Change from baseline in HbA1c (%) to Week 24 during the controlled assessment period is reported as adjusted least square (LS) mean values. Baseline was defined as the last non-missing assessment (scheduled or unscheduled) on or prior to the first dose of randomized study medication. A mixed model with repeated measures (MMRM) analysis was performed, excluding data collected after initiation of rescue medication or premature discontinuation of study medication.

The Evaluable Analysis Set consisted of all randomized patients who received at least 1 dose of randomized study medication and had at least 1 baseline and post-baseline HbA1c assessment.

A controlled assessment period AE was defined as an AE starting on or after day of first dose of study medication up to but not including Week 24 for patients entering the extension period. For patients not entering the extension period, the period was defined up to and including last dose of study medication + 7 days (+ 90 days for serious AEs [SAEs] and other clinically significant or related AEs). The Investigator assessed AEs for causal relationship to study drug medication.

The Safety Analysis Set consisted of all patients who received at least 1 dose of study medication. One patient who was randomized to placebo received a dose of exenatide in error and was subsequently reassigned to the exenatide treatment group for analyses based on actual treatment.

Percentage of patients positive for ADAs up to Week 24 for the exenatide treatment group is reported. Baseline was the antibody measurement at Week 0 (Day 1). A negative or missing antibody measurement was considered negative at baseline. High positive = antibody titers ≥ 625, including baseline assessment. Low positive = antibody titers < 625, including baseline assessment. A patient was said to have treatment-emergent ADA positive at a visit if the antibody test was positive after the first dose of exenatide following a negative or missing antibody measurement, or the titer increased by at least 1 titration category from a detectable measurement prior to first dose of randomized study medication.

The Safety Analysis Set consisted of all patients who received at least 1 dose of study medication. Only patients receiving exenatide in the controlled assessment period were included in the analysis.

Change from baseline in FPG to Week 24 during the controlled assessment period is reported as adjusted LS mean values. Baseline was defined as the last non-missing assessment (scheduled or unscheduled) on or prior to the first dose of randomized study medication. A MMRM analysis was performed, excluding data collected after initiation of rescue medication or after premature discontinuation of study medication.

The ITT Analysis Set consisted of all randomized patients who received at least 1 dose of randomized study medication.

Change from baseline in body weight to Week 24 during the controlled assessment period is reported as adjusted LS mean values. Baseline was defined as the last non-missing assessment (scheduled or unscheduled) on or prior to the first dose of randomized study medication. A MMRM analysis was performed, excluding data collected after initiation of rescue medication or after premature discontinuation of study medication.

The ITT Analysis Set consisted of all randomized patients who received at least 1 dose of randomized study medication.

Change from baseline in fasting insulin to Week 24 during the controlled assessment period is reported as adjusted LS mean values. Baseline was defined as the last non-missing assessment (scheduled or unscheduled) on or prior to the first dose of randomized study medication. A MMRM analysis was performed, excluding data collected after initiation of rescue medication or after premature discontinuation of study medication.

The ITT Analysis Set consisted of all randomized patients who received at least 1 dose of randomized study medication.

The percentage of patients achieving HbA1c goals of < 6.5%, ≤ 6.5%, and < 7.0% at Week 24 during the controlled assessment period is reported. A Cochran-Mantel-Haenszel (CMH) analysis was performed with missing data treated as non-responder, and excluding data collected after initiation of rescue medication or after premature discontinuation of study medication.

The Evaluable Analysis Set consisted of all randomized patients who received at least 1 dose of randomized study medication and had at least 1 baseline and post-baseline HbA1c assessment. Only patients with data available were included in the analysis.

Change from baseline in lipid profiles to Week 24 during the controlled assessment period is reported as mean values (Standard International [SI] units). The following lipids were assessed: total cholesterol, high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), and triglycerides. All lipids presented were taken in a fasted state. Baseline was defined as the last non-missing assessment (scheduled or unscheduled) on or prior to the first dose of randomized study medication. Data collected after initiation of rescue medication or after premature discontinuation of study medication were excluded.

The ITT Analysis Set consisted of all randomized patients who received at least 1 dose of randomized study medication. Only patients with data available were included in the analysis.

