Least squares (LS) means were calculated using analysis of covariance (ANCOVA) and last observation carried forward (LOCF) imputation with country and treatment as fixed effects and baseline HbA1c as a covariate.

All randomized participants in the selected (1.5 mg, 0.75 mg LY2189265) and active comparator (Sitagliptin) arms who had evaluable HbA1c data. Last observation carried forward was used to impute missing postbaseline values. If there were no data after the date of randomization, the endpoint was considered missing.

Change from baseline in HbA1c was 1 of the 4 measures included in the clinical utility index (CUI) used to evaluate the dose decision. The maximum duration of exposure to LY2189265, Sitagliptin, or Placebo (across all treatment arms) at the decision point was 27.4 weeks.

All participants randomized before the dose decision point who had evaluable HbA1c data.

Least squares (LS) means were calculated using analysis of covariance (ANCOVA) and last observation carried forward (LOCF) imputation with country and treatment as fixed effects and baseline HbA1c as a covariate.

All randomized participants in the selected (1.5 mg, 0.75 mg LY2189265) and comparator (Sitagliptin, Placebo/Sitagliptin) arms who had evaluable HbA1c data. Last observation carried forward was used to impute missing postbaseline values. If there were no data after the date of randomization, the endpoint was considered missing.

Durability of effect on HbA1c was assessed by comparing the differences in mean change from baseline in HbA1c at 1 time point versus an earlier time point. Least squares (LS) means of change from baseline HbA1c data were calculated using a mixed-effects model for repeated measures (MMRM) analysis with treatment, country, visit, and treatment-by-visit interaction as fixed effects and baseline as a covariate.

All participants in the selected (1.5 mg, 0.75 mg LY2189265) and comparator (Sitagliptin, Placebo/Sitagliptin) arms randomized after the dose decision point who had evaluable HbA1c data. If there were no data after the date of randomization, the endpoint was considered missing.

Least squares (LS) means of change from baseline were calculated using mixed-effects model for repeated measures (MMRM) with treatment, country, visit, and treatment-by-visit interaction as fixed effects and baseline as a covariate.

All randomized participants in the selected (1.5 mg, 0.75 mg LY2189265) and comparator (Sitagliptin, Placebo/Sitagliptin) arms who had evaluable fasting plasma glucose data. If there were no data after the date of randomization, the endpoint was considered missing.

Least squares (LS) means of change from baseline fasting insulin data were calculated using a mixed-effects model for repeated measures (MMRM) analysis with treatment, country, visit, and treatment-by-visit interaction as fixed effects and baseline as a covariate.

All participants in the selected (1.5 mg, 0.75 mg LY2189265) and comparator (Sitagliptin, Placebo/Sitagliptin) arms randomized after the dose decision point who had evaluable fasting insulin data. If there were no data after the date of randomization, the endpoint was considered missing.

Change from baseline in body weight was 1 of the 4 measures included in the clinical utility index (CUI) used to evaluate the dose decision. The maximum duration of exposure to LY2189265, Sitagliptin, or Placebo (across all treatment arms) at the decision point was 27.4 weeks.

All participants randomized before the dose decision point who had evaluable body weight data.

Least squares (LS) means of change from baseline body weight were calculated using analysis of covariance (ANCOVA) and last observation carried forward (LOCF) imputation with country and treatment as fixed effects and baseline as a covariate.

All randomized participants in the selected (1.5 mg, 0.75 mg LY2189265) and comparator (Sitagliptin, Placebo/Sitagliptin) arms who had evaluable body weight data. Last observation carried forward was used to impute missing postbaseline values. If there were no data after the date of randomization, the endpoint was considered missing.

Durability of effect on body weight was assessed by comparing the differences in mean change from baseline in body weight at 1 time point versus an earlier time point. Least squares (LS) means of change from baseline body weight data were calculated using a mixed-effects model for repeated measures (MMRM) analysis with treatment, country, visit, and treatment-by-visit interaction as fixed effects and baseline as a covariate.

All participants in the selected (1.5 mg, 0.75 mg LY2189265) and comparator (Sitagliptin, Placebo/Sitagliptin) arms randomized after the dose decision point who had evaluable body weight data. If there were no data after the date of randomization, the endpoint was considered missing.

Least squares (LS) means of change from baseline were calculated using a mixed-effects model for repeated measures (MMRM) with treatment, country, visit, and treatment-by-visit interaction as fixed effects and baseline as a covariate.

All randomized participants in the selected (1.5 mg, 0.75 mg LY2189265) and comparator (Sitagliptin, Placebo/Sitagliptin) arms who had evaluable waist circumference data. If there were no data after the date of randomization, the endpoint was considered missing.

The percentage of participants achieving HbA1c levels <7.0% and ≤6.5% was analyzed using a logistic regression model and last observation carried forward (LOCF) imputation with baseline, country, and treatment as factors included in the model.

