click to enlarge Figure 1. Human cardiac myocytes were treated with a variety of targeted therapies in oncology followed by isolation of total cell lysates and Western blot analysis for phosphorylation of ACC, an indicator of AMPK pathway activation and cessation of fatty acid synthesis.

There is considerable interest in drug safety among U.S. stakeholders, including the Food and Drug Administration (FDA), pharmaceutical companies, patient advocacy groups, prescribing physicians and their patients.¹ While awareness that pharmacotherapy carries a degree of risk is not new, evaluations of drug safety are now more rigorous and comprehensive. Such assessments are made throughout the drug development lifecycle. Accordingly, evaluation of cardiac safety biomarkers and hard clinical endpoints are both important components of mitigating risk of drug cardiotoxicity. However, routine implementation of a non-clinical cardiac toxicity testing program far earlier in the drug development process can identify candidates that pose a higher than acceptable risk for the patient population which, if kept from advancement, could prevent considerable losses in time and money, and maintain a positive reputation for the pharmaceutical company, all while minimizing exposure of the patients enrolled into clinical trial from potentially unsafe drugs.

Regulated drug cardiotoxicity testing
Cardiotoxicity testing has become a central component of drug development, and one that is considered right from the start of the lifecycle development process. Various noninvasive methodologies have been used to monitor cardiotoxicity, including radionuclide ventriculography, electrocardiography (ECG), and stress myocardial perfusion imaging. While such identification and monitoring can be a great help in treatment of these cardiovascular conditions, it would clearly be preferable to prevent—or at least reduce—their occurrence by identifying cardiotoxic potential of the drug prior to patient treatment; ideally in the preclinical development setting. However, current preclinical approaches to cardiac safety testing perhaps focus too much on ion channel testing (e.g. hERG) and need to broaden the in vitro test menu to assess other cellular functions that are critical to cardiac cell health, particularly in the oncology drug arena.

Anti-cancer targeted therapies

With the advent of targeted therapy for cancer patient treatment, options are available that not only target the oncogenic drivers, but also minimize the toxicities often brought about with traditional radiation or chemotherapies. However, these targeted strategies are not wholly free from secondary toxicities. The therapeutic molecules prescribed may be designed to inhibit a target that is also necessary for the health and function of normal tissues and organs. Of relevance here is that cancer cells and cardiomyocytes share signaling pathways, such that a drug leading to a beneficial intervention in cancer cells can have deleterious effects on cardiomyocytes.²Therefore, the negative effects these drugs have on cardiac cell health can exist far beyond the interference of ion channel blockage and inhibition.

Small-molecule inhibitors refer to compounds that are designed to enter the cell and disrupt a specific cellular process for which the drug is designed. The process usually involves the inhibition of a disease-promoting tyrosine kinase enzyme within the target pathway. However, studies have shown that regardless of the kinases the TKI is designed to inhibit, there are off-target effects on additional kinases.³ As proper health and function of the cardiac cell is dependent on a variety of enzyme-driven signal transduction pathways, many of which are shared with actively growing tumor cells (e.g. HER2 and C-Abl), a comprehensive screening process is crucial to identify additional kinases the small molecule inhibits. This information is valuable in lead optimization, as well as providing a framework to the mechanisms of how the drug can affect both the heart and the tumor.

A closer look at AMPK 
AMPK is a metabolic sensor of cellular ATP and controls fatty acid oxidation and glucose uptake in skeletal muscle, heart, and liver.⁴ The AMPK enzyme has gained recent attention in the context of sunitinib-induced inhibition of AMPK and cardiotoxicity.⁵ A recent study also showed that targeted therapy can indirectly modulate AMPK activity.⁶ It was shown that a small molecule designed to inhibit HER2 and EGFR had also mobilized calcium within the cell, resulting in the activation of AMPK through the CAMKK pathway.⁶ What was profound from this study is that the activation of AMPK was also observed in treated human cardiac myocytes (HCMs), which resulted in induction of fatty acid beta-oxidation that protected the cardiac cells from TNF(alpha)-induced reduction of cellular ATP and cell death.⁶ In an internal pilot study that surveyed several approved targeted therapies in oncology, it was demonstrated that the modulation of the AMPK pathway is neither a class effect nor can it be predicted based solely on the known targets of the drugs (Table 1 and Figure 1).

