Dec 7, 2006 - the first selective protein kinase inhibitors devel- oped for the treatment of chronic myelogenous leukemia (CML). The principal target of imatinib.
The
n e w e ng l a n d j o u r na l
of
m e dic i n e
clinical implications of basic research
Another Look at Imatinib Mesylate Klaus Strebhardt, Ph.D., and Axel Ullrich, Ph.D. Genetic abnormalities that occur during neoplastic transformation can cause the dysregulation of protein kinases — a critical event in tumorigenesis. The inhibition of protein kinases is one of the most impressive new approaches to targeted cancer therapy. Imatinib mesylate (Gleevec, Novartis; formerly known as STI571) was one of the first selective protein kinase inhibitors developed for the treatment of chronic myelogenous leukemia (CML). The principal target of imatinib is BCR-ABL, a fusion protein made up of part of the breakpoint cluster region (BCR) protein and part of the tyrosine kinase Abelson murine leukemia (ABL). Imatinib directly binds to the tyrosine kinase domain of BCR-ABL and inhibits its oncogenic activity in CML. (It is also effective against other tyrosine kinases, including ABL, KIT, and the platelet-derived growth factor receptor.) This orally administered drug has astonishing efficacy in CML, and clinical studies have not shown substantial toxic effects. Like the benefits of cytotoxic chemotherapies and other drug interventions, however, the benefits of treatment with imatinib are accompanied by adverse effects that must be managed to facilitate a patient’s adherence to therapy. Kerkelä et al.1 recently described a potential new adverse event: left ventricular dysfunction and congestive heart failure in 10 patients treated with imatinib; in these patients, the ejection fraction (an echocardiogram-based measurement of the heart’s pumping capacity) dropped significantly during therapy. Kerkelä and colleagues went on to examine the myocardial histologic features in patients treated with imatinib and in mice that received clinically relevant doses of this drug. In both cases, the authors observed membrane whorls in the sarcoplasmic reticulum and pronounced mitochondrial abnormalities of the heart tissue. These are early signs of toxin-induced myopathies, possibly due to dysregulated cellular energy homeostasis. Analyses of isolated cardiomyocytes from mice that had received imatinib showed low levels of cellular ATP and the collapse of the mitochondrin engl j med 355;23
al electrochemical gradient. Kerkelä et al. reported a modest activation of the apoptotic cascade and signs of necrosis. Several conditions of cellular stress, such as perturbation of calcium homeostasis or the redox status, can lead to the accumulation of misfolded proteins in the lumen of the endoplasmic reticulum, activation of the endoplasmic reticulum stress response, and, consequently, cell death (Fig. 1). The authors showed that imatinib causes stress in the endoplasmic reticulum, which in turn induces cardiomyocyte cell death, suggesting that the endoplasmic reticulum stress response is pivotal to the cardiotoxicity associated with imatinib. A previous study2 of rat pancreatic cells has shown that chronic stress is accompanied by an activation of the Jun N-terminal kinase (JNK) signaling pathways, which leads to cell death. Kerkelä et al. showed that they could inhibit the imatinibinduced collapse of the membrane potential in mitochondria (and thus obviate cell death) by repressing JNK activity. The other effectors of the stress response, however, remained unchanged, suggesting that JNK signaling is a consequence of stress. Kerkelä and colleagues also tested whether the observed toxicity is mediated by the inhibition of known imatinib targets (ABL, KIT, or the platelet-derived growth factor receptor). Transfection with an imatinib-resistant mutant of c-ABL prevented the release of cytochrome c and rescued cells from imatinib-induced cell death; these findings suggest a novel and vital role of c-ABL in cardiomyocytes. How ABL fulfills this role is not clear; perhaps it does so by “turning down the volume” of the stress effector pathways (Fig. 1). Targeting the array of kinases in cancer is a rapidly expanding strategy for developing drugs for the treatment of cancer. The potential and specificity of small-molecule compounds directed against protein kinases are typically evaluated by means of in vitro testing with kinase panels, which usually include only a subgroup of the entire human kinase complement (518 enzymes).
www.nejm.org
december 7, 2006
The New England Journal of Medicine Downloaded from nejm.org at UNIVERSITAETSBIBLIOTHEK on January 23, 2014. For personal use only. No other uses without permission. Copyright © 2006 Massachusetts Medical Society. All rights reserved.
