Duration of Hospitalization is Associated with the Gut Microbiome in Patients Undergoing Hematopoietic Stem Cell Transplantation: Early Results from a Randomized Trial of Home Versus Hospital Transplantation.

Home-based hematopoietic stem cell transplantation (HCT) is an innovative care model with growing interest, but its impact on the gut microbiome remains unexplored in a randomized setting. We present interim results from the first randomized controlled trials (RCT) evaluating the effect of HCT location-home versus hospital-on gut microbial diversity and antimicrobial resistance (AMR) gene carriage. We hypothesize that patients randomized to undergo home HCT would have higher gut taxonomic diversity and lower AMR gene abundance compared to those undergoing standard hospital HCT. We analyzed stool samples from the first 28 patients enrolled in ongoing Phase II RCTs comparing home (n = 16) and hospital (n = 12) HCT at Duke University using shotgun metagenomic sequencing to compare taxa and AMR gene composition between groups. We also performed a secondary analysis comparing patients who received transplants at outpatient infusion clinics versus inpatient standard HCT to evaluate the influence of hospitalization duration. In the primary RCT analysis, taxonomic and AMR gene α- and β-diversity were comparable between home and hospital groups, reflecting similar durations of hospitalization despite group allocation. In contrast, secondary analyses demonstrated that patients transplanted in outpatient infusion clinics who experienced significantly reduced hospitalization had higher gut taxonomic α-diversity and differential β-diversity, although AMR gene diversity remained unchanged. In summary, randomization by transplant location did not impact the gut microbiota to the same extent as the duration of hospitalization, although secondary analyses were heavily confounded. Even when taxonomic differences were observed, AMR genes were similar between groups. This RCT represents a novel investigation into how care setting influences the gut microbiome during HCT. Our findings suggest that hospital duration, rather than randomization allocation alone, is the primary driver of microbial disruption. These results underscore the potential for reducing hospital duration to mitigate microbiome injury, thereby informing future interventions to reduce infection risk and improve patient outcomes.