Plasma fasting TG and HDL concentrations were measured to determine the TG:HDL ratio.

The primary end point analysis set consisted of 413 patients, i.e., all patients who received study medication at least once and had baseline and post-baseline TG and HDL data. Data of patients with missing TG or HDL values or who started or changed lipid-lowering medication during the trial were excluded from analysis.

The TG level was assessed under fasted conditions.

The secondary end point analysis set included all patients who received study medication at least once and provided ≥1 post-baseline efficacy measurement. Therefore, the total number of patients in the analysis set depends on the number of patients having post-baseline data for the particular secondary efficacy parameter under discussion.

The HDL level was assessed under fasted conditions.

The secondary end point analysis set included all patients who received study medication at least once and provided ≥1 post-baseline efficacy measurement. Therefore, the total number of patients in the analysis set depends on the number of patients having post-baseline data for the particular secondary efficacy parameter under discussion.

The total cholesterol level was assessed under fasted conditions.

Safety data were analyzed using the Intent-to-Treat (ITT) analysis set for safety and consisted of 459 patients, i.e., all patients who received study medication at least once and provided any post-baseline safety data. Changes from baseline could only be calculated for patients with paired data.

The level of low density lipoprotein cholesterol was calculated using the Friedwald QT formula.

Safety data were analyzed using the ITT analysis set for safety and consisted of 459 patients, i.e., all patients who received study medication at least once and provided any post-baseline safety data. Changes from baseline could only be calculated for patients with paired data.

The insulin level was assessed under fasted conditions.

Safety data were analyzed using the ITT analysis set for safety and consisted of 459 patients, i.e., all patients who received study medication at least once and provided any post-baseline safety data. Changes from baseline could only be calculated for patients with paired data.

Safety data were analyzed using the ITT analysis set for safety and consisted of 459 patients, i.e., all patients who received study medication at least once and provided any post-baseline safety data. Changes from baseline could only be calculated for patients with paired data.

HOMA-%B is used to assess beta-cell function. HOMA-%B is a dimensionless measure of beta-cell function (higher values present increased insulin secretion for a given glucose level).

HOMA-%B is normalized so that lean, healthy individuals will have values of HOMA-%B close to 100%.

The secondary end point analysis set included all patients who received study medication at least once and provided ≥1 post-baseline efficacy measurement. Therefore, the total number of patients in the analysis set depends on the number of patients having post baseline data for the particular secondary efficacy parameter under discussion.

HOMA-IR is used to assess insulin resistance (IR). HOMA-IR is a dimensionless measure of insulin resistance (higher values present more insulin resistance. HOMA-IR are normalized so that lean, healthy individuals will have values of HOMA-IR close to 1.

The secondary end point analysis set included all patients who received study medication at least once and provided ≥1 post-baseline efficacy measurement. Therefore, the total number of patients in the analysis set depends on the number of patients having post-baseline data for the particular secondary efficacy parameter under discussion.

Fasting plasma glucose ≥126 mg/dL or 2-hour post-load plasma glucose ≥200 mg/dL during an oral glucose tolerance test (OGTT) or initiated use of glucose-lowering agents during the course of the study.

The secondary end point analysis set included all patients who received study medication at least once and provided ≥1 post-baseline efficacy measurement.Therefore, the total number of patients in the analysis set depends on the number of patients having post-baseline data for the particular secondary efficacy parameter under discussion.

Glucose ≥140 mg/dL, <200 mg/dL after a 75g OGTT.

The secondary end point analysis set included all patients who received study medication at least once and provided ≥1 post-baseline efficacy measurement.Therefore, the total number of patients in the analysis set depends on the number of patients having post-baseline data for the particular secondary efficacy parameter under discussion.

Post-baseline glucose level under fasted conditions ≥100 mg/dL but <126 mg/dL.

The secondary end point analysis set included all patients who received study medication at least once and provided ≥1 post-baseline efficacy measurement. Therefore, the total number of patients in the analysis set depends on the number of patients having post-baseline data for the particular secondary efficacy parameter under discussion.

The insulinogenic index, defined as (insulin at 30 min - insulin at 0)/(glucose at 30 min [G(30)] - glucose at 0 [G(0)]) was used as a measure of early insulin secretion in response to the OGTT. Because the index is undefined when G(30)-G(0)=0, and poorly defined when G(30)-G(0)<0, the index was only calculated when G(30)>G(0).

The secondary end point analysis set included all patients who received study medication at least once and provided ≥1 post-baseline efficacy measurement. Therefore, the total number of patients in the analysis set depends on the number of patients having post-baseline data for the particular secondary efficacy parameter under discussion.

As another measure of beta-cell function, the relationship between plasma insulin and glucose concentrations during the OGTT was calculated using a simplified version of the method described by Mari et al. (Mari A, Sallas WM, He YL, Watson C, Ligueros-Saylan M, Dunning BE, Deacon CF, Holst JJ, Foley JE. Vildagliptin, a dipeptidyl peptidase-IV inhibitor, improves model-assessed beta-cell function in patients with type 2 diabetes. J Clin Endocrinol Metab. 2005; 90:4888-4894.).

The secondary end point analysis set included all patients who received study medication at least once and provided ≥1 post-baseline efficacy measurement. Therefore, the total number of patients in the analysis set depends on the number of patients having post-baseline data for the particular secondary efficacy parameter under discussion.

Patients were weighed lightly clothed. The same amount of clothing had to be worn each time.

Safety data were analyzed using the ITT analysis set for safety and consisted of 459 patients, i.e., all patients who received study medication at least once and provided any post-baseline safety data. Changes from baseline could only be calculated for patients with paired data.

BMI is calculated by dividing the body weight (in kg) by the square of height (in meters).

Patients had to be instructed to stand erect with abdomen relaxed, arms at sides, feet together, and weight divided equally over both legs. The tape measure was placed around the bare abdomen midway between the palpated iliac crest and the palpated lowest rib margin in the left and right mid-axillary lines. A nonstretchable tape was evenly placed around the natural waist covering the left and right natural-waist marks. The measurement scale had to face outward, and there could not be any twists in the tape. The tape had to be just touching the skin but not compressing the soft tissue.

Safety data were analyzed using the ITT analysis set for safety and consisted of 459 patients, i.e., all patients who received study medication at least once and provided any post-baseline safety data. Changes from baseline could only be calculated for patients with paired data.

Metabolic syndrome is defined according the Third Report of the National Cholesterol Education Program Expert Panel on

Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (NCEP/ATPIII) of which 3 out of 5 criteria must be met:

• waist circumference men > 102 cm; waist circumference women > 88 cm
• TG ≥ 150 mg/dL
• HDL cholesterol men <40 mg/dL; HDL cholesterol women <50 mg/dL
• Blood pressure systolic ≥ 130 mmHg; Blood pressure diastolic ≥ 85 mmHg
• Fasting glucose ≥ 110 mg /dL

The secondary endpoint analysis set included all patients who received study medication at least once and provided ≥1 post-baseline efficacy measurement. In this case, the total number of patients in the analysis set depends on the number of patients without metabolic syndrome at baseline.

PANSS is an investigator-rated 30-item scale to assess the neuropsychiatric symptoms of schizophrenia. The PANSS provided a total score and scores for 3 subscales, the positive subscale (7 items), the negative subscale (7 items), and the general psychopathology subscale (16 items), each rated on a scale of 1 (absent) to 7 (extreme).

The secondary endpoint analysis set included all patients who received study medication at least once and provided ≥1 post-baseline efficacy measurement. Therefore, the total number of patients in the analysis set depends on the number of patients having post-baseline data for the particular secondary efficacy parameter under discussion.