Change from baseline in SBP and DBP to Week 24 during the controlled assessment period is reported as adjusted LS mean values. Baseline was defined as the last non-missing assessment (scheduled or unscheduled) on or prior to the first dose of randomized study medication. A MMRM analysis was performed, excluding data collected after initiation of rescue medication or after premature discontinuation of study medication.

The ITT Analysis Set consisted of all randomized patients who received at least 1 dose of randomized study medication.

Number of patients needing rescue medication at Week 24 and at each intermediate visit during the controlled assessment period is reported. Patients with a loss of glycemic control, defined as either an increase from baseline in HbA1c values by ≥ 1.0% at 2 consecutive clinic visits that were at least 1 month apart, or a fasting plasma glucose value ≥ 250 mg/dL or random blood glucose value > 300 mg/dL for 4 days during a 7 day period, received rescue medication. Data collected after premature discontinuation of study medication were excluded.

The ITT Analysis Set consisted of all randomized patients who received at least 1 dose of randomized study medication. Only patients with data available at each specified visit were included in the analysis.

Change from baseline in HOMA-B and HOMA-S in patients who were not taking insulin to Week 24 during the controlled assessment period is reported as adjusted LS mean values. Baseline was defined as the last non-missing assessment (scheduled or unscheduled) on or prior to the first dose of randomized study medication. A MMRM analysis was performed, excluding data collected after initiation of rescue medication or after premature discontinuation of study medication.

The ITT Analysis Set consisted of all randomized patients who received at least 1 dose of randomized study medication. Homeostasis model assessments were only performed in patients who were not taking insulin.

Percentage of patients reporting injection site reactions at Week 24 and at each intermediate visit during the controlled assessment period is reported. Injection site reactions were presented from the AE case report form (CRF), based on the "Injection site reactions" higher level term. A controlled assessment period AE was defined as an AE starting on or after day of first dose of study medication up to but not including Week 24 for patients entering the extension period. For patients not entering the extension period, the period was defined up to and including last dose of study medication + 7 days (+ 90 days for SAEs and other clinically significant or related AEs).

The Safety Analysis Set consisted of all patients who received at least 1 dose of study medication. Only patients with data available at each specified visit were included in the analysis.

Change from baseline in HbA1c (%) to Week 52 among patients who received open-label exenatide during the treatment period is reported as mean values. The treatment period was defined as the controlled assessment period and extension period combined. Baseline was defined as the last non-missing assessment (scheduled or unscheduled) on or prior to the first dose of randomized study medication. Data collected after initiation of rescue medication or after premature discontinuation of study medication were excluded.

The Evaluable Analysis Set consisted of all randomized patients who received at least 1 dose of randomized study medication and had at least 1 baseline and post-baseline HbA1c assessment. Only patients with observed baseline and Week 52 values, and who received open-label exenatide were included in the analysis.

Change from baseline in FPG to Week 52 among patients who received open-label exenatide during the treatment period is reported as mean values. The treatment period was defined as the controlled assessment period and extension period combined. Baseline was defined as the last non-missing assessment (scheduled or unscheduled) on or prior to the first dose of randomized study medication. Data collected after initiation of rescue medication or after premature discontinuation of study medication were excluded.

The ITT Analysis Set consisted of all randomized patients who received at least 1 dose of randomized study medication. Only patients with observed baseline and Week 52 values, and who received open-label exenatide were included in the analysis.

Change from baseline in body weight to Week 52 among patients who received open-label exenatide during the treatment period is reported as mean values. The treatment period was defined as the controlled assessment period and extension period combined. Baseline was defined as the last non-missing assessment (scheduled or unscheduled) on or prior to the first dose of randomized study medication. Data collected after initiation of rescue medication or after premature discontinuation of study medication were excluded.

The ITT Analysis Set consisted of all randomized patients who received at least 1 dose of randomized study medication. Only patients with observed baseline and Week 52 values, and who received open-label exenatide were included in the analysis.

Change from baseline in fasting insulin to Week 52 among patients who received open-label exenatide during the treatment period is reported as mean values. The treatment period was defined as the controlled assessment period and extension period combined. Baseline was defined as the last non-missing assessment (scheduled or unscheduled) on or prior to the first dose of randomized study medication. Data collected after initiation of rescue medication or after premature discontinuation of study medication were excluded.

The ITT Analysis Set consisted of all randomized patients who received at least 1 dose of randomized study medication. Only patients with observed baseline and Week 52 values, and who received open-label exenatide were included in the analysis.

