A Pilot Study of Donor Enteral Human Milk to Modulate the Gut Microbiome in Children Receiving Stem Cell Transplant

Background: The gut microbiome is an important immune modulator, and previous work has shown important changes in the gut microbiome associated with the occurrence of GVHD.  Studies in premature babies at risk for necrotizing enterocolitis have shown that enteral human milk reduces gut inflammation and bacterial translocation (Morrow et al, 2012).  We hypothesized that enteral human milk given in the peri-transplant period would be well tolerated and would modify the microbiome and reduce inflammation. 

Methods: In a pilot study, we treated 10 children <2 yrs with donor enteral milk, and collected stool samples for microbiome analysis and blood samples for analysis of inflammatory markers.  Milk was obtained from the Mother’s Milk Bank of the Northeast.  Milk donors underwent standard donor screening and milk was pasteurized before use. As prebiotic and anti-inflammatory benefits of are thought to derive from human milk oligosaccharides produced by fucosyltransferase2 (FUT2) and related gene biosynthesis enzymes (which are highly polymorphic), all milk was screened to ensure the presence of these oligosaccharides.

Results: Ten children were treated with milk, 8 boys and 2 girls.  Ages at transplant ranged from 6 to 40 months. Diagnoses included non-malignant hematopoietic disorders (n=8) and leukemia (n=2).  Three children had sibling donors, 1 a matched aunt and 6 unrelated donors; all received bone marrow grafts. Treated cases were compared with 4 young children transplanted in the same time period who received standard nutrition under the supervision of the BMT dietician.  Milk was given to provide 40% of caloric needs (as breast milk cannot provide all nutritional needs for children older than 6 months), from day -3 through day +14 after transplant.  Enteral human milk was well tolerated with all children receiving all the milk planned for administration.  Microbiome analysis showed no differences between study groups at the start of transplant.  Significant differences were observed between milk treated and control patient microbiomes at day 10 after transplant, when, consistent with microbiome changes previously reported with GVHD risk (Jenq et al 2012), control patients had significantly increased facultative anaerobes Streptococcaceae and Actinomycetaceae compared to milk fed patients. Measurement of sIL2r levels showed elevation of levels at day 14 in controls compared with cases.

Conclusions: This preliminary study suggests a possible role for human milk in modulation of the microbiome and reduction of a pro-inflammatory environment and a randomized study is now in progress.  Synthetic human milk oligosaccharides have the potential to extend this strategy to adults, and are currently being tested in a mouse model of GVHD.