Identification Of Patients (pts) With Chronic Myeloid Leukemia (CML) At High Risk Of Artery Occlusive Events (AOE) During Treatment With The 2nd Generation Tyrosine Kinase Inhibitor (TKI) Nilotinib, Using Risk Stratification For Cardiovascular Diseases (CVD)

Abstract

Background Nilotinib is approved for use in pts with CML after failure of imatinib and in newly diagnosed CP-CML. However, several studies report a nilotinib-associated risk of AOE, especially in pts with preexisting risk factors for CVD. Since CVD are a major cause of disability and death worldwide and result from complex interactions between multiple risk factors, we aimed at determining whether CVD risk estimation using the 2012 European Society of Cardiology (ESC) classification could be useful to identify patients at high risk of AOE during nilotinib therapy. Methods Pts (n=75) treated with nilotinib upfront or after failure of prior TKI at our institution were included provided that baseline risk factors for CVD and prior history of established CVD could be retrospectively collected. Risk factors included age, tobacco use, diabetes mellitus (DM), arterial hypertension (AH), dyslipidemia and increased body mass index (BMI). Patients were categorized into 2 groups according to ESC 2012 classification: low/moderate (L/M) and high/very high (H/VH) CVD risk. H/VH included pts with any of the following: established CVD, DM, severe AH, familial dysplipidemia or a SCORE (systematic coronary risk evaluation project) ≥5%. Results At nilotinib initiation, median age was 51 years (19-76), 41 pts (54.7%) were males. Nilotinib was given upfront in 28 pts (37%) and after failure of imatinib or dasatinib following imatinib in 47 pts (63%). Median time from diagnosis was 1 month (0.3-4) in the former group and 25 months (2-130) in the latter. Median duration of TKI exposure prior to nilotinib in the latter group was 22 months (0.4-91). Initial nilotinib dosing regimen was 400mg BID in 42 pts (56%) and 300mg BID in 33 pts (44%). Median duration of nilotinib treatment of 28 months (3-76) and median follow-up was 30 months (3-77). At baseline, medical history revealed H/VH risk category in 15 pts (20%) including established CVD in 6 pts (8%) (all diagnosed before CML), DM in 10 pts (13.3%), severe AH in 1 pt (1.3%), familial dyslipidemia in 1 pt (1.3%) and a SCORE ≥5% in 2 pts (2.6%). Median number of CVD risk factors was 1 (0-6) including age ≥ 45 years in men and ≥ 55 years in women in 37 pts (49.3%), active smoking or ceased during the previous year in 12 pts (16.7%), DM in 10 pts (13.3%), AH in 14 pts (18.7%), dyslipidemia in 10 pts (13.3%) and BMI ≥ 25 kg/m2in 20 pts (38.6). Nilotinib was discontinued in 23 pts (30.7%) due to adverse events (10 pts), resistance (7 pts), TKI discontinuation study (4 pts) or death (2 pts). AOE occurred in 12 pts after a median duration of nilotinib treatment of 24 months (9-47). AOE included myocardial infarction (MI) or coronary heart disease (CHD) (n=3), cerebrovascular events (CeVD) (n=3) and peripheral artery disease (PAD) (n=6). Overall, the cumulative incidence of AOE was 2.91% (95% CI: 0.74-11.42) by 12 months, 9.81% (95% CI: 4.57-21.08) by 24 months, 19.29 (95% CI: 10.77-34.56) by 36 months and 27.7% (95% CI: 16.2-47.35) by 48 months (discontinuation of nilotinib and death as competing risks). Cumulative incidence of AOE by 48 months was 72.22% (95% CI: 47.46-100) in the H/VH group and only 12.13% (95% CI: 4.32-34.08) in the L/M group. Log Rank comparison of Kaplan Meier analysis of 48-month survival without AOE showed a significant difference between the 2 groups (27.78% (95% CI: 0-58.9) versus 84.38% (95% CI: 67.04-100) p=0.0001). Sensitivity of the ESC classification in nilotinib-treated patients was 67% and specificity 89%. It is important to note that all pts with a history of AOE had recurrent AOE on nilotinib. Nilotinib was discontinued in 11 pts with AOE after a median time of 288 days (1-688), 7 pts had recurrent or worsening AOE before nilotinib discontinuation and 2 pts died from AOE. Conclusions In our retrospective study, CVD risk estimation according to the 2012 ESC classification reveals that pts who belong to the H/VH risk group at baseline are at very high risk of AOE during nilotinib therapy. These findings now need to be validated in a prospective fashion. Nevertheless, we readily recommend that assessment of CVD risk should be performed in all pts considered for nilotinib therapy. Alternative TKI may be chosen whenever possible in pts at H/VH risk of CVD. In those treated with nilotinib, CVD risk should be reassessed throughout therapy and risk factors should be tightly controlled according to current guidelines. Disclosures: Rea: BMS: Honoraria; Novartis: Honoraria; Pfizer: Honoraria; ARIAD: Honoraria; Teva: Honoraria. Messas:Novartis: Honoraria.