All randomized participants in the selected (1.5 mg, 0.75 mg LY2189265) and comparator (Sitagliptin, Placebo/Sitagliptin) arms who had evaluable HbA1c data. Last observation carried forward was used to impute missing postbaseline values. If there were no data after the date of randomization, the endpoint was considered missing.

Hypoglycemic episodes (HE) were classified as severe (defined as episodes requiring assistance from another person to actively administer resuscitative actions), documented symptomatic (defined as any time a participant feels that he/she is experiencing symptoms and/or signs associated with hypoglycemia and has a plasma glucose level of ≤3.9 millimoles per liter [mmol/L]), asymptomatic (defined as episodes not accompanied by typical symptoms of hypoglycemia but with a measured plasma glucose of ≤3.9 mmol/L), nocturnal (defined as any episode that occurred between bedtime and waking), or probable symptomatic (defined as episodes during which symptoms of hypoglycemia were not accompanied by a plasma glucose determination). The number of participants with self-reported hypoglycemic events is summarized cumulatively.

All randomized participants in the selected (1.5 mg, 0.75 mg LY2189265) and comparator (Sitagliptin, Placebo/Sitagliptin) arms.

Hypoglycemic episodes (HE) were classified as severe (defined as episodes requiring assistance from another person to actively administer resuscitative actions), documented symptomatic (defined as any time a participant feels that he/she is experiencing symptoms and/or signs associated with hypoglycemia and has a plasma glucose level of ≤3.9 millimoles per liter [mmol/L]), asymptomatic (defined as episodes not accompanied by typical symptoms of hypoglycemia but with a measured plasma glucose of ≤3.9 mmol/L), nocturnal (defined as any episode that occurred between bedtime and waking), or probable symptomatic (defined as episodes during which symptoms of hypoglycemia were not accompanied by a plasma glucose determination). The 1-year adjusted rate of HE is summarized cumulatively.

All randomized participants in the selected (1.5 mg, 0.75 mg LY2189265) and comparator (Sitagliptin, Placebo/Sitagliptin) arms.

The homeostatic model assessment (HOMA) is a method used to quantify insulin resistance and beta (β)-cell function. HOMA2-%B is a computer model that uses fasting plasma insulin and glucose concentrations to estimate steady state beta cell function (%B) as a percentage of a normal reference population (normal young adults). HOMA2-%S is a computer model that uses fasting plasma insulin and glucose concentrations to estimate insulin sensitivity (%S), as percentages of a normal reference population (normal young adults). The normal reference population for both HOMA2-%B and HOMA2-%S were set at 100%. Least squares (LS) means of change from baseline of C-peptide based HOMA2-%B and HOMA2-%S were calculated using a mixed-effects model for repeated measures (MMRM) with treatment, country, visit, and treatment-by-visit interaction as fixed effects and baseline as a covariate.

All participants in the selected (1.5 mg, 0.75 mg LY2189265) and comparator (Sitagliptin, Placebo/Sitagliptin) arms randomized after the dose decision point and who had evaluable HOMA2 data. If there were no data after the date of randomization, the endpoint was considered missing.

A treatment-emergent adverse event (TEAE) was defined as an event that first occurs or worsens (increases in severity) after baseline regardless of causality or severity. The number of participants with 1 or more TEAEs is summarized cumulatively. A summary of serious and other non-serious adverse events regardless of causality is located in the Reported Adverse Events module.

All randomized participants in the selected (1.5 mg, 0.75 mg LY2189265) and comparator (Sitagliptin, Placebo/Sitagliptin) arms.

A treatment-emergent adverse event (TEAE) was defined as an event that first occurs or worsens (increases in severity) after baseline regardless of causality or severity. The number of participants with 1 or more TEAEs is summarized cumulatively. A summary of serious and other non-serious adverse events regardless of causality is located in the Reported Adverse Events module.

All randomized participants in the LY2189265 and active comparator (Sitagliptin) arms.

A treatment-emergent adverse event (TEAE) was defined as an event that first occurs or worsens (increases in severity) after baseline regardless of causality or severity. The number of participants with 1 or more TEAEs is summarized cumulatively. A summary of serious and other non-serious adverse events regardless of causality is located in the Reported Adverse Events module.

All randomized participants in the selected (1.5 mg, 0.75 mg LY2189265) and active comparator (Sitagliptin) arms.

