseline and 70 end-of-study).
  • Complete metagenomic, clinical, and pharmacologic data for 119 patients.
  • Paired baseline and end-of-study samples were available for 46 patients.

The 16S rRNA gene sequences from a subset of these patients were previously deposited in the NCBI Sequence Read Archive (SRA) under specific BioProject IDs. The shotgun metagenomic sequences from this study were also deposited in the NCBI SRA.

Key Insights and Ideas

  • Antibiotic Diversity vs. Resistance: A greater number of unique antibiotic classes administered was associated with a loss of unique antibiotic resistance genes (ARGs). This suggests that broader antibiotic exposure might not necessarily lead to more resistance genes, but rather a reduction in the diversity of resistance mechanisms present.
  • Linezolid's Dual Role: While oxazolidinones (like linezolid) were associated with a loss of microbial diversity, each additional day of linezolid administration was linked to a reduced risk of AR colonization (ARC) and a decreased odds of acquiring macrolide resistance genes. This highlights that some antibiotics, despite impacting diversity, can have protective effects against specific resistance threats.
  • Not All Diversity Loss is Equal: The study found that while antibiotic-induced loss of microbiome diversity can occur, it's not always directly associated with an increase in AR-threat events. This suggests that the type of diversity lost and the specific ARGs present are crucial factors.
  • Complexity of Antibiotic Impact: The findings underscore that the impact of antibiotics on the microbiome and resistome is complex and not a simple cause-and-effect relationship. The type, duration, and combination of antibiotics all play a role.

Key Results

  • A greater number of unique antibiotic classes administered correlated with a loss of unique ARGs (R = −0.39, p = 0.008).
  • Increased oxazolidinone administrations were correlated with a greater loss of microbial diversity (R = -0.58, p < 0.001).
  • Each additional day of linezolid administration reduced the risk of ARC by approximately 30% (HR: 0.663, p = 0.047).
  • Linezolid administration also decreased the odds of acquiring genes predicted to confer macrolide resistance by about 50% (HR: 0.50, p = 0.026).
  • Eight patients had confirmed AR-threat infections (ARI), with multidrug-resistant Pseudomonas aeruginosa (MDRP) being the most common.
  • Fourteen patients had gastrointestinal colonization with AR-threat pathogens (ARC), with vancomycin-resistant Enterococcus (VRE) being prevalent.

Limitations and Future Work

The authors acknowledge several limitations:

  • Retrospective Design: The study's retrospective nature limits the ability to track previous antimicrobial exposure and long-term consequences of ARG expansion.
  • Limited Exposure Data: They only had access to antimicrobial exposures during the study, not prior to it.
  • Complex Antibiotic Regimens: The study did not fully account for the multifaceted nature of frequent combinations or sequential antimicrobial exposures.
  • Synergistic Effects: Few publications explore synergistic antimicrobial effects on the microbiome or resistome, which could be a factor.
  • Single-Center Study: The findings are from a single center, which may limit generalizability.

Future work could involve larger, multi-institutional cohorts, longer follow-ups, and improved modeling of synergistic antimicrobial effects to better understand the lasting impact of antibiotic use on the microbiome and resistance. Further investigation into additional risk factors beyond antibiotic administration could also help healthcare providers tailor regimens more effectively.

Summary updated

Abstract

The number of antibiotic administrations and the types of antibiotics used can influence the risk of antibiotic resistance gene (ARG) expansion and ARC events in AML patients undergoing RIC. While certain antibiotics may reduce microbial diversity, they are not always linked to an increase in ARGs or ARC events.
BibTeX
@inproceedings{franklin2025characterization,
  author = {Samantha Franklin and C. Ramont and Maliha Batool and Stephanie Hunter McMahon and Pranoti Sahasrabhojane and John C. Blazier and Dimitrios P. Kontoyiannis and Yang Ni and Jessica Galloway-Peña},
  title = {{Characterization of Antibiotic Administration Factors Associated with Microbiome Disruption and Subsequent Antibiotic-Resistant Infection and Colonization Events in Acute Myeloid Leukemia Patients Receiving Chemotherapy}},
  booktitle = {Antibiotics},
  year = {2025},
  doi = {10.3390/antibiotics14080770},
}
Figure