An adverse event (AE) was any untoward medical occurrence in a participant or clinical investigation participant administered a study drug, which did not necessarily have a causal relationship with the treatment. An AE could therefore have been any unfavorable and unintended sign (including an abnormal laboratory finding), symptom, or disease temporally associated with the use of the study drug whether or not it was considered related to the study drug. The number of participants who experienced an AE is reported.

All participants that received ≥1 dose of study treatment

An adverse event (AE) was any untoward medical occurrence in a participant or clinical investigation participant administered a study drug, which did not necessarily have a causal relationship with the treatment. An AE could therefore have been any unfavorable and unintended sign (including an abnormal laboratory finding), symptom, or disease temporally associated with the use of the study drug whether or not it was considered related to the study drug. The number of participants who discontinued study treatment due to an AE is reported.

All participants that received ≥1 dose of study treatment

An erythroid response in NTD participants was defined as a mean hemoglobin increase of ≥1.0 g/dL from baseline over an 8-week interval compared to baseline. Baseline hemoglobin for NTD participants was the average of two or more measurements performed during the screening period of either the base study (A536-04) or the extension study (A536-06). Hemoglobin measurements within 2 weeks following RBC transfusion were excluded from the analysis. NTD participants were participants who had received <4 units of red blood cells (RBCs) within 8 weeks prior to the first dose of study drug. An 8-week interval was defined as any consecutive 8 weeks during the study. The percentage of participants with a hemoglobin increase of ≥1.0 g/dL from baseline is presented.

All participants who received ≥1 dose of study treatment and received <4 units of RBCs within 8 weeks prior to the first dose of study drug and had data available for the analyses

An erythroid response in NTD participants was defined as a mean hemoglobin increase of ≥1.0 g/dL from baseline over a 12-week interval compared to baseline. Baseline hemoglobin for NTD participants was the average of two or more measurements performed during the screening period of either the base study (A536-04) or the extension study (A536-06). Hemoglobin measurements within 2 weeks following red blood cell (RBC) transfusion were excluded from the analysis. NTD participants were participants who had received <4 units of RBCs within 8 weeks prior to the first dose of study drug. A 12-week interval was defined as any consecutive 12 weeks during the study. The percentage of participants with a hemoglobin increase of ≥1.0 g/dL from baseline is presented.

All participants who received ≥1 dose of study treatment and received <4 units of RBCs within 8 weeks prior to the first dose of study drug and had data available for the analyses

An erythroid response in NTD participants was defined as a mean hemoglobin increase of ≥1.0 g/dL from baseline measured during Weeks 13 to 24. Baseline hemoglobin for NTD participants was the average of two or more measurements performed during the screening period of either the base study (A536-04) or the extension study (A536-06). Hemoglobin measurements within 2 weeks following red blood cell (RBC) transfusion were excluded from the analysis. NTD participants were participants who had received <4 units of RBCs within 8 weeks prior to the first dose of study drug. The percentage of participants with a hemoglobin increase of ≥1.0 g/dL from baseline is presented.

All participants who received ≥1 dose of study treatment and received <4 units of RBCs within 8 weeks prior to the first dose of study drug and had data available for the analyses

An erythroid response in NTD participants was defined as a mean hemoglobin increase of ≥1.0 g/dL measured during Weeks 37 to 48. Baseline hemoglobin for NTD participants was the average of two or more measurements performed during the screening period of either the base study (A536-04) or the extension study (A536-06). Hemoglobin measurements within 2 weeks following red blood cell (RBC) transfusion were excluded from the analysis. NTD participants were participants who had received <4 units of RBCs within 8 weeks prior to the first dose of study drug. The percentage of participants with a hemoglobin increase of ≥1.0 g/dL from baseline is presented.

All participants who received ≥1 dose of study treatment and received <4 units of RBCs within 8 weeks prior to the first dose of study drug and had data available for the analyses

An erythroid response in NTD participants was defined as a mean hemoglobin increase of ≥1.5 g/dL from baseline over an 8-week interval compared to baseline. Baseline hemoglobin for NTD participants was the average of two or more measurements performed during the screening period of either the base study (A536-04) or the extension study (A536-06). Hemoglobin measurements within 2 weeks following red blood cell (RBC) transfusion were excluded from the analysis. NTD participants were participants who had received <4 units of RBCs within 8 weeks prior to the first dose of study drug. An 8-week interval was defined as any consecutive 8 weeks during the study. The percentage of participants with a hemoglobin increase of ≥1.5 g/dL from baseline will is presented.

