Keywords: Biomarker, non-parametric, bootstrap simumulation
• Biomarkers have been increasingly used in drug discovery and Phase 1 studies and are becoming an essential part of clinical trials and drug development. Biomarkers have often been utilized as a tool to diagnose disease risk, evaluate treatment effect and risks, patient benefits (personalized medicine) and also, as potentially useful surrogate endpoints. • Changes in cardiovascular biomarkers and their relationship with cardiovascular outcomes have been comprehensively evaluated by clinical cardiologists to understand the underlying cardiovascular risk and identify patients who may be at risk for cardiovascular disease (CVD). Diabetes Mellitus is a major risk factor for CVD. Based on the observed glycemic and also non-glycemic (promoting weight loss, blood pressure reduction etc.) efficacy of certain antidiabetic drugs, it has been proposed that these drugs may also provide a range of cardiometabolic benefit. • In this poster, quantitative approaches to analyze the effect of two antidiabetic drugs, an SGLT2 inhibitor and a sulfonylurea, in patients with Type 2 Diabetes Mellitus (T2DM), on the CVD biomarkers known to be predictive for cardiovascular stress, inflammation and chemokines/adipokines will be presented. Remedial measures for non-normally distributed biomarker data including the use of non-parametric methods to estimate the median change and variability will be compared and contrasted with the estimates obtained via simulations using bootstrap methods.