2481
Clinical Implications of Basic Research
Imatinib mesylate
Cell membrane
c-ABL
Stress in endoplasmic reticulum
IRE1
Dimerization
Membrane of endoplasmic reticulum
Imatinib mesylate
BAX
Activation of JNK pathway
Ultrastructural changes, collapse of membrane potential
Cell death
Figure 1. Imatinib Mesylate and Signaling. A recent study by Kerkelä et al.1 indicates an association between imatinib mesylate and congestive heart failure and provides insight into the mechanisms underlying this association. Stress on the cell, such as energy depletion, leads to the accumulation of unfolded proteins in the endoplasmic reticulum that, in turn, activates stress-response mechanisms and apoptosis. Kerkelä et al. have shown that imatinib may trigger the death of cardiomyocytes by repressing the activity of c-ABL. When unfettered, c-ABL mutes stress in the endoplasmic reticulum; thus, imatinib would seem to augment this stress (and, ultimately, apoptosis and necrosis) by removing the “braking” effect of c-ABL. They also have shown that imatinib may act at a second point, downstream of c-ABL. When unfolded proteins accumulate in the lumen of the endoplasmic reticulum, IRE1 dimerizes, leading to Jun N-terminal kinase (JNK) activation and hence the translocation of BAX (a proapoptotic protein) to the mitochondria. The mitochondrial membrane subsequently collapses, and cytochrome c is released, causing the cell to undergo apoptosis. Kerkelä et al. have shown that imatinib may mimic stress by activating JNK signaling.
Secondary modifications and different conformations that affect kinase activity in vivo make it more difficult, if not impossible, to interpret the relevance of in vitro assays of function. Proteomic strategies are therefore attractive. One such strategy provides a “readout” of the entire protein com-
2482
n engl j med 355;23
plement of the cell once it has bound a candidate small-molecule drug. Using this approach, investigators have shown that selective inhibitors of receptor tyrosine kinases involved in tumor vascularization also inhibit kinases that mediate other processes such as control of the cell cycle.3 A reevaluation of the considerable data available for small-molecule inhibitors such as imatinib is necessary. Also underscoring the wisdom of a reevaluation are the observations that c-ABL mediates the tumor-suppressor effects of the EphB4 receptor in breast cancer cells and that imatinib impairs this cellular defense.4 The potential of imatinib to promote epithelial tumor progression and to induce heart failure should not be ignored in future clinical trials. The way in which candidate drugs are tested before they are assessed in clinical trials also warrants reevaluation. In spite of our vastly expanded understanding of the function and physiological relevance of every member of the kinase family, we are just beginning to comprehend their importance at the level of systems biology. It is therefore critical that different approaches are stringently and systematically applied in the preclinical workup of candidate drugs. These approaches include long-term evaluations of new drugs at clinically relevant doses in valid animal models of cancer. Dr. Ullrich reports receiving lecture fees from Pfizer. No other potential conflict of interest relevant to this article was reported. We thank Yves Matthess for his assistance. From the Department of Obstetrics and Gynecology, School of Medicine, J.W. Goethe University, Frankfurt, Germany (K.S.); the Singapore Oncogenome Laboratory, Centre of Molecular Medicine, Institute of Molecular and Cell Biology, Proteos, Singapore (A.U.); and the Department of Molecular Biology, Max Planck Institute of Biochemistry, Martinsried, Germany (A.U.). 1. Kerkelä R, Grazette L, Yacobi R, et al. Cardiotoxicity of the cancer therapeutic agent imatinib mesylate. Nat Med 2006;12: 908-16. 2. Urano F, Wang X, Bertolotti A, et al. Coupling of stress in the ER to activation of JNK protein kinases by transmembrane protein kinase IRE1. Science 2000;287:664-6. 3. Godl K, Wissing J, Kurtenbach A, et al. An efficient proteomics method to identify the cellular targets of protein kinase inhibitors. Proc Natl Acad Sci U S A 2003;100:15434-9. 4. Noren NK, Foos G, Hauser CA, Pasquale EB. The EphB4 receptor suppresses breast cancer cell tumorigenicity through an Abl-Crk pathway. Nat Cell Biol 2006;8:815-25. Copyright © 2006 Massachusetts Medical Society.
www.nejm.org
december 7, 2006
The New England Journal of Medicine Downloaded from nejm.org at UNIVERSITAETSBIBLIOTHEK on January 23, 2014. For personal use only. No other uses without permission. Copyright © 2006 Massachusetts Medical Society. All rights reserved.
