Examining dietary and metabolomics patterns on cardiometabolic outcome variability in response to calorie restriction - PROJECT SUMMARY
Cardiovascular disease (CVD) is the leading cause of mortality and morbidity worldwide, with age being the
strongest independent non-modifiable risk factor. Calorie restriction (CR), the reduction of total food intake less
than the recommended energy requirement, is the most efficacious dietary intervention to prolong lifespan and
decrease CVD risk. In the largest clinical trial of CR (CALERIETM 2), healthy adults were randomized to either 2
years of CR (n=143) or an ab libitum (AL) (n=75) control group. The CR group achieved ~12% CR over 2 years
and had improvements in cardiometabolic health outcomes such as the change in blood pressure, blood lipids,
and fasting glucose levels from baseline to 2 years. However, preliminary analyses show large interindividual
variability in the magnitude and direction of cardiometabolic health outcomes in response to the CR impacting
intervention efficacy. There was no specific diet prescribed to CR participants in CALERIETM 2, but rather
participants were educated on how to choose their own foods to achieve satiety, mitigate hunger, and to reach
the Dietary Recommended Intakes. Preliminary analysis indeed shows that there was interindividual variability
in the change in diet quality (measured by the Healthy Eating Index) from baseline to 2 years of CR.
Consuming nutrient-rich foods such as fruits, vegetables, whole grains, and fish is beneficial for
cardiometabolic health and are consumed throughout the longest-lived populations. Thus, we hypothesize that
diet quality contributes to the cardiometabolic health outcome variability observed with the 2-year CR
intervention. We will compute novel dietary scores (e.g., Healthy Eating Index and Dietary Inflammatory Index)
from 6-day food records taken at baseline and months 12 and 24 (food records were also taken at months 6
and 18 for the CR group) to measure diet quality. The primary outcome is change in blood pressure, and
secondary outcomes are changes in blood lipids, glucose, insulin, VO2MAX, 10-y CVD risk score, and
biomarkers of inflammation and oxidative stress. Then, we will determine whether diet quality scores are
associated with the change in the primary and secondary cardiometabolic health outcomes across each
timepoint. We will also merge untargeted metabolomics data from CALERIETM 2 participants at baseline and
months 12 and 24 with the diet quality scores to identify unique metabolomic cluster signatures for each diet
score. Applying artificial intelligence and machine learning approaches, we aim to build prediction models that
will determine the most optimal dietary pattern in the context of CR to improve cardiometabolic health at the
individual level. This project will contribute to the emerging field of precision nutrition research by identifying
factors contributing to outcome variability to dietary interventions. Furthermore, I will learn invaluable skills to
advance my goal of becoming an independent clinical researcher in precision nutrition for healthy aging.
