Inferior survival after microbiota injury: A multicenter allo-HCT study

7015 Background: Relationships between microbiota composition and clinical outcomes following allogeneic hematopoietic cell transplantation (allo-HCT) have been described in single-center studies. Geographic variations in human microbial communities and differences in clinical practices across institutions raise the question of whether these associations are generalizable. We report the first multi-center study of the intestinal microbiota in allo-HCT. Methods: Intestinal communities in 8,768 fecal samples from 1,362 allo-HCT patients at 4 centers on 3 continents were profiled by 16S sequencing. Associations between microbiota composition and clinical outcomes were analyzed with proportional-hazards analysis in an observational study. Results: We observed reproducible patterns of microbiota injury characterized by loss of diversity and domination by single taxa. Low diversity in the neutrophil engraftment period was reproducibly associated with increased risk of death (multivariate HR 0.48 [0.30-0.77] p = 0.002 in the largest cohort). These reductions in OS were in part due to an increased risk of transplant-related mortality and graft-vs-host disease. Baseline pre-HCT samples already bore evidence of microbiome disruption; low diversity prior to transplantation was associated with poor survival. A bacterial-composition risk score that was trained in one cohort predicted mortality in the other three cohorts (multivariate HR 1.42 [1.04-1.93] p = 0.03), indicating that not only a diversity metric but also a signature of specific bacterial abundances is informative about post-HCT mortality risk across independent institutions. Conclusions: We demonstrate a relationship between microbiota and survival after allo-HCT that is independent of transplant center and geographic location. The diversity of clinical practices across institutions imposed significant heterogeneity in the study, yet we observed reproducible microbiota injury patterns and associations with outcomes. This concordance suggests that approaches to manipulate the intestinal microbiota in allo-HCT may be generalizable.