AUTHORS
Elisha Pinker, MD, Teng Fei, PhD2, Anqi Dai, MS3, Peter Adintori, MS RD-AP CSO CDN CNSC4, Miguel-Angel Perales, MD5, Antonio Gomes, PhD6, Sergio Giralt, MD7, Nicholas Waters, PhD8, Marcel R.M. van den Brink Jr, M.D., Ph.D.9, Jonathan Peled, MD/PhD10
AFFILIATIONS
1 Memorial Sloan Kettering Cancer Center, New York, New York, United States, NYU Grossman School of Medicine, New York, New York, United States
2 Memorial Sloan Kettering Cancer Center, New York, New York, United States
3 Memorial Sloan Kettering Cancer Center, New York, New York, United States
4 Mount Sinai Queens Cancer Center, New York, New York, United States
5 Memorial Sloan Kettering Cancer Center, New York, New York, United States, Weill Cornell Medical College, New York, New York, United States
6 City of Hope National Medical Center, Duarte, California, United States
7 Memorial Sloan Kettering Cancer Center, New York, New York, United States, Weill Cornell Medical College, New York, New York, United States
8 Memorial Sloan Kettering Cancer Center, New York, New York, United States
9 City of Hope National Medical Center, Duarte, California, United States
10 Memorial Sloan Kettering Cancer Center, New York, New York, United States
ABSTRACT
Introduction Gut microbiome disruption commonly accompanies alloHCT; higher α-diversity predicts favorable outcomes. Longitudinal dietary intake paired with stool sampling provides a unique opportunity to study diet-microbiome interactions during profound immune and microbial perturbation. We hypothesized that micronutrient and caloric intake are associated with microbiome composition and GVHD.
Methods We conducted a retrospective study of 173 alloHCT recipients who recorded dietary intake from over 10,000 bedside-recorded meals. Foods were mapped to USDA's Food and Nutrient Database for Dietary Studies (FNDDS), from which 52 micronutrients were quantified. Stool samples (n=1,009) from 158 patients were profiled by 16S rRNA gene sequencing. Of these, 153 patients had overlapping dietary and microbiome data and comprised the primary analytic cohort. Micronutrient diversity was calculated as Shannon diversity of calorie-adjusted nutrient densities across four windows (pre-transplant [day −7 to 0] and weeks 1–3). Microbiome α-diversity was quantified using the Simpson reciprocal index (IQR 3.0–9.7) and analyzed using generalized estimating equations (GEE), which account for repeated stool and meal observations per patient. Analyses were adjusted for non-prophylactic broad-spectrum antibiotic (BSA) exposure, defined as exposure to any BSA between day −7 and day +30. Enterococcus domination was defined as >30% relative abundance in any post-transplant sample.
Results
Nutrient composition was mainly comprised of fatty acids (oleic acid, 30.6%; palmitic acid, 19.0%; linoleic acid, 16.1%), with potassium and sodium as the largest non-lipid contributors (3.8% each). Inter-patient variability in intake (coefficient of variation) was highest for lycopene, theobromine, and omega-3 fatty acids including docosahexaenoic acid (DHA), docosapentaenoic acid (DPA), and eicosapentaenoic acid (EPA), as well as vitamins D, C, and K.
Caloric intake declined considerably in week 1 (pre-transplant 1,023 vs. 672 kcal), with a corresponding modest reduction in micronutrient diversity (0.227 vs. 0.202). Higher pre-transplant caloric intake was associated with greater week 1 α-diversity (β=1.02; p=0.031), persisting at day +14 (β=1.23; p=0.006), but not with α-diversity in weeks 2 or 3 (p=0.104 and p=0.131, respectively). Similarly, higher pre-transplant micronutrient diversity was associated with greater week-1 α-diversity (β=1.0; p=0.008) after adjustment for calories and antibiotic exposure. No significant association was observed in week 2 (p=0.60), though a non-significant trend was present in week 3 (β=0.54; p=0.066).
In a micronutrient-wide screen, omega-3 fatty acids were positively associated with week-1 α-diversity: DPA (β=1.66; p=0.012), EPA (β=1.51; p=0.026), and DHA (β=1.55; p=0.029). Higher pre-transplant DPA intake was independently associated with higher post-transplant Faecalibacterium abundance in week 1 (β=0.017; q<0.001). Higher micronutrient diversity was associated with lower Enterococcus domination risk (OR 0.27 comparing above-vs below median micronutrient-diversity groups; p=0.034). Although associations between micronutrient intake and GVHD outcomes were not significant after correction for multiple hypothesis testing, in an exploratory analysis, a borderline protective association was observed between higher pre-transplant micronutrient diversity and lower GI GVHD (OR 0.17; p=0.056).
Conclusions Higher pre-transplant caloric intake and micronutrient diversity were independently associated with fecal microbiota α-diversity during post-HCT neutrophil nadir. Pre-transplant DPA intake was specifically associated with preservation of Faecalibacterium, a key butyrate producer. We hypothesize that omega-3 fatty acids contribute to microbiome resilience post-HCT, and these findings support prospective investigation of omega-3 fatty acid supplementation as a strategy to preserve microbial diversity and abundance of key fermenting organisms in the peri-transplant period.
DISCLOSURES
Elisha Pinker, MD: No relevant conflicts of interest to declare.