41. Impact of Gut Microbiome Changes on Hematopoietic Stem Cell Transplantation Outcomes in Children

Background In adults undergoing allogeneic hematopoietic cell transplantation (HCT), higher gut microbiome diversity is associated with reduced bloodstream infections (BSI) and improved overall survival (OS). Rifaximin prophylaxis in adult HCT helps to maintain microbiome diversity. We examine changes in microbiome in a cohort of pediatric patients undergoing HCT. Methods Patients were enrolled in an institutional biorepository (n=82) with a subset enrolled in an ongoing trial using rifaximin (n=21) between 2013–2020. All patients had HCT for a hematologic malignancy, using myeloablative conditioning. Patients in the rifaximin trial received rifaximin starting 7 days before HCT (D-7) through D+28, otherwise, no prophylactic antibiotics were used. Systemic antibiotic timing was categorized as none, early (≤ Day 0, day of HCT), and late (> D0). We performed 16s rRNA sequencing from stool for 73 subjects, at baseline (D-7), and weekly through D+28 (engraftment). Microbiome diversity was assessed by Shannon index. Results Median age was 9 years (range 1–20), 59% male, 41% Caucasian and 29% Black. There were no differences in BSI or mortality by age, sex, or race. Microbiome diversity changed significantly over time (p=0.008). Drop in diversity was most notable in patients who had early antibiotics (Mean=1.4, CI -0.15, 2.94, p=0.077). Higher diversity was seen when patients received none or late versus early antibiotics, but this was not statistically significant (Figure 1, p=0.23). Piperacillin-tazobactam was used empirically in 91% of patients. OS at 1 year was 88.5% (CI 68.4%, 96.1%) for patients with high (≥ median) D+28 diversity compared to 60% (CI 38.4%, 76.1%) for patients with low diversity (Figure 2, p=0.018) Only 1 of 21 (4.8%) in the rifaximin group developed a BSI with a gut bacterium compared to 8 of 61 (13.1%) not on rifaximin within the first 30 days (trial enrollment ongoing). Figure 1. Effect of systemic antibiotic timing on microbiome diversity over time. Figure 2. One-year overall survival of patients with high (>2.77) versus low (<2.77) diversity defined by median Shannon-Index. Conclusion We have shown a significant correlation between engraftment microbiome diversity and 1-year OS. Early antibiotic exposure was detrimental to microbiome diversity. Approaches to preserve microbiome diversity and prevent BSI are likely to improve HCT outcomes. Our ongoing trial using rifaximin will provide preliminary data regarding this approach. Disclosures All Authors: No reported disclosures