Real-Life Comparison Of Severe Vascular Events and Other Non-Hematological Complications In CML Patients Treated With Second Line Nilotinib Or Dasatinib

Despite very good long-term results of imatinib (IM) in chronic myeloid leukemia (CML) patients about 30% of them will need the new treatment option. For these patients, two effective second-generation tyrosine kinase inhibitors (TKI) are available: dasatinib and nilotinib. Although both have good toxicity profile and side effects are usually mild and manageable, some patients have major problems and develop intolerance or even life-threatening adverse events (AEs). A unique spectrum of non-hematological AEs has been reported for both TKI like pleural infusion for dasatinib and elevation of glucose, pancreatic and liver enzymes for nilotinib. Recently several studies have described the clinical manifestation of peripheral artery occlusive disease (PAOD) in patients receiving nilotinib. The risk of developing PAOD on TKI therapy is unknown and causality has not been established yet.

We retrospectively analyzed the rates of grade 2-5 non-hematologic AEs in CML patients treated with dasatinib or nilotinib as second line therapy in “real-life” setting. The analysis included 105 patients from 5 Polish Hematological Centers, 57 men (54%) and 48 women (46%), with median age of 57 (range 21-88). Median time of observation was 28 months (range 1-93 months). Fifty-five (52%) patients were treated with nilotinib and 50 (48%) with dasatinib. The reason for stopping first line therapy was resistance to IM in 84% and intolerance in 16% (the proportions were similar in dasatinib and nilotinib group, p=0.77). At the time of the present analysis 73 patients (68%) continued second line therapy while in 28 patients (32%) therapy was changed because of resistance or intolerance. Estimated median time for second line therapy withdrawal duration was 35 months. Interestingly, estimated median time to treatment change was significantly longer for dasatinib (42 months) than for nilotinib (29 months) (log-rank test p=0.17). The non-hematological AEs (grade 2-4) were observed in 22 (40%) patients treated with nilotinib and 21 (42%) patients in dasatinib arm (p=0.83). The median estimated time to AE onset was 70 months for nilotinib and 53 months for dasatinib (log-rank test, p=0.35). Vascular events occurred in 8 patients (8%) in both treatment groups, included 6 (11%) in nilotinib group and 2 (4%) in dasatinib group (p=0.16). In nilotinib group one patient with previously diagnosed hypertension developed PAOD which required stent implantation, 2 patients had myocardial infarction, 2 patients had arrythmia and 1 had ischemic stroke. In dasatinib gruoup one patients had myocardial infarction and 1 ischemic stroke. Among other clinically significant AEs pleural effusion (grade 2-4) occurred more often in dasatinib group than in nilotinib group (26% vs. 2%) (p=0.003). Moreover, in dasatinib group two patients (4%) developed pulmonary hypertension, in both cases completely reversible after stopping the drug. In one of these patients sclerodermia was diagnosed before dasatinib treatment. In conclusion we found that despite generalized opinion on good toxicity profile of second generation TKI about 40% of patients treated with second line nilotinib or dasatinib outside clinical studies suffered from grade 2-4 non-hematological AE. It is worth underlining that most AEs occurred late, after more than 2 years of treatment. Importantly, the total frequency of different AEs did not differ significantly and the number of severe vascular events seems comparable on nilotinib and dasatinib, though larger studies may be needed.