Personalized Nutrition Recommendations Improve Plasma Metabolite Concentrations Related to Dietary Intake
Abstract Objectives The objectives of our proof-of-concept study was to assess the efficacy of personalized nutrition interventions on diet and chronic disease risk. Methods Fasting plasma samples were collected at day 1 and day 100 of a cohort of 148 adults (aged 23–65y) volunteers with a median (range) BMI of 25.8 (17.2–48.3). At both time points 119 metabolites were quantitated using LC-MS/MS. Based on their metabolite concentrations and dietary preferences, each participant received their own personalized nutrition recommendations through an AI-assisted online platform and were advised to follow the recommendations for 100 days. Plasma metabolite concentrations from Day 1 and Day 100 were compared using a paired t-test with Holm-Bonferroni correction (P < 0.05). Results After 100 days, statistically significant changes in acylcarnitine, phosphatidylcholine and amino acid concentrations indicated participants had increased their intakes of omega-3 fats and whole grains and decreased their intakes of saturated fat. For example, Betaine, a biomarker of whole grain intake, increased significantly in concentration from Day 1 to Day 100 [mean (SD): 34.3 (13.2) to 45.3 (15.6) umol/l]. Overall, 55 of the analyzed 119 metabolites’ (46%) concentrations had previously been linked to dietary intake according to a systematic literature search that was used to generate evidence-based personalized nutrition recommendations. Concentrations of 33 (60%) of these 55 metabolites changed significantly. The majority [26 (79%)] of the metabolite concentrations changed in a predicted manner consistent with the literature, particularly those metabolites associated to chronic disease risk. For example, a significant decrease in asymmetric dimethylarginine concentration [mean (SD): 0.54 (0.15) to 0.48 (0.12) umol/L] which is understood to result in a decreased cardiovascular disease risk. Conclusions Adherence with an evidence-based personalized nutrition plan based on a panel of serum metabolomic data can significantly modify serum metabolite concentrations in a direction that can reduce the risk of chronic disease. Further analysis in how these changes relate to chronic disease risk is warranted. Funding Sources This study was supported by Mitacs and Molecular You.