These results established that certain targeted therapies can unexpectedly modulate the activity of the AMPK pathway. In the case of AMPK inhibition, the molecule can expose the heart to cardiotoxic risk; particularly in the context of cellular stresses that often accompany chemotherapy or stem from cardiac co-morbidities such as high blood pressure, coronary artery disease, or diabetes. As AMPK activators gain attention in the context of tumor growth inhibition, such as the diabetic drug metformin,⁷ targeted therapies that activate the AMPK pathway may not only provide protection to the heart, but may help to restrict the growth potential of the tumor.^6,8 

click to enlarge Figure 2. Primary fetal HCMs were treated for 48 hours with DMSO vehicle (left), sunitinib (middle), or erlotinib (right) followed by lipid staining by Oil Red-O. Sunitinib treatment induces lipid accumulation in cardiac cells. Bright-field pictures of stained cells and tissues were taken at 40X and 100X (inset).

Targeted therapy-induced cardiotoxicity
Trastuzumab (Herceptin) is a monoclonal antibody that targets the human epidermal growth factor receptor tyrosine kinase HER2/ErbB2, and has shown a significant antitumor effect for patients with HER2-positive breast cancer.⁹ However, inhibition of HER2, a receptor required for healthy heart function,¹⁰ by trastuzumab has been shown to cause a high incidence of congestive heart failure (CHF)¹¹ when combined with doxorubicin, a regular component of breast cancer therapy. Doxorubicin undergoes electron reductive activation that results in the accumulation of reactive oxygen species (ROS) that in turn causes cellular stress and damage.¹² However, patients affected by trastuzumab-related cardiotoxicity do not exhibit the cellular death and ultrastructural changes associated anthracyclins. As it has been shown that development and function of the heart requires HER2, some believe that inhibition of HER2 by trastuzumab or other HER2 inhibiting compounds may lead to a HER2-inhibitor class effect of cardiotoxicity by preventing the heart from mounting a proper stress response; particularly in the context of anthracycline-induced ROS production. This point can also be particularly important in the context of cardiac-related comorbidities, but results of clinical trials designed to assess this hypothesis are ongoing. Because there has not been substantial irreversible cardiotoxicity identified in breast cancer patients treated with the HER2-inhibiting small molecule lapatinib (Tykerb),¹³ it is believed that some of the cardiotoxicity associated with trastuzumab may be associated with an immunological response mounted by the inhibitory antibody binding the HER2-expressing cardiac cells.

Sunitinib (Sutent) is a multi-kinase inhibitor that was designed to inhibit the angiogenic processes of vascular tumors through the inhibition of VEGF and PDGF receptors.¹⁴ However, increased occurrence of cardiotoxicity, including left ventricular dysfunction and CHF5  have lead sunitinib to be increasingly scrutinized as a therapy choice. Kinome analysis has suggested that sunitinib inhibits AMPK.³ This observation has been corroborated in the treatment of human cardiac myocytes (HCMs) with sunitinib.⁸ Sunitinib-treated HCMs also dramatically increased intercellular lipid accumulation in both primary HCMs in vitro⁸ (Figure 2) and Sprague Dawley rat hearts in vivo, decreased mitochondrial membrane potential (Figure 3), and increased ROS. All of these observations contribute to a reduction in the energy generating capacity of the heart; a mechanism that is crucial for supplying the metabolically active organ with ATP.

Non-clinical safety evaluation for cardiotoxicity 

click to enlarge Figure 3. Primary fetal human cardiac myocytes were treated with DMSO vehicle (green), sunitinib (red), or erlotinib (blue) for 24 hours followed by analysis of mitochondria membrane potential changes by flow cytometry. A decrease in mitochondrial polarization is associated with de-coupling of the electron transport chain responsible for the generation of ATP and apoptosis (program-mediated cell death). An increase in mitochondrial polarization has been observed in the context of a high energy state of the cell (high ATP). Treatment of human cardiac myocytes with sunitinib causes massive depolarization of the mitochondria.

A viable non-clinical safety program would evaluate new and existing drugs in a preclinical setting by a prospective screening approach to identify whether a drug molecule has the potential to display unacceptable cardiotoxic risk.

With a growing list of pharmaceutical compounds that are under scrutiny and investigation for causing prolonged QT interval, left ventricular dysfunction, and CHF—such as sunitinib and trastuzumab when combined with anthracyclines—it is critical to screen a library of compounds in the early phases of development in order to choose the appropriate candidate for Phase 1 and Phase 2 clinical trials. This approach can also be leveraged for compounds that are well into their development program to identify and potentially ameliorate negative influences the drug may have on the heart.