All participants who received ≥1 dose of study treatment and received <4 units of RBCs within 8 weeks prior to the first dose of study drug and had data available for the analyses

An erythroid response in NTD participants was defined as a mean hemoglobin increase of ≥1.5 g/dL from baseline over a 12-week interval compared to baseline. Baseline hemoglobin for NTD participants was the average of two or more measurements performed during the screening period of either the base study (A536-04) or the extension study (A536-06). Hemoglobin measurements within 2 weeks following red blood cell (RBC) transfusion were excluded from the analysis. NTD participants were participants who had received <4 units of RBCs within 8 weeks prior to the first dose of study drug. A 12-week interval was defined as any consecutive 12 weeks during the study. The percentage of participants with a hemoglobin increase of ≥1.5 g/dL from baseline is presented.

All participants who received ≥1 dose of study treatment and received <4 units of RBCs within 8 weeks prior to the first dose of study drug and had data available for the analyses

An erythroid response in NTD participants was defined as a mean hemoglobin increase of ≥1.5 g/dL from baseline measured during Weeks 13 to 24. Baseline hemoglobin for NTD participants was the average of two or more measurements performed during the screening period of either the base study (A536-04) or the extension study (A536-06). Hemoglobin measurements within 2 weeks following red blood cell (RBC) transfusion were excluded from the analysis. NTD participants were participants who had received <4 units of RBCs within 8 weeks prior to the first dose of study drug. The percentage of participants with a hemoglobin increase of ≥1.5 g/dL from baseline is presented.

All participants who received ≥1 dose of study treatment and received <4 units of RBCs within 8 weeks prior to the first dose of study drug and had data available for the analyses

An erythroid response in NTD participants was defined as a mean hemoglobin increase of ≥1.5 g/dL from baseline measured during Weeks 37 to 48. Baseline hemoglobin for NTD participants was the average of two or more measurements performed during the screening period of either the base study (A536-04) or the extension study (A536-06). Hemoglobin measurements within 2 weeks following red blood cell (RBC) transfusion were excluded from the analysis. NTD participants were participants who had received <4 units of RBCs within 8 weeks prior to the first dose of study drug. The percentage of participants with a hemoglobin increase of ≥1.5 g/dL from baseline is presented.

All participants who received ≥1 dose of study treatment and received <4 units of RBCs within 8 weeks prior to the first dose of study drug and had data available for the analyses

Transfusion burden for TD participants was defined as the ratio of RBC transfusion units (or milliliters) transfused during an 8-week interval divided by the duration of that interval compared to the ratio of RBC transfusion units 8 weeks prior to the start of treatment (baseline). TD participants were participants who had received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug). An 8-week interval was defined as any consecutive 8 weeks during the study. The percentage of participants with a ≥20% reduction in RBC transfusion burden from baseline is presented.

All participants who received ≥1 dose of study treatment and received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug) and had data available for the analyses

Transfusion burden for TD participants was defined as the ratio of RBC transfusion units (or milliliters) transfused during an 8-week interval divided by the duration of that interval compared to the ratio of RBC transfusion units 8 weeks prior to the start of treatment (baseline). TD participants were participants who had received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug). An 8-week interval was defined as any consecutive 8 weeks during the study. The percentage of participants with a ≥33% reduction in RBC transfusion burden from baseline is presented.

All participants who received ≥1 dose of study treatment and received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug) and had data available for the analyses

Transfusion burden for TD participants was defined as the ratio of RBC transfusion units (or milliliters) transfused during an 8-week interval divided by the duration of that interval compared to the ratio of RBC transfusion units 8 weeks prior to the start of treatment (baseline). TD participants were participants who had received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug). An 8-week interval was defined as any consecutive 8 weeks during the study. The percentage of participants with a ≥50% reduction in RBC transfusion burden from baseline is presented.

All participants who received ≥1 dose of study treatment and received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug) and had data available for the analyses

Transfusion burden for TD participants was defined as the ratio of RBC transfusion units (or milliliters) transfused during a 12-week interval divided by the duration of that interval compared to the ratio of RBC transfusion units 12 weeks prior to the start of treatment (baseline). TD participants were participants who had received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug). A 12-week interval was defined as any consecutive 12 weeks during the study. The percentage of participants with a ≥20% reduction in RBC transfusion burden from baseline is presented.

