Percentage of participants who were in deep molecular response MR4.5 (IS) at 24 months measured in a standardized EUTOS (European Treatment and Outcome Study for CML) MR4.5 laboratory. MR4.5 was defined as either (i) detectable disease ≤ 0.0032% BCR-ABL (fusion gene from breakpoint cluster region and Abelson genes) (IS) or (ii) undetectable disease in cDNA with 32000-99999 ABL1 transcripts or 77000-239999 glucuronidase beta (GUSB) transcripts.

Responders: Participants with a MR4.5 at 24 months, or if the assessment at this time point was missing, with a MR4.5 at 21 months Non-responders: Participants dropping out early or not providing sufficient data for any other reason. Participants who achieved MR4.5 before 24 months, but was no longer in MR4.5 at 24 months or progressed (or was no longer in MR4.5 at 21 months if evaluation at 24 months was missing).

Confidence intervals were calculated based on the Exact Clopper-Pearson method.

Full Analysis Set (FAS): all participants who entered study and received at least one dose of study drug and had at least one post-baseline assessment of the primary efficacy variable. Only participants in the FAS with typical BCR-ABL transcripts (b2a2 and/or b3a2) were included in the analysis. Participants with atypical transcripts were excluded from the analysis, because their transcripts could not be used for molecular response determination

Percentage of participants with MR4 at 24 months of study treatment. MR4 (IS) is defined as either (i) detectable disease ≤0.01% BCR-ABL by IS or (ii) undetectable disease in cDNA with 10000 - 31999 ABL1 transcripts or 24000 - 76999 GUSB transcripts.

Confidence intervals were calculated based on the Exact Clopper-Pearson method.

Full Analysis Set (FAS): all participants who entered study and received at least one dose of study drug and had at least one post-baseline assessment of the primary efficacy variable. Only participants in the FAS with typical BCR-ABL transcripts (b2a2 and/or b3a2) were included in the analysis. Participants with atypical transcripts were excluded from the analysis, because their transcripts could not be used for molecular response determination

Percentage of participants with MMR at 12 months of study treatment. MMR is defined as ≤ 0.1% BCR-ABL by IS, or equivalent to ≥ 3 log reduction of BCR-ABL transcript from standardized baseline.

Confidence intervals were calculated based on the Exact Clopper-Pearson method.

FAS: all participants who entered study and received at least one dose of study drug and had at least one post-baseline assessment of the primary efficacy variable. Only participants in the FAS with typical BCR-ABL transcripts (b2a2 and/or b3a2) were included in the analysis. Participants with atypical transcripts were excluded from the analysis, because their transcripts could not be used for molecular response determination

Percentage of participants with CCyR at 6 months of study treatment. Cytogenetic response was assessed as the percentage of Philadelphia positive (Ph+) metaphases in the bone marrow (a review of a minimum of 20 metaphases was required). CCyR was defined as a value of 0% Ph+ metaphases in bone marrow.

Confidence intervals were calculated based on the Exact Clopper-Pearson method.

FAS: all participants who entered study and received at least one dose of study drug and had at least one post-baseline assessment of the primary efficacy variable. Only participants with review of a minimum of 20 metaphases at month 6 were included in the analysis

Progression-free survival is defined as the time from the date of start of study treatment to the date of the first documented disease progression to accelerated phase (AP)/ blast crisis (BC) or death from any cause, whichever is earlier.

AP is defined as:

• 15% blasts in the peripheral blood or one marrow aspirate, but <30% blasts in both the peripheral blood and bone marrow aspirate
• 30% blasts plus promyelocytes in peripheral blood or bone marrow aspirate
• 20% basophils in the peripheral blood or bone marrow Thrombocytopenia (<100 x 109/Liter) that is unrelated to therapy Evidence of clonal evolution

BC is defined as:

≥ 30% blasts in peripheral blood or bone marrow aspirate Appearance of extramedullary involvement other than hepatosplenomegaly proven by biopsy (i.e., chloroma)

FAS: all participants who entered study and received at least one dose of study drug and had at least one post-baseline assessment of the primary efficacy variable.

Time to progression to AP/BC is defined as the time from the date of start of study treatment to the date of earliest transformation to AP/BC, or CML-related death.

AP is defined as:

• 15% blasts in the peripheral blood or one marrow aspirate, but <30% blasts in both the peripheral blood and bone marrow aspirate
• 30% blasts plus promyelocytes in peripheral blood or bone marrow aspirate
• 20% basophils in the peripheral blood or bone marrow Thrombocytopenia (<100 x 109/L) that is unrelated to therapy Evidence of clonal evolution (with consensus of SC only)

BC is defined as:

≥ 30% blasts in peripheral blood or bone marrow aspirate Appearance of extramedullary involvement other than hepatosplenomegaly proven by biopsy (i.e., chloroma)

FAS: all participants who entered study and received at least one dose of study drug and had at least one post-baseline assessment of the primary efficacy variable.