The central components of a non-clinical cardiac safety program are often conducted using cell-based assays on drug-treated primary fetal HCMs, neonatal rat ventricular myocytes, or immortalized cardiac cell lines.  However, these cell types are wrought with weaknesses including fibroblast contamination, stemming from non-human sources, and the lack of electrophysical characteristics of real cardiac myocyte health and function.  Fortunately, stem-cell technologies now offer induced pluripotent stem cell-derived (iPSC) cardiac myocytes that are highly pure, functionally relevant (exhibit electrical profiles in culture and are amenable to patch-clamp-like studies that monitor electrical potentials and voltage-gated ion channel function), and are human in origin. The following would comprise an effective preclinical cardiac safety testing program that utilizes iPSC-derived cardiac myocytes:

• Determining influences on key cardiac metabolic pathways focusing on AMPK;
• Evaluating changes in fatty acid beta-oxidation;
• Measuring changes in mitochondrial health , ROS production, and ATP levels;
• Assessing drug-induced apoptosis;
• Survey potential off-target effects using a comprehensive kinase profiling platform.

Additional assays to consider are assessment of cellular hypertrophy, gene expression profiling for oxidative stress and hypertrophy, non-invasive long-term monitoring of adverse events using impedance technology, and evaluation of ion channel function by multi-electrode array (MEA).

Summary
Prospective identification and potential amelioration of cardiotoxicity is a critical component of contemporary drug development, particularly for targeted therapies in oncology that are designed to inhibit critical signaling pathways shared by both the tumor cell and the cardiac myocyte. Accordingly, effective non-clinical cardiotoxicity screening programs need to be implemented earlier in the development process to provide insight into potential cardiotoxic compounds.  Additionally, the program should identify pharmaceutical drugs that demonstrate cardio-protective effects with regard to mitochondrial health and energy homeostasis.

About the Authors
Scott Shell directs and manages both the sponsored and internal pre-clinical R&D projects that result in identification and validation of biomarkers best suited for specific oncology drug development programs.

Robert Wappel is the main technical expert for the CardioCheck program.  His responsibilities include the end-to-end operations of CardioCheck, data analysis, and result reporting. 

Rick Turner leads the Interdisciplinary Cardiac Safety Services Team at Quintiles, which provides consultation and guidance for sponsors on study design issues, operational efficiencies, and regulatory submissions. His business travel schedule enables him to meet and keep close contacts with members of regulatory agencies worldwide.  He is a regular presenter at international cardiac & cardiovascular safety meetings.  He writes a “Cardiovascular Safety Watch” column for the Journal for Clinical Studies and publishes regularly in peer-reviewed journals. 

Sarah Bacus is responsible for overseeing all of the preclinical R&D projects in understanding new targeted therapy mechanism of action and proof of concept.  She originated the CardioCheck program through a paper in the Proceedings of the National Academy of Science. 

References
1. Turner J. Drug safety, medication safety, patient safety: An overview of recent FDA initiatives and guidances. Regulatory Rapporteur. 2009: 4-8.
2. Force T,  Krause DS,  Van Etten RA. Molecular mechanisms of cardiotoxicity of tyrosine kinase inhibition. Nat Rev Cancer. 2007;7(5):  332-44.
3. Karaman MW, et al. A quantitative analysis of kinase inhibitor selectivity. Nat Biotechnol.2008; 26(1): 127-32.
4. Hardie DG. AMP-activated/SNF1 protein kinases: conserved guardians of cellular energy. Nat Rev Mol Cell Biol.2007;8(10): 774-85.
5. Chu TF, et al. Cardiotoxicity associated with tyrosine kinase inhibitor sunitinib. Lancet. 2007;370(9604): 2011-9.
6. Spector  NL, et al. Activation of AMP-activated protein kinase by human EGF receptor 2/EGF receptor tyrosine kinase inhibitor protects cardiac cells. Proc Natl Acad Sci USA.2007; 104(25): 10607-12.
7. Ben Sahra I, et al, Metformin in cancer therapy: a new perspective for an old antidiabetic drug? Mol Cancer Ther.2010;9(5): 1092-9.
8. Shell SA, et al. Activation of AMPK is necessary for killing cancer cells and sparing cardiac cells. Cell Cycle. 2008;7(12): 1769-75.
9. Ewer SM,  Ewer MS. Cardiotoxicity profile of trastuzumab. Drug Saf. 2008;31(6): 459-67.
10. Crone SA, et al. ErbB2 is essential in the prevention of dilated cardiomyopathy. Nat Med. 2002;8(5): 459-65.
11. Slamon  DJ, et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med. 2001;344(11): 783-92.
12. Minotti G, et al, Anthracyclines: molecular advances and pharmacologic developments in antitumor activity and cardiotoxicity. Pharmacol Rev. 2004;56(2):185-229.
13. Perez EA, et al. Cardiac safety of lapatinib: pooled analysis of 3689 patients enrolled in clinical trials. Mayo Clin Proc. 2008;83(6): 679-86.
14. Demetri GD, et al. Efficacy and safety of sunitinib in patients with advanced gastrointestinal stromal tumour after failure of imatinib: a randomised controlled trial. Lancet. 2006;368(9544):1329-38.