DENTAL BIOFILM MICROBIOTA DYSBIOSIS PREDICTS AGVHD RISK AFTER ALLOGENEIC HEMATOPOIETIC STEM-CELL TRANSPLANTATION

Allogeneic hematopoietic stem-cell transplantation (allo-HSCT) is the best curative treatment for several hematologic disorders. Acute Graft Versus Host Disease (aGVHD) occurs in half of allo-HSCT patients and it is the second leading cause of non-relapse mortality. Recently, aGVHD onset was causally linked to gastrointestinal microbiota due to the modulation of intestinal epithelium turnover and immune homeostasis by bacterial metabolites. However, those studies have focused on the intestinal microbiota (IM), neglecting other bacterial-rich gastrointestinal sites, such as the mouth. The dental biofilm microbiota (DBM) is among the most diverse human microbiotas and, like the IM, interacts with host cells and modulates immune homeostasis. In this study, we characterized changes in the DBM of patients during allo-HSCT and evaluated whether DBM was associated with the aGVHD risk. We collected dental biofilm samples from 30 patients undergoing allo-HSCT and characterized their DBM at 3 timepoints (preconditioning, aplasia and engraftment) by 16S rRNA sequencing. Fifteen patients developed aGVHD (grade II-IV) and, among those, 6 developed severe aGVHD (grade III-IV). We observed clear DBM dysbiosis during the transplantation course, including loss of diversity, reduction of commensal (e.g., Actinomyces) and increase of potentially pathogenic bacterial genera (e.g., Enterococcus). At preconditioning, a higher relative abundance of Streptococcus and Corynebacterium genera were associated with increased aGVHD risk (HR 2.89, 95%CI 1.07 7.79, P = 0.036; HR 2.74, 95%CI 1.05 7.15, P = 0.04, respectively), while a higher relative abundance of Veillonella genus was associated with decreased aGVHD risk (HR 0.24, 95%CI 0.08 0.7, P = 0.009). We also observed that 6/30 patients showed a sudden expansion of the Enterococcus genus (relative abundance <1% at preconditioning to ≥30% at aplasia or engraftment). This expansion, known as bloom, has been previously associated with an increased risk of aGVHD in IM studies. Likewise, DBM Enterococcus faecalis bloom was associated with an increased risk of aGVHD (HR 4.07, 95%CI 1.82 9.14, P < 0.001) and an increased risk of severe aGVHD (HR 6.82, 95%CI 1.48 31.41, P = 0.014). This was the first study evaluating DBM in the allo-HSCT context. Prospective trials are needed to confirm the potential of DBM dysbiosis as a predictive biomarker of aGVHD risk and test whether interventions to correct DBM dysbiosis could reduce aGVHD risk.