The EORTC QLQ-C30 is a patient completed 30 item questionnaire that is composed of both multi-item scales and single-item measures. These include five functional scales, three symptom scales, six single items and a GHS/QoL scale.

The GHS/QoL scale has 7 possible scores of responses (1=very poor to 7=excellent). Scores were averaged and transformed to 0 to 100. Higher scores indicate better quality of life.

A positive change from Baseline indicates improvement.

FAS: all participants who entered study and received at least one dose of study drug and had at least one post-baseline assessment of the primary efficacy variable. Number analyzed indicated number of participants with available data for this outcome measure at specified timepoints.

The EORTC QLQ-C30 is a patient completed 30 item questionnaire that is composed of both multi-item scales and single-item measures. These include five functional scales, three symptom scales, six single items and a global health status/QoL scale.

For the physical functioning scale, participants self-rated levels of difficulty in doing strenuous activities, taking a walk, how much they needed to stay in bed or a chair, or needed help with eating, dressing, bathing, using the toilet. The physical functioning scale had 4 possible scores (1=not at all, 2=a little, 3=quite a bit, 4=very much). Scores were averaged and transformed to 0 to 100. Higher scores indicate better functioning. A positive change from baseline indicates improvement in physical functioning.

FAS: all participants who entered study and received at least one dose of study drug and had at least one post-baseline assessment of the primary efficacy variable. Number analyzed indicated number of participants with available data for this outcome measure at specified timepoints.

The EORTC QLQ-C30 is a patient completed 30 item questionnaire that is composed of both multi-item scales and single-item measures. These include five functional scales, three symptom scales, six single items and a global health status/QoL scale.

For the role functioning scale, participants self-rated how much they were limited in doing work or daily activities, or in pursuing hobbies or other leisure time activities during the past week. The role functioning scale had 4 possible scores (1=not at all, 2=a little, 3=quite a bit, 4=very much). Scores were averaged and transformed to 0 to 100. Higher scores indicate better functioning. A positive change from baseline indicates improvement in role functioning.

FAS: all participants who entered study and received at least one dose of study drug and had at least one post-baseline assessment of the primary efficacy variable. Number analyzed indicated number of participants with available data for this outcome measure at specified timepoints.

The EORTC QLQ-C30 is a patient completed 30 item questionnaire that is composed of both multi-item scales and single-item measures. These include five functional scales, three symptom scales, six single items and a global health status/QoL scale.

For the emotional functioning scale, participants self-rated how much they felt tense, worried, irritable or depressed during the past week. The emotional functioning scale had 4 possible scores (1=not at all, 2=a little, 3=quite a bit, 4=very much). Scores were averaged and transformed to 0 to 100. Higher scores indicate better functioning. A positive change from baseline indicates improvement in emotional functioning.

FAS: all participants who entered study and received at least one dose of study drug and had at least one post-baseline assessment of the primary efficacy variable. Number analyzed indicated number of participants with available data for this outcome measure at specified timepoints.

The EORTC QLQ-C30 is a patient completed 30 item questionnaire that is composed of both multi-item scales and single-item measures. These include five functional scales, three symptom scales, six single items and a global health status/QoL scale.

For the cognitive functioning scale, participants self-rated the extent of difficulty in concentrating on things or remembering things during the past week. The cognitive functioning scale had 4 possible scores (1=not at all, 2=a little, 3=quite a bit, 4=very much). Scores were averaged and transformed to 0 to 100. Higher scores indicate better functioning. A positive change from baseline indicates improvement in cognitive functioning.

FAS: all participants who entered study and received at least one dose of study drug and had at least one post-baseline assessment of the primary efficacy variable. Number analyzed indicated number of participants with available data for this outcome measure at specified timepoints.

The EORTC QLQ-C30 is a patient completed 30 item questionnaire that is composed of both multi-item scales and single-item measures. These include five functional scales, three symptom scales, six single items and a global health status/QoL scale.

For the social functioning scale, participants self-rated how much their physical condition or medical treatment interfered with their family life and social activities during the past week. The social functioning scale had 4 possible scores (1=not at all, 2=a little, 3=quite a bit, 4=very much). Scores were averaged and transformed to 0 to 100. Higher scores indicate better functioning. A positive change from baseline indicates improvement in social functioning.

FAS: all participants who entered study and received at least one dose of study drug and had at least one post-baseline assessment of the primary efficacy variable. Number analyzed indicated number of participants with available data for this outcome measure at specified timepoints.

