Genetic Risk Scores for Obesity and the Effectiveness of a Diet and Exercise Intervention Program: A Historical Cohort Study.

Abstract

The role of genetic risk scores (GRS) in predicting responses to the obesity interventions in Japanese individuals remains underexplored. We aimed to examine GRS and introduce and evaluate a novel efficiency score derived from data envelopment analysis (DEA) for predicting the outcomes of a combined diet and exercise program. Data from 145 individuals who completed an 8- to 12-week low-carbohydrate diet and resistance training intervention were retrospectively analyzed. GRS were calculated from 75 SNPs identified in a previous Japanese genome-wide association study. DEA was applied to generate efficiency scores using GRS as an input reflecting genetic load and initial BMI and body fat percentage as outputs reflecting phenotypic status. Linear regression analyses were performed to compare the predictive utility of GRS and efficiency scores. BMI and other measures decreased following the intervention. Linear-regression analyses revealed that DEA-derived efficiency scores strongly predicted the change in BMI percentage, whereas GRS did not. Weight loss was achieved independently of GRS. GRS-derived efficiency score offered superior prediction of intervention effectiveness to that of GRS alone, suggesting that while GRS based on cross-sectional data may not predict responsiveness to interventions, integrated metrics also incorporating phenotypic data show potential for personalizing weight management strategies, pending further validation.