Nilotinib Exerts Direct Effects on Vascular Endothelial Cells and May Act As a Co-Trigger of Atherosclerosis in Patients with Ph+ CML

Abstract ²⁷⁵³

The second-generation BCR/ABL tyrosine kinase inhibitor nilotinib is increasingly used for the treatment of patients (pts) with newly diagnosed Ph⁺ CML or imatinib-resistant CML. So far, nilotinib has been considered a well-tolerated drug that shows little if any relevant side effects. However, long-term results are not available yet. Known recurrent side effect of nilotinib include an increase in pancreatic enzymes (lipase, amylase) and an increase in fasting glucose levels. More recently, several centers have reported on the occurrence of progressive peripheral arterial occlusive disease (PAOD) in some of their pts treated with nilotinib. The exact relationship with drug intake and the potential mechanisms underlying PAOD development remain unknown. To address these questions, we have extended and updated our retrospective analysis and examined the in vitro effects of nilotinib. A total number of 34 pts (second line nilotinib, n=33; first line nilotinib, n=1; females, n=15; males, n=19, median age: 57.5 years) receiving nilotinib in our center were examined (observation period 2006–2011). Vascular events related to atherosclerosis were recorded in 9 pts (26.5%). Of these, 6 (17.6%) had PAOD, including 2 cases of severe PAOD requiring repeated surgical interventions or amputation (n=1). One patient had a myocardial infarction, one an acute coronary artery syndrome requiring bypass surgery, and one a spinal infarction. Although in all 9 pts, one or more pre-existing risk factors for atherosclerosis were identified, none of these pts had developed a vascular event during prior imatinib therapy. The JAK2 V617F mutant was not found in any of the 9 pts. Three pts died during the observation period, two from major bleeding episodes, and one from sudden death of unknown etiology. In 5 cases, an increase in blood glucose levels was found, and 1 patient developed overt diabetes mellitus. The frequency of PAOD development in age-, observation time-, and risk factor-matched control cohorts (CML pts receiving imatinib, lymphoproliferative neoplasms, JAK2 V617F+ myeloproliferative neoplasms, each n=34 pts) was significantly lower (<5%) than in the nilotinib group (p<0.05 by Chi Square test with Boneferroni correction). As assessed by ³H-thymidine uptake, nilotinib was found to inhibit the growth of the human microvascular endothelial cell line HMEC-1 (IC₅₀: 0.5–1.5 μM), whereas no substantial effect was seen with imatinib. Furthermore, nilotinib, but not imatinib, was found to inhibit umbilical vein endothelial cell (HUVEC) migration and sprouting in a scratch-wound healing assay. Finally, nilotinib but not imatinib, was found to block the growth of various human pancreatic cell lines, including PANC-1 and the insulin- producing cell line 1.4E7. Together, our data show that nilotinib exerts direct inhibitory effects on endothelial cells and pancreatic cells. Whether these drug effects contribute to vascular events seen in our nilotinib-treated pts remains at present unknown. In addition, the exact pro-atherogenic effect of nilotinib in second-line and first-line treated patients with CML remains unknown since no data from prospective clinical trials focusing on this issue, are available. Until these questions and safety-issues have been clarified, it may be advisable to take certain co-morbidities into account when considering the use of nilotinib in CML patients. In addition, the follow up of long-term treated patients should include parameters that can detect vascular events and predisposing co-morbidities as early as possible.