Participants were monitored for adverse events, serious adverse events and death throughout the study.

Safety analysis set: The safety analysis set included all randomized participants who received at least one dose of study drug.

Two axial, ECG-gated, steady state free precession (SSFP) 'cine' images were acquired during breath-hold to determine aortic distensibility. The first image was obtained at the level of the right pulmonary artery through the ascending and proximal descending aorta and the second through the distal aorta below the diaphragm. Imaging of the aorta also enabled evaluation of the plaque burden and additional vascular function measures.

Only participants from the imaging analysis set, who had evaluable data at both baseline and the given post-baseline time point, were included in the analysis for that post baseline time point. The imaging analysis set included randomized participants who received at least one dose of study medication.

For assessment of atherosclerotic plaque burden of the aorta, vessel wall images of the aorta were acquired with an ECG gated double-inversion recovery (black blood) fast spin echo sequence applied breath-holding. Using an oblique sagittal image of the aorta as a pilot, serial axial images were acquired to cover a section of the descending thoracic aorta. The midpoint of the right pulmonary artery in cross section was used as the anatomical reference for the first slice in baseline and follow-up scans. For assessment of the atherosclerotic plaque burden in the carotids, vessel wall images were acquired with an axial ECG gated PD (proton density) weighted black blood sequence. The carotid bifurcation was used as the anatomical reference for all three imaging time points (baseline, 12 weeks, 48 weeks) with axial slice planes acquired below the bifurcation region. The mean values reported here for the carotid are reported for the proximal common carotid region.

Only participants from the imaging analysis set, who had evaluable data at both baseline and the given post-baseline time point, were included in the analysis for that post baseline time point. The imaging analysis set included randomized participants who received at least one dose of study medication.

Utilizing the SphygmoCor Device, ECG leads placed at the carotid and femoral arteries provided the measure of the pulse wave at that particular arterial location. The distance between the two vascular beds divided by the pulse wave time shift provided a measure of the pulse wave velocity.

Only participants from the imaging analysis set, who had evaluable data at both baseline and the given post-baseline time point, were included in the analysis for that post baseline time point. The imaging analysis set included randomized participants who received at least one dose of study medication.

During the carotid MRI acquisition, in addition to the PD weighted ECG gated double inversion fast spin echo sequences T1 and T2 weighted sequences were acquired. In combination with the PD weighted images, the multi-contrast images were analyzed to determine regions of interest with contrast patterns consistent with the presence of necrotic lipid core, calcification and fibrous tissue in participants who had complex carotid plaque present in the bifurcation region.

Only participants from the imaging analysis set, who had evaluable data at both baseline and the given post-baseline time point, were included in the analysis for that post baseline time point. The imaging analysis set included randomized participants who received at least one dose of study medication.

Arterial strain was computed directly from the cine SSFP images and the change in lumen diameters over the cardiac cycle. The value was independent of pulse pressure and is unitless ratio derived from the maximum to minimum lumen diameters diastole and systole, respectively..

Only participants from the imaging analysis set, who had evaluable data at both baseline and the given post-baseline time point, were included in the analysis for that post baseline time point. The imaging analysis set included randomized participants who received at least one dose of study medication.

Blood samples were collected to analyze hsCRP.

Only participants from the PD analysis set, who had evaluable data at both baseline and the given post-baseline time point, were included in the analysis for that post baseline time point. The PD analysis set included randomized participants who received at least one dose of study medication.

Blood samples were collected to analyze fasting plasma glucose.

Only participants from the PD analysis set, who had evaluable data at both baseline and the given post-baseline time point, were included in the analysis for that post baseline time point. The PD analysis set included randomized participants who received at least one dose of study medication.

Blood samples were collected to analyze HbA1c.

Only participants from the PD analysis set, who had evaluable data at both baseline and the given post-baseline time point, were included in the analysis for that post baseline time point. The PD analysis set included randomized participants who received at least one dose of study medication.

Blood samples were collected to analyze the 2 hour glucose post OGTT.

Only participants from the PD analysis set, who had evaluable data at both baseline and the given post-baseline time point, were included in the analysis for that post baseline time point. The PD analysis set included randomized participants who received at least one dose of study medication.

Blood samples were collected to analyze beta cell function. Beta cell function was calculated by the Homeostasis Model Assessments (of beta cell function (HOMA-B) as follows: HOMA-B: The product of 20 and basal insulin (µU/mL) levels divided by the value of basal glucose (mmol/L) concentrations minus 3.5 [i.e., HOMA-B = 20*basal insulin/(basal glucose-3.5)].

Only participants from the PD analysis set, who had evaluable data at both baseline and the given post-baseline time point, were included in the analysis for that post baseline time point. The PD analysis set included randomized participants who received at least one dose of study medication.

Blood samples were collected to analyze insulin resistance. Insulin resistance was calculated by the Homeostasis Model Assessments of insulin resistance (HOMA-IR)) as follows: HOMA-IR: The product of basal glucose (mmol/L) and insulin (µU/mL) levels divided by 22.5 [i.e., HOMA-IR = basal glucose*basal insulin/22.5].

Only participants from the PD analysis set, who had evaluable data at both baseline and the given post-baseline time point, were included in the analysis for that post baseline time point. The PD analysis set included randomized participants who received at least one dose of study medication.

Blood samples were collected to analyze the ACZ885 serum concentrations.

Only participants from the PK analysis set, who had evaluable data at each time point, were included in the analysis for that time point. The PK analysis set included randomized participants from the ACZ885 arm who received at least one dose of study medication.