Safety and Efficacy Study of Roxadustat to Treat Anemia in Patients With Chronic Kidney Disease (CKD), Not on Dialysis

Baseline Hb was defined as the mean of the last 3 central laboratory Hb values from the screening and randomization visits. Mean change in Hb from baseline to mean value from Week 28 to Week 52 was analyzed using a missing at random (MAR) based multiple imputation analysis of covariance (ANCOVA) model with baseline Hb, baseline estimated glomerular filtration rate (eGFR), cardiovascular (CV) history, geographic region and treatment group as fixed effect covariates. The adjusted least squares (LS) mean estimates of change from baseline to mean during Week 28 to Week 52 are presented.

Hb response was defined as:

• Hb ≥ 11.0 g/dL and Hb increase from baseline by ≥ 1.0 g/dL for participants with baseline Hb > 8.0 g/dL; or
• Hb increase from baseline by ≥ 2.0 g/dL, for participants with baseline Hb ≤ 8.0 g/dL at 2 consecutive visits (with available data) separated at least 5 days during the first 24 weeks of treatment without having received rescue therapy (red blood cell [RBC] transfusion, erythropoietin analogue, or intravenous [IV] iron) prior to Hb response. The percentage of participants with an Hb response during the first 24 weeks of treatment is presented.

Baseline hsCRP was quantified from stored biomarker samples obtained at randomization. Baseline Hb was defined as the mean of the last 3 central laboratory Hb values from the screening and randomization visits. Mean change in Hb from baseline to mean value during Week 28 to Week 52 was analyzed using a MAR based multiple imputation ANCOVA model with baseline Hb, baseline eGFR, CV history, geographic region and treatment group as fixed effect covariates. The adjusted LS mean estimates of change from baseline in participants with baseline hsCRP >ULN to mean during Week 28 to Week 52 are presented.

Proportion of total time of interpolated Hb values ≥10 g/dL was calculated as the time the linearly interpolated curve between measurements ≥10 g/dL divided by the time between measurements from Week 28 to Week 52. Proportion of total time was analyzed using an ANCOVA model with baseline Hb and baseline eGFR as covariates, and CV history, geographic region and treatment group as fixed effects. The adjusted LS mean estimates of change from Week 28 to Week 52 are presented.

Proportion of total time of interpolated Hb values within the interval of 10 to 12 g/dL was calculated as the time the linearly interpolated curve between measurements were within 10 to 12 g/dL divided by the time between measurements from Week 28 to Week 52. Proportion of total time was analyzed using an ANCOVA model with baseline Hb and baseline eGFR as covariates, and CV history, geographic region and treatment group as fixed effects. The adjusted LS mean estimates of change from Week 28 to Week 52 are presented.

Baseline LDL was defined as the last result obtained prior to randomization. Mean changes in LDL cholesterol from baseline to Week 24 was analyzed using an ANCOVA model with baseline LDL, baseline Hb and baseline eGFR as covariates, and CV history, geographic region and treatment group as fixed effects. The adjusted LS mean estimates of change from baseline to Week 24 are presented.

Time-to-first rescue therapy (IV iron, RBC transfusion or erythropoietin analogue) was calculated as (date of first rescue therapy, or date of censoring if no rescue therapy was taken) minus (date of first dose of IP) +1. Event rate was calculated as (number of participants with event) divided by (the total number of days at risk for event across all participants in given group divided by 365.25) multiplied by 100. The event rate is presented for participants with events.

Time-to-first RBC rescue therapy was calculated as (date of first RBC rescue therapy, or date of censoring if no rescue therapy was taken) minus (date of first dose of IP) +1. Event rate was calculated as (number of participants with event) divided by (the total number of days at risk for event across all participants in given group divided by 365.25) multiplied by 100. The event rate is presented for participants with events.

SF-36 is a Quality of Life (QoL) scale comprising 8 domains of health status: Physical Functioning, Role-Physical, Bodily Pain, General Health, Vitality, Social Functioning, Role Emotional and Mental Health. Each domain score is on a scale from 0-100 (worst health possible to best health possible); higher scores indicate better health status. Mean change in SF-36 Vitality sub-score from baseline to mean from Week 12 to Week 28 was analyzed using a mixed model for repeated measures (MMRM) with terms for baseline score, treatment group, baseline Hb, baseline eGFR, CV history, geographic region, visit and treatment-by-visit interaction as fixed effects and participant as random effect. The adjusted LS mean estimates of change from baseline to mean score from Week 12 to Week 28 are presented.

Baseline eGFR was defined as the mean of all available central laboratory values prior to or at randomization. Rate of change in eGFR from baseline during the entire treatment period (in millilitres/minute/1.73 meters squared/years [mL/min/1.73m^2/years]) was estimated using a random effects model using all post-baseline eGFR values prior to initiation of dialysis/transplant. Baseline eGFR, baseline Hb, geographic region, CV history, treatment group and post-baseline eGFR measurement time were used as fixed effects and participant and time (years) as random effects, ie, random intercept and slope.

SF-36 is a QoL scale comprising 8 domains of health status: Physical Functioning, Role-Physical, Bodily Pain, General Health, Vitality, Social Functioning, Role Emotional and Mental Health. Each domain score is on a scale from 0-100 (worst health possible to best health possible); higher scores indicate better health status. Mean change in SF-36 Physical Functioning sub-score from baseline to mean from Week 12 to Week 28 was analyzed using a MMRM with terms for baseline score, treatment group, baseline Hb, baseline eGFR, CV history, geographic region, visit and treatment-by-visit interaction as fixed effects and participant as random effect. The adjusted LS mean estimates of change from baseline to mean score from Week 12 to Week 28 are presented.