The study was a retrospective, non-interventional patient chart review and used a panel of oncologists/hematologists from the US to collect real-world clinical outcomes of patients with CML-CP in 3L+ and those with the T315I mutation.

This study included two distinct cohorts of CML-CP patients; that is, patients with CML-CP who initiated 3L for CML-CP (i.e., 3L cohort) and patients with CML-CP with T315I mutation (i.e., T315I cohort).

Study design for the analyses of the 3L cohort:

• The index date: date of initiation of 3L therapy for CML-CP
• The study period: period of ≥ 24 months following the index date unless the patient died before
• Patient characteristics were measured at CML diagnosis and at the index date
• The clinical outcomes of interest were measured during the study period

Study design for the analyses of the T315I cohort:

• The index date: date of initiation of a line of therapy identified as the T315I line of interest (i.e., identification of T315I mutation before initiation or over the course of line of therapy)
• The study period: period of ≥ 24 months following the index date unless the patient died before
• Patient characteristics were measured at CML diagnosis and at the index date.
• The clinical outcomes of interest were measured during the study period

Inclusion Criteria:

Physician selection

Physicians were eligible to participate in the study if they fulfilled all of the following criteria:

• Completed medical subspecialty training
• Reported hematology, medical oncology, or any other oncology subspecialties as the primary medical subspecialty
• Were responsible for treatment decisions and follow-up for ≥ 1 adult patient with Ph+ CML-CP who received a 3L or those with the T315I mutation since January 2013 (the date from which molecular monitoring response on the International Scale (IS) became a more standard procedure/commonly available
• Had access to molecular monitoring results reported on the IS, and with a sensitivity level of precision for molecular response of MR3 (BCR ABL1/ABL1 ≤ 0.1% or 3-log reduction) or better

Patient selection Participating physicians were directed to provide information on patients who were included into the following separate cohorts. Each participating physician contributed up to 5 patient medical charts from each cohort.

For the 3L cohort:

• Adult patients diagnosed with Ph+ CML-CP who initiated a 1L therapy, switched to a 2L therapy, and initiated a 3L therapy for CML-CP
• All lines of therapy (TKIs or other CML treatments) received outside of an interventional clinical trial setting
• 3L therapy was initiated on or after January 1st, 2013 (when molecular monitoring became a common practice in CML monitoring) and no later than November 30th, 2018, to have a minimum of 2 years of follow-up after therapy initiation, except if the patient died before

For the T315I cohort:

• Adult patients diagnosed with Ph+ CML-CP who initiated ≥ 1 line of therapy for Ph+ CML-CP and T315I mutation was identified
• All lines of therapy (TKIs or other CML treatments) received outside of an interventional clinical trial setting
• Line of therapy identified as the T315I line of interest was initiated on or after January 1st, 2013, and no later than November 30th, 2018, to have a minimum of 2 years of follow-up after therapy initiation, except if the patient died before

For both cohorts:

• Patients with Ph+ CML-CP for whom the physician had complete information on the CML related care from CML diagnosis and for ≥ 2 years after the initiation of line of therapy of interest (i.e., 3L or line of therapy identified as the T315I line of interest), unless the patient died before. Complete information included: CML treatments, treatment duration, routine laboratory (e.g., complete blood count (CBC), BCR-ABL), CML status (e.g., SOKAL risk score, CP/accelerated phase (AP)/ blast crisis (BC)), medications, and clinical status (e.g., history, physical exam)
• The physician had access to molecular monitoring results reported on the IS from initiation of the line of therapy of interest and with a sensitivity level of precision for molecular response of MR3 (BCR-ABL1/ABL1≤0.1% or 3-log reduction) or better Of note, the cohorts were not mutually exclusive such that patients included in the 3L cohort with T315I mutation was included in the T315I cohort. Thereafter, there was an oversampling of patients with T315I mutation. Patients from the T315I cohort from the oversampling with a 3L were not included in the 3L cohort.

Exclusion Criteria:

- Physicians and patients who did not meet study inclusion criteria detailed above were excluded.

Patients with Chronic Myeloid Leukemia and those with T315I Mutation in the United States