1812-LB: Association of Plasma Ceramides with Progression to Diabetes in the Diabetes Prevention Program (DPP) and Outcomes Study (DPPOS)

Objective: Ceramides (CER) have been implicated in the pathogenesis of type 2 diabetes (T2D). Here, we assessed baseline plasma CER profiles in relation to incident T2D in the DPP/DPPOS.

Methods: Adults at high risk for T2D were randomized in DPP to intensive lifestyle, metformin, or placebo to compare their effects on diabetes incidence, and followed long-term in DPPOS. For this DPP ancillary study, we evaluated 198 randomly selected participants from the DPP placebo group, comprising 100 who had progressed to T2D (P) by DPPOS Year 11 (~15 years after randomization) and 98 non-progressors (NP), frequency matched by sex, age group and race/ethnicity. Plasma levels of 30 CER species, including 10 monohexosylceramides (MHC) and 10 lactosylceramides (LAC), were assayed in DPP baseline specimens using LC-MS/MS.

Results: At DPP baseline, mean (<u>+</u> SD) age (51.6 <u>+</u> 8.5 y vs 51.4 <u>+</u> 9.0 y), BMI (34.5 <u>+</u> 7.7 kg/m2 vs. 34.1 <u>+</u> 6. 1 kg/m2), waist circumference, lipids, hsCRP and adiponectin levels were similar, but fasting plasma glucose (FPG) (110 <u>+</u> 8.8 mg/dl vs 104 <u>+</u> 6.3 mg/dl, P<0.0001) and 2-hour plasma glucose (2hPG) (167<u>+</u> 15.8 mg/dl vs 160 <u>+</u> 16.4 mg/dl, P=0.0007) were higher in P vs. NP. Baseline levels of total CER, long-chain CER C20:0 - C26:0 and MHC C22:0 were higher, and LAC C18:0, C20:0 and C26:1 were lower, in P vs. NP, adjusted for age, race, sex, BMI, FPG and 2hPG (P’s = 0.02 to <0.0001). The ratios of CER C26:0/C26:1 (P=0.005) and MHC C24:0/C24:1 (P= 0.008) were higher in P vs. NP. In logistic regression models, baseline plasma ratios of CER C26:0/C26:1 predicted incident T2D: OR 1.26 per 0.1 unit, 95% CI 1.09-1.49, P=0.004, adjusted for baseline age, race, sex, BMI, FPG, and 2hPG

Conclusion: Higher baseline plasma levels of long-chain CER and MHC, and lower levels of LAC, were associated with incident T2D during ~15 years of follow-up in the DPP/DPPOS. The saturated-to-unsaturated CER C26:0/C26:1 ratio is a potential biomarker of incident T2D risk among individuals with pre-diabetes.