SAT-639 - Comparing Different Metabolic Indices for Predicting Conversion to Diabetes in the Sangre Por Salud Latino Cohort

Objective:
Latinos are at increased risk for metabolic diseases including type 2 diabetes (T2DM), when compared to White populations. Early identification of patients at risk for progression to T2DM in this population is critical. We utilized a biobank consisting of nearly 4000 Latino adults followed over 10 years. The aim of this study was to identify which baseline risk factors—hemoglobin A1c (A1c), fasting plasma glucose (FPG), 2-hour postprandial glucose (2hPPG), family history of diabetes, or smoking status—were most predictive of progression to T2DM in this population.

Methods:
We utilized a retrospective cohort design looking at the Sange Por Salud (SPS) biobank of Latino adults (age: 18–85 years) with baseline FPG, 2hPPG, and A1c, being followed for 10 years with confirmation of incident T2DM. Patients with baseline diabetes or less than 3 years of follow-up were excluded. Patients who developed a new diagnosis of T2DM were matched to non-diabetic controls (1:3) based on age (± 5 years) sex, and BMI (± 3 units). Independent variables including baseline A1c, 2hPPG, FPG, family history of diabetes, and current smoking status were compared using conditional univariate logistic regression to analyze the association between each measure and progression to T2DM. Time-to-event analysis was conducted using univariate Cox proportional hazards regression (with stratification for matching) for each variable.

Results:
283 T2DM patients were identified and matched to 816 controls. Mean (SD) age was 44.1 (11.2) years, BMI 32.0 (5.5) and 79% were female. 56% of T2DM patients had a family history of diabetes versus 46% of controls.
A1c was most predictive by logistic / time-to-event model (C-statistic 0.78, C-index 0.77), followed by FPG (C-statistic 0.71, C-index 0.71) and 2hPPG (C-statistic 0.68, C-index 0.67). Family history (C-statistic 0.56, C-index 0.56) and current smoking (C-statistic 0.51, C-index 0.51) were poor predictors of conversion to T2DM. Odds ratio (95% confidence interval) for A1c (per 0.5 unit increase) was 5.89 (4.26, 8.14), FPG (per 10.0 unit increase) was 2.15 (1.80, 2.57) and 2hPPG (per 10.0 unit increase) was 1.21 (1.16, 1.27).
Median time to T2DM was 55.4 months (≈ 4.5 yrs) amongst the cases and median follow-up for controls was 94.9 months (≈ 7.9 years).

Conclusion:
In this Latino cohort, baseline A1c was the strongest predictor of conversion to T2DM, when compared to the other metabolic and clinical factors. This study supports the importance of dynamic glucose monitoring for early identification of high-risk individuals for progression to T2DM. Our group is now working to compare the above indices to validated polygenic risk scores for prediction of progression to type T2DM.

*Unless otherwise noted, all abstracts presented at ENDO must not be released to the press or the public until the date and time of presentation. For oral presentations, the abstracts are embargoed until the session begins. The Endocrine Society reserves the right to lift the embargo on specific abstracts that are selected for promotion prior to or during ENDO.*