All participants who received ≥1 dose of study treatment and received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug) and had data available for the analyses

Transfusion burden for TD participants was defined as the ratio of RBC transfusion units (or milliliters) transfused during a 12-week interval divided by the duration of that interval compared to the ratio of RBC transfusion units 8 weeks prior to the start of treatment (baseline). TD participants were participants who had received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug). A 12-week interval was defined as any consecutive 12 weeks during the study. The percentage of participants with a ≥33% reduction in RBC transfusion burden from baseline is presented.

All participants who received ≥1 dose of study treatment and received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug) and had data available for the analyses

Transfusion burden for TD participants was defined as the ratio of RBC transfusion units (or milliliters) transfused during a 12-week interval divided by the duration of that interval compared to the ratio of RBC transfusion units 12 weeks prior to the start of treatment (baseline). TD participants were participants who had received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug). A 12-week interval was defined as any consecutive 12 weeks during the study. The percentage of participants with a ≥50% reduction in RBC transfusion burden from baseline is presented.

All participants who received ≥1 dose of study treatment and received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug) and had data available for the analyses

Transfusion burden for TD participants was defined as the ratio of RBC transfusion units (or milliliters) transfused during a Weeks 13 to 24 divided by the duration of that interval compared to the ratio of RBC transfusion units 12 weeks prior to the start of treatment (baseline). TD participants were participants who had received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug). The percentage of participants with a ≥20% reduction in RBC transfusion burden from baseline is presented.

All participants who received ≥1 dose of study treatment and received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug) and had data available for the analyses

Transfusion burden for TD participants was defined as the ratio of RBC transfusion units (or milliliters) transfused during a Weeks 13 to 24 divided by the duration of that interval compared to the ratio of RBC transfusion units 12 weeks prior to the start of treatment (baseline). TD participants were participants who had received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug). The percentage of participants with a ≥33% reduction in RBC transfusion burden from baseline is presented.

All participants who received ≥1 dose of study treatment and received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug) and had data available for the analyses

Transfusion burden for TD participants was defined as the ratio of RBC transfusion units (or milliliters) transfused during Weeks 13 to 24 divided by the duration of that interval compared to the ratio of RBC transfusion units 12 weeks prior to the start of treatment (baseline). TD participants were participants who had received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug). The percentage of participants with a ≥50% reduction in RBC transfusion burden from baseline is presented.

All participants who received ≥1 dose of study treatment and received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug) and had data available for the analyses

Transfusion burden for TD participants was defined as the ratio of RBC transfusion units (or milliliters) transfused during Weeks 37 to 48 divided by the duration of that interval compared to the ratio of RBC transfusion units 12 weeks prior to the start of treatment (baseline). TD participants were participants who had received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug). The percentage of participants with a ≥20% reduction in RBC transfusion burden from baseline is presented.

All participants who received ≥1 dose of study treatment and received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug) and had data available for the analyses

Transfusion burden for TD participants was defined as the ratio of RBC transfusion units (or milliliters) transfused during Weeks 37 to 48 divided by the duration of that interval compared to the ratio of RBC transfusion units 12 weeks prior to the start of treatment (baseline). TD participants were participants who had received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug). The percentage of participants with a ≥33% reduction in RBC transfusion burden from baseline is presented.

All participants who received ≥1 dose of study treatment and received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug) and had data available for the analyses

Transfusion burden for TD participants was defined as the ratio of RBC transfusion units (or milliliters) transfused during Weeks 37 to 48 divided by the duration of that interval compared to the ratio of RBC transfusion units 12 weeks prior to the start of treatment (baseline). TD participants were participants who had received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug). The percentage of participants with a ≥50% reduction in RBC transfusion burden from baseline is presented.

All participants who received ≥1 dose of study treatment and received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug) and had data available for the analyses

Transfusion independence for TD participants was defined as the percentage of participants who did not require RBC transfusion units (or milliliters) transfused for ≥8 weeks in the study after start of treatment. TD participants were participants who had received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug). The percentage of participants who maintained transfusion independence for ≥8 weeks is presented.

