Longitudinal, prospective cardiovascular and metabolic risk in treatment-naive patients with chronic myeloid leukemia in chronic phase (CML-CP) starting tyrosine kinase inhibitor (TKI) therapy in a real-world setting

7053 Background: Multiple TKIs used in the treatment (tx) of CML have been associated with the development of cardiovascular/metabolic (CV/Met) comorbidities, yet no study to date has prospectively and systematically assessed CV/Met parameters after TKI initiation. CA180-653 (NCT03045120) is a non-interventional, prospective, real-world study characterizing the effects of TKIs (dasatinib [DAS], imatinib [IM], nilotinib [NIL], and bosutinib [BOS]) on CV/Met risk factors in patients (pts) with CML-CP in routine care. Baseline CV/Met risks were previously reported (Mauro, GCOS 2019). Here, we report data from the final 2-year follow-up. Methods: Adult pts with newly diagnosed CML-CP due to start tx with DAS, IM, NIL, or BOS were enrolled; the BOS cohort was not analyzed independently due to small sample size. Pts were followed for 24 mo. Blood/urine samples, CV/Met assessments, coronary calcium scoring, echocardiography, and cytogenetic/molecular response assessments were collected at routine office visits at predetermined timepoints. The primary endpoint was change in CV/Met risk from baseline. Results: At the end of follow-up, 118 pts were enrolled; 49 received DAS, 47 IM, 16 NIL, and 6 BOS. Of these, 83 (70.3%) pts completed the study and 35 (29.7%) discontinued; the most common reason for discontinuation was withdrawal of consent (n = 19). No pts died from CML. At baseline, median age was 57 (range, 20–97) y; 59.3% of pts were male and 83.9% were White. Sixty-two (52.5%) pts had CV/Met risk factors at baseline, including hypertension (n = 45/62; 72.6%), hyperlipidemia (n = 33/62; 53.2%), diabetes (n = 19/62; 30.6%), and arrhythmia (n = 10/62; 16.1%). Over the study period, 1 (3.4%) patient receiving DAS and 2 (6.9%) receiving IM developed coronary artery calcification; 2 (9.1%) pts receiving IM developed pericardial effusion. Median left ventricular ejection fraction on 2D echocardiogram was normal for all TKIs at baseline (65.4% [range, 27–81]) and end of study (65.4% [range, 40–80]). Systolic and diastolic blood pressure increased from baseline only in pts who received NIL. Blood glucose levels increased from baseline for DAS, IM, and NIL (median: 1.9, 4.9, and 5.8 mg/dL, respectively), and low-density lipoprotein levels decreased from baseline in all 3 cohorts (median: 72.7, 52.1, and 56.3 U/L, respectively); data from the NIL group were limited by the small sample size. There was 1 death from myocardial infarction related to NIL. Conclusions: These findings suggest more than half of pts with CML-CP have CV/Met risk factors at baseline. At end of study, there was indication of CV/Met effects based on TKI selection. This highlights the importance of appropriate TKI selection based on pt comorbidity characteristics, along with close monitoring and management of CV/Met risks throughout tx. Clinical trial information: NCT03045120 .