Microbiome-based prediction of allogeneic hematopoietic stem cell transplantation outcome

This study aimed to investigate the use of stool and saliva microbiome features as biomarkers for predicting the timing of seven post-HSCT events, utilizing the RATIO (suRvival Analysis lefT barrIer lOss) model specifically tailored for HSCT cohorts characterized by competing events. Through longitudinal analysis of a comprehensive dataset comprising 514 stool and 1291 saliva samples collected from allogeneic HSCT recipients (N=204), coupled with detailed clinical information over the initial year post-HSCT, we observed notable similarities in the microbial dysbiosis trends between the gut and oral cavity following HSCT. By employing the RATIO model, we significantly enhanced the accuracy of survival analysis predictions for both the stool and saliva cohorts. Furthermore, we identified critical time points that allow for an accurate assessment of the risk associated with each complication. Utilizing SHAP (SHapley Additive exPlanations) analysis, we determined the key microbial taxa contributing to each complication prediction. This study provides valuable insights into the potential utility of the microbiome as a predictive tool for post-HSCT complications, highlighting the prospects for targeted interventions and personalized treatments to improve patient outcomes in the context of HSCT.