Antibiotic Resistance Genes Peak before Hematopoietic Stem Cell Transplant in Pediatric Blood Cancer Patients

Background- Infections, especially from multi-drug resistant organisms (MDROs), are a leading cause of death in pediatric hematopoietic cell transplant (HCT) recipients. The gut microbiome can harbor MDROs, with pathogenic bacteria found in stool prior to bacteremia in up to 50% of cases (Perez et al. 2020 Hematology, Transfusion and Cell Therapy; Tamburini et al. 2018 Nature Medicine). Additionally, more antibiotic resistance genes (called resistome) in the gut bacteria after HCT is linked to an increased risk of acute graft versus host disease (aGvHD) (D’Amico et al. 2019 Scientific Reports). Understanding changes in the gut resistome between HCT and neutrophil engraftment is critical to hone antibiotic stewardship efforts and reduce complications.

Methods- Stool samples from 166 HCT recipient children were collected before HCT, 30 days after, and post-neutrophil engraftment at 2 transplant centers (approved by IRBs at St. Jude Children’s Research Hospital and University of Bologna). These children, median age 8.4, mainly underwent HCT for Acute Myeloid Leukemia (AML, 38%), with others having Acute Lymphoblastic Leukemia (ALL, 33.7%), other malignancies (11.4%), or non-malignant diseases (16.9%). Most received myeloablative conditioning (65.7%), with the rest receiving reduced intensity conditioning (33.7%). Graft was sourced from a matched related (45.8%); or a matched unrelated (15.7%) or a haploidentical (38%) donor. Graft was collected either from bone marrow (57.8%) or peripheral blood stem cells (41.6%). Ex-vivo T cell depletion was performed in 50% of HCTs. Patients received multiple antibiotic doses (median 42 doses) for fever and neutropenia, infections, and prophylaxis (trimethoprim-sulfamethoxazole at all institutions; fluoroquinolones exclusively at St Jude). For comparison, stool samples from patients were compared to stool samples from 347 healthy children (median age 3.2, with approval from Seattle Children’s Hospital IRB) and 91 children undergoing treatment for ALL (median age 9, with approval from St. Jude Children’s Research Hospital IRB). DNA from these samples underwent Illumina metagenome sequencing, yielding 25 million reads/sample. Antibiotic resistance genes (ARGs) were identified using MEGARes (v 3.0). ARG abundance was adjusted for sequencing depth and gene length, and changes were analyzed based on antibiotic exposure using linear mixed models.

Results- Between HCT and neutrophil engraftment the gut resistomes of HCT recipients showed little change. Surprisingly, common antibiotics like Glycopeptides, Pseudomonal Cephalosporins, Aminoglycosides and Fluoroquinolones, did not increase the resistance in the patients’ gut microbiomes. However, antibiotics with anaerobic activity increased MDRO-associated genes including Extended Spectrum Beta Lactamases (ESBL), vanA, and multi-drug efflux pumps. Comparing these patients with healthy children showed that significant changes in resistance genes had occurred before HCT. Analysis of leukemia patients undergoing treatment found most increases in antibiotic resistance genes happened prior to HCT.

Conclusions- These results suggest that by the time pediatric blood cancer patients reach HCT, saturation of resistance genes in their gut microbiome has already occurred and additional antibiotic exposure has only a minor impact. Factors like radiation exposure, conditioning agents, inflammation, turnover, and nutrition may also impact the resistome. Therefore, antibiotic stewardship before HCT is crucial to prevent MDROs in pediatric patients’ gut microbiomes.