Change from baseline in HbA1c is reported as adjusted least square (LS) mean values at Week 28. Baseline is defined as the last non-missing assessment (scheduled or unscheduled) on or prior to the first dose of study medication. A mixed model with repeated measures (MMRM) analysis was performed, excluding measurements after initiation of rescue medication and study drug discontinuation.

The Evaluable Analysis Set included all randomized participants who received at least 1 dose of randomized study medication and had a baseline and at least 1 post-baseline HbA1c assessment. For analysis of efficacy, a decision was made with agreement by the European Medicines Agency and the Food and Drug Administration to pool participants from both exenatide groups (Total EBID) and compare this pooled group with placebo.

Post-treatment adverse events (AEs) were defined as AEs that started or worsened during the off-treatment period (Safety Follow-up Period), which was defined as the day after the Week 28/early discontinuation (ED) visit to the date of completion of the Safety Follow-up Period. The AESIs recorded were as follows: hematological malignancies, thyroid neoplasms, pancreas neoplasms, aplastic anemia, pancreatitis, pregnancy and pregnancy outcomes (including congenital anomalies).

The Safety Follow-up Analysis Set included all participants who had at least 1 safety follow-up period assessment visit.

The percentage of participants achieving HbA1c goals of < 7%, <= 6.5%, and < 6.5% through Week 28 were compared between treatments using the Cochran-Mantel-Haenszel (CMH) procedure, in which screening HbA1c strata and background diabetes therapy strata served as the stratification factors. The CMH analysis excluded measurements after initiation of rescue medication and study drug discontinuation with missing data treated as non-responder.

The Evaluable Analysis Set included all randomized participants who received at least 1 dose of randomized study medication and had a baseline and at least 1 post-baseline HbA1c assessment. For analysis of efficacy, a decision was made with agreement by the European Medicines Agency and the Food and Drug Administration to pool participants from both exenatide groups (Total EBID) and compare this pooled group with placebo.

Change from baseline in body weight is reported as adjusted LS mean values at Weeks 4, 12, 20 and 28. Baseline is defined as the last non-missing assessment (scheduled or unscheduled) on or prior to the first dose of study medication. A MMRM analysis was performed, excluding measurements after initiation of rescue medication and study drug discontinuation.

The Full Analysis Set included all randomized participants who received at least 1 dose of randomized study medication. For analysis of efficacy, a decision was made with agreement by the European Medicines Agency and the Food and Drug Administration to pool participants from both exenatide groups (Total EBID) and compare this pooled group with placebo.

Change from baseline in FSG is reported as adjusted LS mean values at Week 28. Baseline is defined as the last non-missing assessment (scheduled or unscheduled) on or prior to the first dose of study medication. An analysis of covariance (ANCOVA) analysis was performed, excluding measurements after initiation of rescue medication and study drug discontinuation.

The Full Analysis Set included all randomized participants who received at least 1 dose of randomized study medication. For analysis of efficacy, a decision was made with agreement by the European Medicines Agency and the Food and Drug Administration to pool participants from both exenatide groups (Total EBID) and compare this pooled group with placebo.

Change from baseline in SMBG measurements are reported as adjusted LS mean values at Week 28. SMBG measurements were taken before (pre-prandial) and 2 hours after (post-prandial) the 2 main meals of the day on 3 separate days during the week before baseline (Day 1) and Week 28. Post-prandial excursions were calculated as the difference between the pre-prandial and post-prandial blood glucose concentrations (post-prandial - pre-prandial) and averaged (mean) over the 2 main meals over the 3 separate days in each period. Baseline is defined as the last non-missing assessment (scheduled or unscheduled) on or prior to the first dose of study medication. An ANCOVA analysis was performed, excluding measurements after initiation of rescue medication and study drug discontinuation.

The Full Analysis Set included all randomized participants who received at least 1 dose of randomized study medication. For analysis of efficacy, a decision was made with agreement by the European Medicines Agency and the Food and Drug Administration to pool participants from both exenatide groups (Total EBID) and compare this pooled group with placebo.

Change from baseline in fasting serum insulin is reported as adjusted LS mean values at Week 28. Baseline is defined as the last non-missing assessment (scheduled or unscheduled) on or prior to the first dose of study medication. An ANCOVA analysis was performed, excluding measurements after initiation of rescue medication and study drug discontinuation.

The Full Analysis Set included all randomized participants who received at least 1 dose of randomized study medication. For analysis of efficacy, a decision was made with agreement by the European Medicines Agency and the Food and Drug Administration to pool participants from both exenatide groups (Total EBID) and compare this pooled group with placebo.

Change from baseline in HOMA-B and HOMA-S are reported as adjusted LS mean values at Week 28. Baseline is defined as the last non-missing assessment (scheduled or unscheduled) on or prior to the first dose of study medication. An ANCOVA analysis was performed, excluding measurements after initiation of rescue medication and study drug discontinuation.

The Full Analysis Set included all randomized participants who received at least 1 dose of randomized study medication. For analysis of efficacy, a decision was made with agreement by the European Medicines Agency and the Food and Drug Administration to pool participants from both exenatide groups (Total EBID) and compare this pooled group with placebo.

Participants were discontinued from the study due to failure to maintain glycemic control if either discontinuation reason on summary case report form was "Loss of glucose control" or AE with lower level Medical Dictionary for Regulatory Activities (MedDRA) term "Loss of control of blood sugar" or "Hyperglycaemia" leading to study drug discontinuation, using MedDRA Version 23.0.

The Full Analysis Set included all randomized participants who received at least 1 dose of randomized study medication. For analysis of efficacy, a decision was made with agreement by the European Medicines Agency and the Food and Drug Administration to pool participants from both exenatide groups (Total EBID) and compare this pooled group with placebo.