710-P: Comparing Models for Predicting Diabetes Progression and Cardiovascular Disease in Risk Stratification and Drug Selection for Prediabetes

Le résumé fourni par la source

Introduction and Objective: To compare models predicting diabetes (DM) progression or incident cardiovascular disease (CVD) in risk stratification and treatment heterogeneity identification in patients with prediabetes. Methods: This study included 2,607 patients from the Diabetes Prevention Program (DPP) and the Diabetes Prevention Program Outcomes Study (DPPOS). Patients with prediabetes were stratified based on DM and CVD progression models with optimal predictive performance. Cox regression and logistic regression were used to evaluate the incidence of diabetes, microvascular, and cardiovascular events across different risk and intervention groups. Results: The machine learning-based diabetes progression model in US people (MLPR-US model) achieved an ROC AUC of 0.79 (95% confidence interval [CI], 0.72-0.86) for predicting diabetes progression, while the PREVENT model showed an ROC AUC of 0.73 (95% CI, 0.71-0.76) for predicting CVD events. The risk of diabetes progression, cardiovascular events, and microvascular events increased by 250% (p<0.001), 27% (p=0.043), and 50% (p=0.004) in the DM high-risk group versus their controls. The corresponding figure was -11% (p=0.276), 284% (p<0.001), and 287% (p<0.001) in the CVD high-risk group. A significant treatment-by-group interaction was observed for diabetes progression in both DM and CVD risk groups, with DM high-risk patients benefiting from metformin and lifestyle therapy, while CVD high-risk patients benefited only from lifestyle therapy. No such interaction was found for microvascular or CVD events. Conclusion: The CVD model identified high-risk individuals for cardiovascular and microvascular events but not diabetes progression. The DM model highlighted patients with prediabetes needing lifestyle and metformin therapy to prevent diabetes progression. Disclosure S. Wang: None. Q. Huang: None. Y. Luo: None. L. Ji: None. X. Zou: None. Funding National Natural Science Foundation of China (T2341011), Noncommunicable Chronic Diseases-National Science and Technology Major Project (2023ZD0508800), Beijing Science and Technology Program (Z181100001618010)

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.