All participants who received ≥1 dose of study treatment and received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug) and had data available for the analyses

The reduction from baseline in red blood cell transfusions for TD participants was defined as the ratio of RBC transfusion units (or milliliters) transfused during an 8-week interval divided by the duration of that interval compared to the ratio of RBC transfusion units 8 weeks prior to the start of treatment (baseline). TD participants were participants who had received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug). An 8-week interval was defined as any consecutive 8 weeks during the study. The percentage change from baseline in RBC transfusions in TD participants is presented.

All participants who received ≥1 dose of study treatment and received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug) and had data available for the analyses

The change from baseline in red blood cell transfusions for TD participants was defined as the ratio of RBC transfusion units (or milliliters) transfused during a 12-week interval divided by the duration of that interval compared to the ratio of RBC transfusion units 12 weeks prior to the start of treatment (baseline). TD participants were participants who had received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug). A 12-week interval was defined as any consecutive 12 weeks during the study. The percentage change from baseline in RBC transfusions in TD participants is presented.

All participants who received ≥1 dose of study treatment and received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug) and had data available for the analyses

Time to erythroid response was defined as the time from first dose to the first date of any rolling 12-week window achieving a hemoglobin increase ≥1.0 g/dL. When there were multiple disjointed intervals with response, the longest interval was used. Participants with response ongoing by the analysis cutoff day were censored. The time to the first date of any rolling 12-week window achieving a hemoglobin increase ≥1.0 g/dL is presented.

All participants who received ≥1 dose of study treatment and received <4 units of red blood cells (RBCs) within 8 weeks prior to the first dose of study drug and had data available for the analyses

Time to erythroid response was defined as the time from first dose to the first date of any rolling 12-week window achieving a hemoglobin increase ≥1.5 g/dL. When there were multiple disjointed intervals with response, the longest interval was used. Participants with response ongoing by the analysis cutoff day were censored. The time to the first date of any rolling 12-week window achieving a hemoglobin increase ≥1.5 g/dL is presented.

All participants who received ≥1 dose of study treatment and received <4 units of red blood cells (RBCs) within 8 weeks prior to the first dose of study drug and had data available for the analyses

Time to erythroid response was defined as the time from first dose to the first date of any rolling 12-week window achieving a red blood cell (RBC) transfusion burden reduction compared to pretreatment of ≥33%. When there were multiple disjointed intervals with response, the longest interval was used. Participants with response ongoing by the analysis cutoff day were censored. The time to the first date of any rolling 12-week window achieving a red blood cell transfusion burden reduction compared to pretreatment of ≥33% is presented.

All participants who received ≥1 dose of study treatment and who have received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug) and had data available for the analyses

Time to erythroid response was defined as the time from first dose to the first date of any rolling 12-week window achieving a red blood cell (RBC) transfusion burden reduction compared to pretreatment of ≥50%. When there were multiple disjointed intervals with response, the longest interval was used. Participants with response ongoing by the analysis cutoff day were censored. The time to the first date of any rolling 12-week window achieving a red blood cell transfusion burden reduction compared to pretreatment of ≥50% is presented.

All participants who received ≥1 dose of study treatment and who have received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug) and had data available for the analyses

Duration of erythroid response was defined as the time from the starting date of the first rolling 12-week window achieving a hemoglobin increase of ≥1.0 g/dL to the last date of the last consecutive rolling 12-week window achieving a hemoglobin increase of ≥1.0 g/dL. When there were multiple disjointed intervals with response, the longest interval was used. Participants with response ongoing by the analysis cutoff day were censored. The duration of response for participants achieving a hemoglobin increase ≥1.0 g/dL is presented.

All participants who received ≥1 dose of study treatment and received <4 units of red blood cells (RBCs) within 8 weeks prior to the first dose of study drug and had data available for the analyses

Duration of erythroid response was defined as the time from the starting date of the first rolling 12-week window achieving a hemoglobin increase of ≥1.5 g/dL to the last date of the last consecutive rolling 12-week window achieving a hemoglobin increase of ≥1.5 g/dL. When there were multiple disjointed intervals with response, the longest interval was used. Participants with response ongoing by the analysis cutoff day were censored. The duration of response for participants achieving a hemoglobin increase ≥1.5 g/dL is presented.

