1645-P: Assessing the Impact of a Plant-Based Diet for Diabetes Prevention: A Pilot Study

Introduction and Objective: Skeletal muscle is a major site of glucose disposal, and interventions increasing insulin sensitivity without weight loss are impactful for diabetes prevention. We studied whether a 4-week, plant-based diet in adults with prediabetes could improve glucose metabolism without energy restriction and completed exploratory analyses of sex-based responses. Methods: We enrolled adults (n=6) aged 30-55 years with prediabetes and obesity habitually following a Western diet. Participants consumed the provided plant-based intervention diet for 4 weeks. OGTT enabled measurement of the Matsuda Insulin Sensitivity Index and HOMA-IR, and android fat was measured by DXA; these were analyzed with paired t-tests. Skeletal muscle biopsies were collected before and after for histology analyses and bulk RNA sequencing. Results: Acceptability of the plant-based diet was high (mean meal taste rating 8.1/10), and adherence was 99%. HOMA-IR decreased significantly (7.27 ± 1.93 vs. 5.58 ± 2.30; p= 0.02), and Matsuda insulin sensitivity tended to increase (1.25 ± 0.85 vs. 1.63 ± 1.13; p=0.07). Android fat mass decreased (5,578 ± 1,200 vs. 5,282 ± 1,232 g; p= 0.03), and total body weight remained unchanged (112.4 ± 13.7 vs. 110.2 ± 12.5 kg; p= 0.13). Metabolic and body composition changes were not attributable to weight loss. Women showed greater increases in the Matsuda index when compared with men (0.70 ± 0.30 vs. 0.05 ± 0.10, respectively; p=0.024). Conclusion: A 4-week, plant-based diet caused significant improvements in HOMA-IR and android fat without weight loss in adults with prediabetes. These findings point to weight-independent metabolic adaptation mechanisms and highlight the importance of considering sex-specific responses to diet interventions. Ongoing histological and bulk RNA sequencing analyses of skeletal muscle samples will further elucidate tissue-level mechanisms underlying the changes. This pilot study demonstrates feasibility, high adherence, and clinically meaningful metabolic effects to inform larger trials for diabetes prevention. Disclosure K.R. Anderson: None. P. Kern: None. Z. Leicht: None. D.E. Long: None. J.L. Fry: None. Funding Gene Haas FoundationBarnstable Brown Diabetes Center National Institutes of Health (UL1TR001998)