The number of participants with treatment-emergent abnormal laboratory results (defined as abnormalities that first occur after baseline) was summarized cumulatively for alkaline phosphatase, alanine aminotransferase or serum glutamic pyruvic transaminase (ALT/SGPT), amylase (pancreatic and total), aspartate aminotransferase or serum glutamic oxaloacetic transaminase (AST/SGOT), basophils, bilirubin (direct and total), calcitonin, chloride, creatine phosphokinase (CPK), creatinine, creatinine clearance, eosinophils, erythrocytes, gamma glutamyltransferase (GGT), hematocrit, hemoglobin, leukocytes, lipase, lymphocytes, mean cell hemoglobin concentration (MCHC), mean cell volume (MCV), monocytes, neutrophils, platelets, potassium, sodium, urea nitrogen, and urine microalbumin-to-creatinine ratio (UMCR).

All randomized participants in the selected (1.5 mg, 0.75 mg LY2189265) and comparator (Sitagliptin, Placebo/Sitagliptin) arms with baseline value not in the specified direction of treatment-emergent abnormality and at least 1 postbaseline result.

The number of participants with treatment-emergent abnormal laboratory results (defined as abnormalities that first occur after baseline) was summarized cumulatively for alkaline phosphatase, alanine aminotransferase or serum glutamic pyruvic transaminase (ALT/SGPT), amylase (pancreatic and total), aspartate aminotransferase or serum glutamic oxaloacetic transaminase (AST/SGOT), basophils, bilirubin (direct and total), calcitonin, chloride, creatine phosphokinase (CPK), creatinine, creatinine clearance, eosinophils, erythrocytes, gamma glutamyltransferase (GGT), hematocrit, hemoglobin, leukocytes, lipase, lymphocytes, mean cell hemoglobin concentration (MCHC), mean cell volume (MCV), monocytes, neutrophils, platelets, potassium, sodium, urea nitrogen, and urine microalbumin-to-creatinine ratio (UMCR) .

All randomized participants in the selected (1.5 mg, 0.75 mg LY2189265) and active comparator (Sitagliptin) arms with baseline value not in the specified direction of treatment-emergent abnormality and at least 1 postbaseline result.

The number of participants with treatment-emergent abnormal laboratory results (defined as abnormalities that first occur after baseline) was summarized cumulatively for alkaline phosphatase, alanine aminotransferase or serum glutamic pyruvic transaminase (ALT/SGPT), amylase (pancreatic and total), aspartate aminotransferase or serum glutamic oxaloacetic transaminase (AST/SGOT), basophils, bilirubin (direct and total), calcitonin, chloride, creatine phosphokinase (CPK), creatinine, creatinine clearance, eosinophils, erythrocytes, gamma glutamyltransferase (GGT), hematocrit, hemoglobin, leukocytes, lipase, lymphocytes, mean cell hemoglobin concentration (MCHC), mean cell volume (MCV), monocytes, neutrophils, platelets, potassium, sodium, urea nitrogen, and urine microalbumin-to-creatinine ratio (UMCR).

All randomized participants in the selected (1.5 mg, 0.75 mg LY2189265) and active comparator (Sitagliptin) arms with baseline value not in the specified direction of treatment-emergent abnormality and at least 1 postbaseline result.

The number of participants with treatment-emergent abnormal lipid test (cholesterol, high density lipoprotein cholesterol [HDL-C], low density lipoprotein cholesterol [LDL-C], and triglycerides [TG]) results (defined as lipid test abnormalities that first occurred after baseline) is summarized cumulatively.

All randomized participants in the selected (1.5 mg, 0.75 mg LY2189265) and comparator (Sitagliptin, Placebo/Sitagliptin) arms with baseline value not in the specified direction of treatment-emergent abnormality and at least 1 postbaseline result.

Sitting pulse rate was measured at the time that the dose decision was made (dose decision point). Change from baseline in pulse rate was 1 of the 4 measures included in the clinical utility index (CUI) used to evaluate the dose decision. The maximum duration of exposure to LY2189265, Sitagliptin, or Placebo (across all treatment arms) at the decision point was 27.4 weeks.

All participants randomized before the dose decision point who had evaluable sitting pulse rate data.

Sitting systolic blood pressure (SBP) and diastolic blood pressure (DBP) were measured at the dose decision point. Change from baseline in DBP was 1 of the 4 measures included in the clinical utility index (CUI) used to evaluate the dose decision. The maximum duration of exposure to LY2189265, Sitagliptin, or Placebo (across all treatment arms) at the time of the decision point was 27.4 weeks.

All participants randomized before the dose decision point who had evaluable sitting SBP and DBP data.

Sitting and standing pulse rate were measured. Least squares (LS) means of change from baseline were calculated using a mixed-effects model for repeated measures (MMRM) with treatment, country, visit, and treatment-by-visit interaction as fixed effects and baseline as covariate.

All randomized participants in the selected (1.5 mg, 0.75 mg LY2189265) and comparator (Sitagliptin, Placebo/Sitagliptin) arms who had evaluable pulse data. If there were no data after the date of randomization, the endpoint was considered missing.

