Real World Treatment Patterns in Chronic Myeloid Leukemia Patients Treated with Tyrosine Kinase Inhibitors in First Line in an Integrated Healthcare System

INTRODUCTION: BCR-ABL Tyrosine Kinase Inhibitors (TKI) have demonstrated efficacy in patients with chronic myeloid leukemia (CML) in the clinical trial setting. However, additional real world data evaluating first line TKI treatment patterns are still needed. Moreover, real world data captured within integrated, managed care organizations are scarce. This study evaluates treatment patterns among newly diagnosed CML patients initiated on frontline TKI therapy within the Kaiser Permanente Southern California (KPSC) integrated managed care organization.

METHODS: A retrospective analysis was conducted using KPSC data between 1/1/2007 to 12/31/2014. Patients were identified with CML diagnosis from the KPSC cancer registry database during 1/1/2007 to 12/31/2013. Patients > 18 years on date of CML diagnosis, with no history of other prior cancers or stem cell transplant during the 12 months prior to diagnosis date were eligible for inclusion. Index date was defined as the date of the first TKI prescription identified during the period 1/1/2007 to 12/31/2013, and no history of TKI use in the prior 12 months. All patients had continuous membership and drug benefit eligibility 12 months prior to index date. Patients were followed from their index date until one of the following outcomes: 1) death, 2) disenrollment from health plan, 3) switch to another TKI, 4) discontinuation of index TKI therapy, or 5) end of study period of 12/31/2014. Discontinuation, switch and persistence of index TKI therapy were defined using a 60 day gap calculation: estimated date of depleting the second to last prescription supply, plus days' supply of the last prescription, plus 60 days. If a patient did not fill any TKI prescription before or during the 60 day gap, this was labeled as a discontinuation. If a patient filled a different TKI prescription before or during the 60 day gap, this was labeled as switch, and if a patient refilled their index TKI prescription, then this was labeled as continuation of index therapy.

RESULTS: 216 patients were identified with incident CML and with first line TKI therapy: 19 (8.8%) on dasatinib, 189 (87.5%) on imatinib, and 8 (3.7%) on nilotinib. The mean age on diagnosis date was 53 years and 63% were male. The mean age for patients who initiated nilotinib was 45.5 years, 48.3 years for dasatinib, and 54.1 years for imatinib. The Charlson comorbidity index was highest for patients on dasatinib (3.0) but similar for the patients initiated on imatinib and nilotinib (2.51 and 2.50). The mean days until an outcome was identified was 347 (SD ± 311). 93 patients (43%) continued their index therapy; 95 patients (44%) discontinued their index therapy, 3 patients (1.4%) switched their index therapy, 7 patients (3.2%) died during follow up, and 18 patients (8.3%) disenrolled from the health plan. The majority of the patients were initiated on imatinib and 45% of these patients discontinued their index therapy per our study definition.

CONCLUSIONS: 44% of incident CML patients initiated on TKIs discontinued therapy by an average of 347 days (less than 1 year) from the start of treatment. Imatinib was the initial TKI therapy for 87.5% of the patients. Additional data will be presented which may assist in healthcare decision making for those newly diagnosed CML patients initiating first line TKI therapy in a managed care setting.

Figure 1: Summary Table of Outcomes from First Line Index TKI Therapy