Preclinical Models of Cancer Therapy-Associated Cardiovascular Toxicity

The progressive increase in our aging population has led to a significant rise in the prevalence of both cancer and heart disease. While these two disciplines have largely existed in distinct clinical domains, there is growing recognition that cardiac risk factors and comorbidities may impact the choices and outcomes of cancer treatment, while cancer therapies may also have long-lasting effects on cardiovascular health. To facilitate the communication and coordination of clinical care, the establishment of guidelines for best practice, as well as the development of new approaches to diagnose and treat cancer patients with cardiovascular complications or underlying cardiovascular diseases, the new field of Cardio-Oncology has recently emerged and is rapidly becoming established as a new sub-specialty within cardiovascular medicine.

As it is an interdisciplinary field, the knowledge base required to be a proficient provider in Cardio-Oncology is vast. Similarly, the understanding of the mechanisms behind cardiovascular complications of cancer treatment requires extensive training in basic and translational research. To address the need for a comprehensive yet compact resource that allows for basic and translational investigators to review the advantages and disadvantages of various model systems available to study the mechanistic basis of cardiotoxicity due to cancer therapy, the Cardio-Oncology Subcommittee of the Council on Genomic and Precision Medicine along with the Basic Cardiovascular Sciences Council of the American Heart Association has recently published a scientific statement to address this knowledge gap.¹ This document serves to illustrate both the breadth of the investigative tools available at our disposal as well as the challenges that investigators often encounter when employing these models. Given the pace of progress in this rapidly moving field, this document will certainly require frequent updates in the coming years. Yet, the illustration of the methodology underlying various experimental model systems to address fundamental questions in Cardio-Oncology should be extremely helpful for young investigators starting their research careers as well as established investigators who are new to the field.

From the perspective of an experimental biologist, this statement provides discussions on model systems ranging from isolated cells in culture—including differentiated cardiovascular cells from human induced pluripotent stem cells (iPSCs), engineered 3-dimensional micro-tissues and organoids—to animal models from rodent and non-human primates. For example, earlier studies of anthracycline-induced cardiomyopathy have relied on cultured cardiomyocytes such as neonatal rat ventricular myocytes to identify mechanisms of cardiotoxicity. More recently, human iPSC-derived cardiomyocyte studies have provided additional insights on patient-specific susceptibility for doxorubicin-induced cardiotoxicity² as well as the development of a cardiotoxicity score for tyrosine kinase inhibitors³. In recognition of the heterogenous cell types and complex environment within a functioning organ, the creation of engineered cardiac micro-tissue and organoid models have gained great interests in recent years. However, beyond simple immortalized cell lines, most of the in vitro model systems described have not been adopted by the FDA for toxicology testing or drug approval process. The most familiar but expensive and time-consuming assays to perform are in vivo animal models for cardiotoxicity. Species ranging from small animals such as zebrafish and rodents to large animals such as swine and non-human primates have been used as models in published studies of cardiotoxicity from cancer therapeutics. The advantages of the in vivo system are the incorporation of drug absorption, distribution, metabolism, and excretion effects that cannot be easily modeled by cells or engineered tissues and the ability to measure cardiac and vascular function within the appropriate hemodynamic context of a beating heart. However, notable species-specific differences between animals and humans must be recognized and appropriately considered when making conclusions regarding the potential involvement of specific molecular pathways in disease pathogenesis.

In addition to discussing the model systems available to study cardiotoxic effects, the statement also addresses the reported mechanisms of toxicity for various major classes of cancer therapeutics. The classically described anthracycline cardiotoxicity mechanisms affecting topoisomerase II and its role in the repair of double-stranded DNA breaks or the role of doxorubicin on mitochondrial reactive oxygen species production have since been expanded greatly to include the effects of doxorubicin on mitochondrial biogenesis, inhibition of PI3Kinase-gamma and autophagy, as well as activation of aryl hydrocarbon receptor and the induction of cellular apoptosis. Unlike the anthracyclines where cardiotoxic mechanisms are largely off-target effects from tumor cell cytotoxicity, toxicity from kinase inhibitors can often be on-target effects due to shared requirements of the same cell signaling pathway on cardiomyocyte survival and tumor growth promotion. The best example of this can be found in trastuzumab cardiotoxicity, where the target cell surface receptor HER2/Neu/Erbb2 is known to be required for both adult cardiomyocyte homeostasis and survival as well as breast cancer cell proliferation. However, the distinction between on-target vs off-target effects of the newer classes of kinase inhibitors is less clear due to their cross-reactivity with multiple signaling pathway receptors (e.g., sunitinib and sorafenib on vascular endothelial growth factor receptor and platelet-derived growth factor receptor). Hence, it is important to select the appropriate cell culture and engineered tissue model that contains the types of cells expressing relevant levels of the growth factor receptor and the right cell phenotype for the study endpoint. Finally, with the recent explosion in the use of immunotherapies in cancer, the toxic side effects of immune checkpoint inhibitors and chimera-antigen receptor-T cell infusion have been increasingly recognized.⁴ While animal models of cardiovascular toxicities for these drugs are still being developed, it is encouraging that new tools for profiling immune cell repertoire such as single cell time-of-flight mass spectrometry (CyTOF) and single cell RNA sequencing can be applied directly to assess for the presence of toxic immune cells from peripheral blood and biopsy tissues of patients.

As we gain a greater understanding of the mechanisms of toxicity of cancer therapeutics on the cardiovascular system and identify potential counter measures to mitigate the damaging effects of drug toxicities, an exciting frontier in Cardio-Oncology has merged that suggests the presence of shared risks for the development of both cancer and cardiovascular disease. The discovery of clonal hematopoiesis of indeterminate potential associated with not only an increased risk for malignancies in hematopoietic cells but also cardiovascular mortality highlights the dependence of both disciplines on the effects of immune cell populations, where on one hand the mutation in a chromatin modifying gene leads to growth advantage in immune cells, while on the other hand, the population expansion of immune cells (or a subset of) may modulate inflammatory events at the vascular wall to increase plaque instability.⁵ Similarly, the elevation of an oncometabolite D-2-hydroxyglutarate has been found to inhibit alpha-ketoglutarate dehydrogenase and inhibit cardiac contractile function in rodent models.⁶ Intriguingly, patients with heart failure may experience stimulated tumor growth by altering the growth factor exposure of cancer cells. To be sure, many more details remain to be uncovered to fill the large knowledge gap that currently exists in our understanding of the shared risks between cancer and cardiovascular disease. Yet, with the growing interest in Cardio-Oncology and the infusion of a new generation of bright young investigators in this area, the future of this new field is very bright.

In summary, the AHA Scientific Statement on Preclinical Models of Cancer Therapy-Associated Cardiovascular Toxicity by Asnani et al provides a compact yet comprehensive look at the mechanisms and models available for understanding cardiovascular toxicities due to cancer therapeutics. While many challenges such as the development of the most appropriate animal models for studying permissive cardiotoxicity and the need to increase the throughput and reduce the effort and cost of animal experiments remain to be addressed, there is tremendous excitement in the field for discovering new pathways and molecules from studying cancer therapeutic effects and side effects that may lead to new ways to improve cardiovascular care in non-cancer patients. In time, it is anticipated that the collaborative efforts from oncologists, cardiologist, cancer biologists, and cardiovascular investigators will lead to improvement in the care of both cancer and cardiovascular disease patients and their outcomes.