Genetic Modifier of the Gut Microbiome, GVHD and Bacterial Translocation Following HSCT

The human gut microbiome is involved in vital biological functions such as maintenance of immune homeostasis, modulation of intestinal development and enhanced metabolic capabilities. Disturbances of intestinal microbiota have been associated with development and progression of inflammatory conditions including GVHD. Non-secretor individuals do not express the H antigen on mucosal surfaces and body fluids due to a homozygous single nucleotide polymorphism in the fucosyltransferase 2 (FUT2) gene (428G>A) and FUT2 genotype has been shown to modify the gut microbiome. We hypothesized that FUT2 genotype influences risk of GVHD and bacterial translocation following allogeneic HSCT.  

FUT2 genotype was determined in 150 consecutive patients receiving allogeneic HSCT at our center.  We abstracted clinical characteristics and outcomes from the transplant database.

Median age at transplantation was 6.31 years and male/female ratio was 97/53. Genetic analysis revealed 23%  recipients had A/A genotype (n=34), 52%  A/G genotype (n=78) and 25% G/G genotype (n=38); Hardy Weinberg equilibrium was confirmed. Hematologic malignancies comprised 30% (n=45), 29%  immune deficiencies (n=43), 22% bone marrow failure (n=33), 12% hemophagocytic lymphohistiocytosis (HLH) (n=18), 4% metabolic diseases (n=6), 3% hemoglobinopathies (n=4), and one patient with Evan's syndrome. 54% received myeloablative conditioning (n=81), while 46% received reduced intensity conditioning (n=68). Stem cell source was bone marrow in 81% (n=122), peripheral blood stem cells in 11% (n=16), cord blood in 7% (n=11), and one patient received both bone marrow and cord blood from a sibling donor. Among donors, 29% were matched sibling donors (n=43), 49%  matched unrelated or other family member donors (n=74) and 22% were mismatched donors (n=33). Acute GVHD occurred in 36% of patients (n=54). Cumulative risk of any acute GVHD varied by FUT2 genotype with decreased risk in those with A/A genotype and increased risk in those with G/G genotype (p=0.04) (Fig. 1). A/A genotype (OR=0.4 p-value=0.046), myeloablation (OR=1.99 p-value=0.029) and matched sibling donor (OR=0.41 p-value=0.026) were identified to be significant GVHD risk factors in multivariate analysis.  Bacteremia occurred in 34% of patients (n=51), and in contrast to our findings in GVHD cumulative incidence was increased in A/A genotype (p-value=0.01) (Fig. 1). A/A genotype (OR=3.94 p-value=0.0047) and A/G genotype (OR=2.46 p-value=0.05) were associated with increased risk in multivariate analysis.

FUT2 genotype influences risk of acute GVHD and bacteremia following HSCT. We hypothesize that the mechanism involves altered composition and diversity of gut microbiome, and limited data indicate increased diversity of the gut microbiome in the A/A genotype, but this requires additional studies.