The percentage of patients achieving HbA1c goals of < 6.5%, ≤ 6.5%, and < 7.0% at Week 52 among patients who received open-label exenatide during the treatment period is reported. The treatment period was defined as the controlled assessment period and extension period combined. Data collected after initiation of rescue medication or after premature discontinuation of study medication were excluded.

The Evaluable Analysis Set consisted of all randomized patients who received at least 1 dose of randomized study medication and had at least 1 baseline and post-baseline HbA1c assessment. Only patients who received open-label exenatide and with data available were included in the analysis.

Change from baseline in lipid profiles to Week 52 among patients who received open-label exenatide during the treatment period is reported as mean values (SI units). The treatment period was defined as the controlled assessment period and extension period combined. The following lipids were assessed: total cholesterol, HDL-C, LDL-C, and triglycerides. All lipids presented were taken in a fasted state. Baseline was defined as the last non-missing assessment (scheduled or unscheduled) on or prior to the first dose of randomized study medication. Data collected after initiation of rescue medication or after premature discontinuation of study medication were excluded.

The ITT Analysis Set consisted of all randomized patients who received at least 1 dose of randomized study medication. Only patients with observed baseline and Week 52 values, and who received open-label exenatide were included in the analysis.

Change from baseline in SBP and DBP to Week 52 among patients who received open-label exenatide during the treatment period is reported as mean values. The treatment period was defined as the controlled assessment period and extension period combined. Baseline was defined as the last non-missing assessment (scheduled or unscheduled) on or prior to the first dose of randomized study medication. Data collected after initiation of rescue medication or after premature discontinuation of study medication were excluded.

The ITT Analysis Set consisted of all randomized patients who received at least 1 dose of randomized study medication. Only patients with observed baseline and Week 52 values, and who received open-label exenatide were included in the analysis.

Number of patients needing rescue medication at Week 52 and at each intermediate visit during the treatment period is reported. The treatment period was defined as the controlled assessment period and extension period combined. Patients with a loss of glycemic control, defined as either an increase from baseline in HbA1c values by ≥ 1.0% at 2 consecutive clinic visits that were at least 1 month apart, or a fasting plasma glucose value ≥ 250 mg/dL or random blood glucose value > 300 mg/dL for 4 days during a 7 day period, received rescue medication. Data collected after premature discontinuation of study medication were excluded.

The ITT Analysis Set consisted of all randomized patients who received at least 1 dose of randomized study medication. Only patients who received open-label exenatide and with data available were included in the analysis.

Change from baseline in HOMA-B and HOMA-S to Week 52 among patients who received open-label exenatide during the treatment period is reported as mean values. The treatment period was defined as the controlled assessment period and extension period combined. Baseline was defined as the last non-missing assessment (scheduled or unscheduled) on or prior to the first dose of randomized study medication. Data collected after initiation of rescue medication or after premature discontinuation of study medication were excluded.

The ITT Analysis Set consisted of all randomized patients who received at least 1 dose of randomized study medication. Only patients with observed baseline and Week 52 values, and who received open-label exenatide were included in the analysis.

Percentage of patients reporting injection site reactions at Week 52 and at each intermediate visit among patients who received open-label exenatide during the treatment period is reported. The treatment period was defined as the controlled assessment period and extension period combined. Injection site reactions were presented from the AE CRF, based on the "Injection site reactions" higher level term. An Extension Period AE was defined as an AE starting on or after day of first dose of open-label exenatide to last dose + 7 days (+ 90 days for SAEs and other clinically significant or related AEs).

The Safety Analysis Set consisted of all patients who received at least 1 dose of study medication. Only patients with data available at each specified visit were included in the analysis.

Geometric mean plasma exenatide concentrations up to Week 52 during the treatment period are reported (for the placebo then exenatide treatment group, only Weeks 24 and 52 were applicable). The treatment period was defined as the controlled assessment period and extension period combined. Data collected after initiation of rescue medication were included. Data collected after discontinuation of study medication were excluded.

The Pharmacokinetic (PK) Analysis Set consisted of all patients who received at least 1 dose of exenatide, for whom any postdose data were available and who did not deviate from the protocol in ways that would significantly affect the PK analyses. Only patients who received open-label exenatide and with data available were included in the analysis.