All participants who received ≥1 dose of study treatment and received <4 units of red blood cells (RBCs) within 8 weeks prior to the first dose of study drug and had data available for the analyses

Duration of erythroid response was defined as the time from the starting date of the first rolling 12-week window achieving a red blood cell (RBC) transfusion burden reduction compared to pretreatment of ≥33% to the last date of the last consecutive rolling 12-week window achieving a RBC transfusion burden reduction compared to pretreatment of ≥33%. When there were multiple disjointed intervals with response, the longest interval was used. Participants with response ongoing by the analysis cutoff day were censored. The duration of response for participants achieving a RBC transfusion burden reduction compared to pretreatment of ≥33% is presented.

All participants who received ≥1 dose of study treatment and who have received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug) and had data available for the analyses

Duration of erythroid response was defined as the time from the starting date of the first rolling 12-week window achieving a red blood cell (RBC) transfusion burden reduction compared to pretreatment of ≥50% to the last date of the last consecutive rolling 12-week window achieving a RBC transfusion burden reduction compared to pretreatment of ≥50%. When there were multiple disjointed intervals with response, the longest interval was used. Participants with response ongoing by the analysis cutoff day were censored. The duration of response for participants achieving a RBC transfusion burden reduction compared to pretreatment of ≥50% is presented.

All participants who received ≥1 dose of study treatment and who have received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug) and had data available for the analyses

Pre-transfusion hemoglobin levels were calculated in TD participants. TD participants were participants who had received ≥4 units of red blood cells (RBC) every 8 weeks (confirmed over 6 months prior to the first dose of study drug). The baseline pre-transfusion hemoglobin level was an average of all hemoglobin values measured before the first dose of study drug given. The post-baseline pre-transfusion hemoglobin levels were calculated using the average of all hemoglobin values recorded before each transfusion that was required after the first dose of study drug was given. All hemoglobin levels measured within the 2 weeks following a red blood cell transfusion were excluded from the analysis. The change from baseline in pre-transfusion hemoglobin levels is presented.

All participants who received ≥1 dose of study treatment and received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug) and had data available for the analyses

Mean change from baseline in hemoglobin levels was calculated for multiple rolling 8-week intervals in NTD participants. NTD participants were participants who had received <4 units of red blood cells (RBC) within 8 weeks prior to the first dose of study treatment. The baseline pre-transfusion hemoglobin level was an average of all hemoglobin values measured before the first dose of study drug given. The post-baseline hemoglobin level was taken after every rolling 8-week interval producing multiple values for the mean change from baseline in hemoglobin. The maximum change from baseline in hemoglobin level among the multiple rolling 8-week intervals is presented.

All participants who received ≥1 dose of study treatment and received <4 units of RBCs within 8 weeks prior to the first dose of study drug and had data available for the analyses

Mean change from baseline in hemoglobin levels was calculated for multiple rolling 12-week intervals in NTD participants. NTD participants were participants who had received <4 units of red blood cells (RBC) within 8 weeks prior to the first dose of study treatment. The baseline pre-transfusion hemoglobin level was an average of all hemoglobin values measured before the first dose of study drug given. The post-baseline hemoglobin level was taken after every rolling 12-week interval producing multiple values for the mean change from baseline in hemoglobin. The maximum change from baseline in hemoglobin level among the multiple rolling 12-week intervals is presented.

All participants who received ≥1 dose of study treatment and received <4 units of RBCs within 8 weeks prior to the first dose of study drug and had data available for the analyses

Percent change from baseline in transfusion burden was calculated for multiple rolling 8-week intervals in TD participants. TD participants were participants who had received ≥4 units of red blood cells (RBC) every 8 weeks (confirmed over 6 months prior to the first dose of study drug). Transfusion burden for TD participants was defined as the ratio of RBC transfusion units (or milliliters) transfused during an 8-week interval divided by the duration of that interval compared to the ratio of RBC transfusion units 8 weeks prior to the start of treatment (baseline). The post-baseline transfusion burden was calculated after every rolling 8-week interval producing multiple values for the percent change from baseline in transfusion burden. The maximum percent change from baseline in transfusion burden among the multiple rolling 8-week intervals is presented.

All participants who received ≥1 dose of study treatment and who have received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug) and had data available for the analyses

Percent change from baseline in transfusion burden was calculated for multiple rolling 12-week intervals in TD participants. TD participants were participants who had received ≥4 units of red blood cells (RBC) every 8 weeks (confirmed over 6 months prior to the first dose of study drug). Transfusion burden for TD participants was defined as the ratio of RBC transfusion units (or milliliters) transfused during a 12-week interval divided by the duration of that interval compared to the ratio of RBC transfusion units 12 weeks prior to the start of treatment (baseline). The post-baseline transfusion burden was calculated after every rolling 12-week interval producing multiple values for the percent change from baseline in transfusion burden. The maximum percent change from baseline in transfusion burden among the multiple rolling 12-week intervals is presented.