Sitting and standing systolic blood pressure (SBP) and diastolic blood pressure (DBP) were measured. Least squares (LS) means of change from baseline were calculated using a mixed-effects model for repeated measures (MMRM) with treatment, country, visit, and treatment-by-visit interaction as fixed effects and baseline as a covariate.

All randomized participants in the selected (1.5 mg, 0.75 mg LY2189265) and comparator (Sitagliptin, Placebo/Sitagliptin) arms who had evaluable SBP and DBP data. If there were no data after the date of randomization, the endpoint was considered missing.

The QT interval is a measure of the time between the start of the Q wave and the end of the T wave and was calculated from electrocardiogram (ECG) data using Fridericia's formula: QTc = QT/RR^0.33. Corrected QT (QTc) is the QT interval corrected for heart rate and RR, which is the interval between two R waves. PR is the interval between the P wave and the QRS complex. Least Squares (LS) means of change from baseline were calculated using a mixed-effects model for repeated measures (MMRM) with treatment, country, visit, and treatment-by-visit interaction as fixed effects and baseline as a covariate.

All randomized participants in the selected (1.5 mg, 0.75 mg LY2189265) and comparator (Sitagliptin, Placebo/Sitagliptin) arms who had evaluable ECG data. If there were no data after the date of randomization, the endpoint was considered missing.

The Impact of Weight on Quality of Life-Lite (IWQoL-Lite questionnaire) is an obesity-specific, 31-item questionnaire designed to measure the impact of weight on participants' quality of life. Items are scored on a 5-point numeric rating scale where 5 = "always true" and 1 = "never true". Items are summed into 6 scales (physical function [11 items], self-esteem [7 items], sexual life [4 items], public distress [5 items], work [4 items], and total score [31 items]) based on the average for the valid responses on that scale multiplied by the number of items on that scale (rounded to the nearest whole integer). Higher scores indicate lower levels of functioning (negative effects). Scores are linearly transformed to a 0 to 100 scale.

All randomized participants in the selected (1.5 mg, 0.75 mg LY2189265) and active comparator (Sitagliptin) arms who had evaluable IWQoL-Lite data. If there were no data after the date of randomization, the endpoint was considered missing.

The EQ-5D questionnaire is a generic, multidimensional, health-related, quality-of-life instrument. It consists of 2 parts. The first part allows participants to rate their health state in 5 health domains: mobility, self-care, usual activities, pain/discomfort, and mood using a three level scale of 1-3 (no problem, some problems, and major problems). These combinations of attributes were converted into a weighted health-state Index Score according to the United Kingdom (UK) population-based algorithm. The possible values for the Index Score ranged from -0.59 (severe problems in all 5 dimensions) to 1.0 (no problem in any dimension). The second part of the questionnaire consists of a 100-millimeter visual analog scale (VAS) on which the participants rated their perceived health state on that day from 0 (worst imaginable health state) to 100 (best imaginable health state).

All randomized participants in the selected (1.5 mg, 0.75 mg LY2189265) and active comparator (Sitagliptin) arms who had evaluable EQ-5D data. If there were no data after the date of randomization, the endpoint was considered missing.

The number of visits to the emergency room (ER) is summarized cumulatively.

All randomized participants in the selected (1.5 mg, 0.75 mg LY2189265) and active comparator (Sitagliptin) arms.

Pharmacokinetic (PK) parameter estimates from LY2189265 concentration data were obtained using a 2-compartment population PK model with first order absorption. Area under the plasma-concentration curve from 0 to 168 hours, steady state (AUC0-168h, ss) of LY2189265 is summarized.

Randomized participants in the selected (1.5 mg, 0.75 mg LY2189265) who had blood samples collected for PK assessments.

The number of participants with postbaseline detection of treatment-emergent antidrug LY2189265 antibodies (ADA) is summarized.

All randomized participants in the LY2189265 arms who had evaluable ADA data. If there were no data after the date of randomization, the endpoint was considered missing.

The number of participants with pancreatitis confirmed by adjudication is summarized cumulatively. A summary of serious and other non-serious adverse events regardless of causality is located in the Reported Adverse Events module.

All randomized participants.

Data on any new cardiovascular (CV) event was prospectively collected using a CV event electronic case report form. At prespecified visits, participants were asked about any new CV event. Deaths and nonfatal cardiovascular adverse events (AEs) were adjudicated by a committee of physicians with cardiology expertise external to the Sponsor. The nonfatal cardiovascular AEs to be adjudicated include myocardial infarction, hospitalization for unstable angina, hospitalization for heart failure, coronary interventions (such as coronary artery bypass graft or percutaneous coronary intervention), and cerebrovascular events including cerebrovascular accident (stroke) and transient ischemic attack. The number of participants with adjudicated CV events is summarized cumulatively. A summary of serious and other non-serious adverse events regardless of causality is located in the Reported Adverse Events module.

All randomized participants.