The EORTC QLQ-CML 24 is an internationally developed disease specific health-related quality of life questionnaire for CML patients. The questionnaire is composed of four multi-item scales and two single-item scales. The module consists of 24 items assessing symptoms burden (13 items), impact on worry/mood (4 items), impact on daily life (3 items), satisfaction with care and information (2 items) body image problems (1 item) and satisfaction with social life (1 item). The items were measured on four levels: 1=not at all, 2=a little, 3=quite a bit, 4=very much. For each domain, scores were averaged and transformed to 0 to 100. A higher score in symptom burden domain indicates a worse outcome. A negative change from baseline indicates improvement.

FAS: all participants who entered study and received at least one dose of study drug and had at least one post-baseline assessment of the primary efficacy variable. Number analyzed indicated number of participants with available data for this outcome measure at specified timepoints.

The EORTC QLQ-CML 24 is an internationally developed disease specific health-related quality of life questionnaire for CML patients. The questionnaire is composed of four multi-item scales and two single-item scales. The module consists of 24 items assessing symptoms burden (13 items), impact on worry/mood (4 items), impact on daily life (3 items), satisfaction with care and information (2 items) body image problems (1 item) and satisfaction with social life (1 item). The items were measured on four levels: 1=not at all, 2=a little, 3=quite a bit, 4=very much. For each scale, scores were averaged and transformed to 0 to 100. A higher score in impact on worry/mood domain indicates a worse outcome. A negative change from baseline indicates improvement.

FAS: all participants who entered study and received at least one dose of study drug and had at least one post-baseline assessment of the primary efficacy variable. Number analyzed indicated number of participants with available data for this outcome measure at specified timepoints.

The EORTC QLQ-CML 24 is an internationally developed disease specific health-related quality of life questionnaire for CML patients. The questionnaire is composed of four multi-item scales and two single-item scales. The module consists of 24 items assessing symptoms burden (13 items), impact on worry/mood (4 items), impact on daily life (3 items), satisfaction with care and information (2 items) body image problems (1 item) and satisfaction with social life (1 item). The items were measured on four levels: 1=not at all, 2=a little, 3=quite a bit, 4=very much. For each domain, scores were averaged and transformed to 0 to 100. A higher score in impact on daily life domain indicates a worse outcome. A negative change from baseline indicates improvement.

FAS: all participants who entered study and received at least one dose of study drug and had at least one post-baseline assessment of the primary efficacy variable. Number analyzed indicated number of participants with available data for this outcome measure at specified timepoints.

The EORTC QLQ-CML 24 is an internationally developed disease specific health-related quality of life questionnaire for CML patients. The questionnaire is composed of four multi-item scales and two single-item scales. The module consists of 24 items assessing symptoms burden (13 items), impact on worry/mood (4 items), impact on daily life (3 items), satisfaction with care and information (2 items) body image problems (1 item) and satisfaction with social life (1 item). The items were measured on four levels: 1=not at all, 2=a little, 3=quite a bit, 4=very much. For each domain, scores were averaged and transformed to 0 to 100. A higher score in satisfaction with care and information domain indicates a higher level of satisfaction. A positive change from baseline indicates increasing satisfaction.

FAS: all participants who entered study and received at least one dose of study drug and had at least one post-baseline assessment of the primary efficacy variable. Number analyzed indicated number of participants with available data for this outcome measure at specified timepoints.

The EORTC QLQ-CML 24 is an internationally developed disease specific health-related quality of life questionnaire for CML patients. The questionnaire is composed of four multi-item scales and two single-item scales. The module consists of 24 items assessing symptoms burden (13 items), impact on worry/mood (4 items), impact on daily life (3 items), satisfaction with care and information (2 items) body image problems (1 item) and satisfaction with social life (1 item). The items were measured on four levels: 1=not at all, 2=a little, 3=quite a bit, 4=very much. For each domain, scores were averaged and transformed to 0 to 100. A higher score in body image problems domain indicates a worse outcome. A negative change from baseline indicates improvement.

FAS: all participants who entered study and received at least one dose of study drug and had at least one post-baseline assessment of the primary efficacy variable. Number analyzed indicated number of participants with available data for this outcome measure at specified timepoints.

The EORTC QLQ-CML 24 is an internationally developed disease specific health-related quality of life questionnaire for CML patients. The questionnaire is composed of four multi-item scales and two single-item scales. The module consists of 24 items assessing symptoms burden (13 items), impact on worry/mood (4 items), impact on daily life (3 items), satisfaction with care and information (2 items) body image problems (1 item) and satisfaction with social life (1 item). The items were measured on four levels: 1=not at all, 2=a little, 3=quite a bit, 4=very much. For each domain, scores were averaged and transformed to 0 to 100. A higher score in satisfaction with social life domain indicates a higher level of satisfaction. A positive change from baseline indicates increasing satisfaction.

FAS: all participants who entered study and received at least one dose of study drug and had at least one post-baseline assessment of the primary efficacy variable. Number analyzed indicated number of participants with available data for this outcome measure at specified timepoints.