All participants who received ≥1 dose of study treatment and who have received ≥4 units of RBCs every 8 weeks (confirmed over 6 months prior to the first dose of study drug) and had data available for the analyses

Blood samples were collected at pre-specified time intervals to determine erythropoietin. The percent change from baseline in mean concentration of erythropoietin was measured. Baseline was the last measurement prior to the first dose of study drug. Per protocol, the analysis was presented by transfusion status (non-transfusion and transfusion dependent).

All participants who received ≥1 dose of study treatment and had data available for erythropoietin analyses

Blood samples were collected at pre-specified time intervals to determine reticulocyte count. The percent change from baseline in reticulocyte count was measured. Baseline was the last measurement prior to the first dose of study drug. Per protocol, the analysis was presented by transfusion status (non-transfusion and transfusion dependent).

All participants who received ≥1 dose of study treatment and had data available for reticulocyte analyses

Blood samples were collected at pre-specified time intervals to determine nRBC count. The percent change from baseline in nRBC count was measured. Baseline was the last measurement prior to the first dose of study drug. Per protocol, the analysis was presented by transfusion status (non-transfusion and transfusion dependent).

All participants who received ≥1 dose of study treatment and had data available for nRBC analyses

Blood samples were collected at pre-specified time intervals to determine soluble transferrin receptor. The percent change from baseline in mean concentration of soluble transferrin receptor was measured. Baseline was the last measurement prior to the first dose of study drug. Per protocol, the analysis was presented by transfusion status (non-transfusion and transfusion dependent).

All participants who received ≥1 dose of study treatment and had data available for soluble transferrin receptor analyses

Blood samples were collected at pre-specified time intervals to determine serum ferritin. The percent change from baseline in mean concentration of serum ferritin was measured. Baseline was the last measurement prior to the first dose of study drug. Per protocol, the analysis was presented by transfusion status (non-transfusion and transfusion dependent).

All participants who received ≥1 dose of study treatment and had data available for serum ferritin analyses

Blood samples were collected at pre-specified time intervals to determine percent transferrin (iron) saturation. The percent change from baseline in mean concentration of percent transferrin (iron) saturation was measured. Baseline was the last measurement prior to the first dose of study drug. Per protocol, the analysis was presented by transfusion status (non-transfusion and transfusion dependent).

All participants who received ≥1 dose of study treatment and had data available for percent transferrin (iron) saturation analyses

Blood samples were to be collected at pre-specified time intervals to determine serum transferrin. Baseline was pre-specified to be the last measurement prior to the first dose of study drug. Per protocol, the analysis was planned to be presented by transfusion status (non-transfusion and transfusion dependent).

No data were collected for Percent Change From Baseline to End of Treatment in Serum Transferrin.

Blood samples were to be collected at pre-specified time intervals to determine serum iron. Baseline was pre-specified to be the last measurement prior to the first dose of study drug. Per protocol, the analysis was planned to be presented by transfusion status (non-transfusion and transfusion dependent).

No data were collected for Percent Change From Baseline to End of Treatment in Serum Iron.

Blood samples were to be collected at pre-specified time intervals to determine serum hepcidin. Baseline was pre-specified to be the last measurement prior to the first dose of study drug. Per protocol, the analysis was planned to be presented by transfusion status (non-transfusion and transfusion dependent).

No data were collected for Percent Change From Baseline to End of Treatment in Serum Hepcidin.

Blood samples were to be collected at pre-specified time intervals to determine serum total iron binding capacity. Baseline was pre-specified to be the last measurement prior to the first dose of study drug. Per protocol, the analysis was planned to be presented by transfusion status (non-transfusion and transfusion dependent).

No data were collected for Percent Change From Baseline to End of Treatment in Serum TIBC.

Blood samples were to be collected at pre-specified time intervals to determine serum non-transferrin bound iron. Baseline was pre-specified to be the last measurement prior to the first dose of study drug. Per protocol, the analysis was planned to be presented by transfusion status (non-transfusion and transfusion dependent).

No data were collected for Percent Change From Baseline to End of Treatment in NTBI.

