The Baseline average was calculated as the average of the Hb values obtained at the screening visit closest to the date of randomization and the randomization visit. The average for the Primary Efficacy Period was calculated as the average Hb value over Weeks 24 to 36. Analysis was conducted using an analysis of covariance (ANCOVA) model with multiple imputation for missing data with Baseline hemoglobin concentration (<10.0 versus ≥10.0 g/dL), geographic region (United States [US] versus European Union [EU] versus Rest of World [ROW]), and New York Heart Association congestive heart failure (NYHA CHF) class (Class 0 [no CHF] or I versus II or III) as covariates.

Randomized Population: All participants randomized. Analyses of this population were based on the randomized treatment.

MACE was defined as all-cause mortality, non-fatal myocardial infarction (MI), or non-fatal stroke. The primary safety outcome was positively adjudicated first MACE, which was defined as any death, Endpoint Adjudication Committee (EAC)-confirmed non-fatal MI, or EAC-confirmed non-fatal stroke occurring between the first dose date and each participant's last participation date. PROTECT MACE results and analysis, by design, was performed on pooled data from studies AKB-6548-CI-0014 (NCT02648347) and AKB-6548-CI-0015 (NCT02680574). Results and statistical analysis from study AKB-6548-CI-0015 has been reported in below table and under section "Statistical Analysis 1". Results and statistical analysis of the pooled data from studies AKB-6548-CI-0014 and AKB-6548-CI-0015 has been reported under section "Statistical Analysis 2" of this outcome measure.

Safety Population (PRO2TECT): All participants from the PRO2TECT (Non-dialysis-dependent chronic kidney disease [NDD-CKD]) population who received 1 or more doses of study drug. Only those participants with MACE events were analyzed for this outcome measure.

The Baseline average was calculated as the average of the Hb values obtained at the screening visit closest to the date of randomization and the randomization visit. The average for the Secondary Efficacy Period was calculated as the average Hb value over Weeks 40 to 52. Analysis was conducted using an ANCOVA model with multiple imputation for missing data with Baseline hemoglobin concentration (<10.0 versus ≥10.0 g/dL), geographic region (US versus EU versus ROW), and NYHA CHF class (Class 0 [no CHF] or I versus II or III) as covariates.

Randomized Population. Analyses of this population were based on the randomized treatment.

MACE was defined as all-cause mortality, non-fatal MI, or non-fatal stroke. Hospitalization for EAC adjudicated heart failure included presentation of participants to an acute care facility requiring an overnight hospitalization (change in calendar day) with an exacerbation of heart failure requiring treatment. EAC confirmed thromboembolic events for this secondary outcome measure included arterial thrombosis, deep vein thrombosis, and pulmonary embolism. PROTECT MACE results and analysis, by design, was performed on pooled data from studies AKB-6548-CI-0014 (NCT02648347) and AKB-6548-CI-0015 (NCT02680574). Results and statistical analysis from study AKB-6548-CI-0015 has been reported in below table and under section "Statistical Analysis 1". Results and statistical analysis of the pooled data from studies AKB-6548-CI-0014 and AKB-6548-CI-0015 has been reported under section "Statistical Analysis 2" of this outcome measure.

Safety Population (PRO2TECT). Only those participants with MACE plus hospitalization for heart failure or thromboembolic event excluding vascular access thrombosis were analyzed for this outcome measure.

MACE was defined as all-cause mortality, non-fatal MI, or non-fatal stroke. Cardiovascular MACE analysis differed from the primary MACE endpoint as it included only deaths adjudicated by the EAC as cardiovascular deaths (i.e, only EAC-confirmed cardiovascular deaths) in addition to first events of non-fatal MI or non-fatal stroke. PROTECT MACE results and analysis, by design, was performed on pooled data from studies AKB-6548-CI-0014 (NCT02648347) and AKB-6548-CI-0015 (NCT02680574). Results and statistical analysis from study AKB-6548-CI-0015 has been reported in below table and under section "Statistical Analysis 1". Results and statistical analysis of the pooled data from studies AKB-6548-CI-0014 and AKB-6548-CI-0015 has been reported under section "Statistical Analysis 2" of this outcome measure.

Safety Population (PRO2TECT). Only those participants with cardiovascular MACE events were analyzed for this outcome measure.

Cardiovascular death included EAC adjudicated fatal MI, pump failure, sudden death, presumed sudden death, fatal stroke, fatal pulmonary embolism, cardiovascular procedure-related death, other cardiovascular death, and presumed cardiovascular death. PROTECT MACE results and analysis, by design, was performed on pooled data from studies AKB-6548-CI-0014 (NCT02648347) and AKB-6548-CI-0015 (NCT02680574). Results and statistical analysis from study AKB-6548-CI-0015 has been reported in below table and under section "Statistical Analysis 1". Results and statistical analysis of the pooled data from studies AKB-6548-CI-0014 and AKB-6548-CI-0015 has been reported under section "Statistical Analysis 2" of this outcome measure.

Safety Population (PRO2TECT). Only those participants with cardiovascular death were analyzed for this outcome measure.