Abstract
Tyrosine kinase inhibitors (TKI) selective for Bcr-Abl, such as dasatinib, imatinib, and nilotinib have had remarkable success in the clinic, potentially shifting the prognosis of chronic myelogenous leukemia (CML) to a manageable chronic disease. With the increase in longevity of CML patients, there is rising concern of co-morbidities that may be influenced by chemotherapy (
Force et al., Nature Rev. 2007;7:332–340
). Recently, congestive heart failure (CHF) and direct cellular cardiotoxicity have been reported in CML patients on imatinib therapy (Kerkela et al., Nature Medicine 2006;12:908–916
). Ultrastructural mitochondrial abnormalities in cardiomyocytes were observed in CML patients with severe CHF and, interestingly, similar abnormalities were observed in cardiomyocytes of imatinib-treated mice, thus providing a prospective in vivo animal model for imatinib-induced cardiotoxicity. Furthermore, correlative findings of mitochondrial membrane potential loss, decreased cell viability, and increased apoptosis resulted from an array of cell-based assays in imatinib-treated primary rat cardiomyocytes, consequentially affording a supportive, if not predictive, in vitro cardiomyocyte toxicity model. Since imatinib-induced inhibition of the native form of c-Abl kinase was speculated to cause the observed cardiotoxicity and c-Abl is a shared target of dasatinib, imatinib, and nilotinib, the in vitro cardiotoxicity potential of dasatinib and nilotinib at pharmacologically relevant concentrations (0.09 μM and 5 μM, respectively) and up to 10-fold higher concentrations were compared side-by-side with imatinib in primary rat cardiomyocytes. Dasatinib did not significantly affect mitochondrial membrane potential, cell viability, apoptosis, or cellular ultrastructure in vitro, whereas imatinib significantly affected these parameters. Nilotinib at pharmacologically relevant concentration demonstrated decreased cell viability, but differed from imatinib in that mitochondrial membrane potential integrity was not affected under identical experimental conditions. Results suggest that at pharmacologically relevant concentrations, dasatinib does not induce cardiotoxicity, as does imatinib and nilotinib, and the molecular mechanisms of the observed cardiotoxicities may differ between imatinib and nilotinib. Of indirect relation, results from assessing another cardiovascular liability, namely hERG K+ channel blockade, demonstrated that dasatinib, imatinib and nilotinib differentially inhibited the hERG currents in vitro with IC50 of 14.3, 15.6 and 0.66 μM, respectively. These in vitro findings occurred at concentration levels approximately 150, 3 and 0.1-fold the expected human Cmax for the three TKIs, respectively. Thus, although TKI therapies may share similar targeting and clinical indications, differentiating specific toxicity profiles may be predictive of differences in potential clinical adversities.Author notes
Disclosure:Employment: All authors are employees of Bristol-Myers Squibb. Ownership Interests:; All authors hold stock/stock options in Bristol-Myers Squibb.
2007, The American Society of Hematology
2007