Blood samples were collected at pre-specified time intervals to determine total bilirubin. The percent change from baseline in total bilirubin was measured. Baseline was the last measurement prior to the first dose of study drug.

All participants who received ≥1 dose of study treatment and had data available for total bilirubin analyses

Blood samples were collected at pre-specified time intervals to determine indirect bilirubin. The percent change from baseline in indirect bilirubin was measured. Baseline was the last measurement prior to the first dose of study drug.

All participants who received ≥1 dose of study treatment and had data available for indirect bilirubin analyses

Blood samples were collected at pre-specified time intervals to determine LDH. The percent change from baseline in LDH was measured. Baseline was the last measurement prior to the first dose of study drug.

All participants who received ≥1 dose of study treatment and had data available for LDH analyses

Blood samples were collected at pre-specified time intervals to determine LIC. The change from baseline in mean LIC for participants with baseline LIC <3 mg/g dry weight was measured using magnetic resonance imaging (MRI). Baseline was the last measurement prior to the first dose of study drug. Per protocol, the analysis was presented by transfusion status (non-transfusion and transfusion dependent).

All participants with baseline LIC <3 mg/g dry weight, who received ≥1 dose of study treatment, and had data available for the analyses

Blood samples were collected at pre-specified time intervals to determine LIC. The change from baseline in mean LIC for participants with baseline LIC ≥3 mg/g dry weight was measured using magnetic resonance imaging (MRI). Baseline was the last measurement prior to the first dose of study drug. Per protocol, the analysis was presented by transfusion status (non-transfusion and transfusion dependent).

All participants with baseline LIC ≥3 mg/g dry weight, who received ≥1 dose of study treatment, and had data available for the analyses

Blood samples were collected at pre-specified time intervals to determine LIC. The change from baseline in mean LIC for participants with baseline LIC <3 mg/g dry weight and who have used ICT within 84 days prior to the first dose of study drug or during the study treatment period was measured using magnetic resonance imaging (MRI). Baseline was the last measurement prior to the first dose of study drug. Per protocol, the analysis was presented by transfusion status (non-transfusion and transfusion dependent).

All participants with baseline LIC <3 mg/g dry weight, who have used ICT, who received ≥1 dose of study treatment, and had data available for the analyses

Blood samples were collected at pre-specified time intervals to determine LIC. The change from baseline in mean LIC for participants with baseline LIC ≥3 mg/g dry weight and who have used ICT within 84 days prior to the first dose of study drug or during the study treatment period was measured using magnetic resonance imaging (MRI). Baseline was the last measurement prior to the first dose of study drug. Per protocol, the analysis was presented by transfusion status (non-transfusion and transfusion dependent).

All participants with baseline LIC ≥3 mg/g dry weight, who have used ICT, who received ≥1 dose of study treatment, and had data available for the analyses

Blood samples were collected at pre-specified time intervals to determine LIC. The change from baseline in mean LIC for participants with baseline LIC <3 mg/g dry weight and who have not used ICT within 84 days prior to the first dose of study drug or during the study treatment period was measured using magnetic resonance imaging (MRI). Baseline was the last measurement prior to the first dose of study drug. Per protocol, the analysis was presented by transfusion status (non-transfusion and transfusion dependent).

All participants with baseline LIC <3 mg/g dry weight, who have not used ICT, who received ≥1 dose of study treatment, and had data available for the analyses

Blood samples were collected at pre-specified time intervals to determine LIC. The change from baseline in mean LIC for participants with baseline LIC ≥3 mg/g dry weight and who have not used ICT within 84 days prior to the first dose of study drug or during the study treatment period was measured using magnetic resonance imaging (MRI). Baseline was the last measurement prior to the first dose of study drug. Per protocol, the analysis was presented by transfusion status (non-transfusion and transfusion dependent).

All participants with baseline LIC ≥3 mg/g dry weight, who have not used ICT, who received ≥1 dose of study treatment, and had data available for the analyses

Blood samples were collected at multiple time points to determine the serum concentration of luspatercept. Serum concentrations that were below the limit of quantitation (LOQ) of the assay prior to the first dose were assigned a numerical value of zero. Post-treatment serum concentrations that were below the LOQ of the assay were treated as missing. Serum concentrations assigned a value of missing were omitted from the analysis. The LOQ of the assay was 50 ng/mL.

All participants who received ≥1 dose of study treatment and had data available for serum concentration analyses