Annotated Bibliography

DOI:10.1093/jnci/dju232 First published online September 10, 2014

1 of 9 Commentary | JNCI Vol. 106, Issue 9 | dju232 | September 10, 2014

Published by Oxford University Press 2014.

Commentary

Cancer treatment–related Cardiotoxicity: Current State of Knowledge and Future research Priorities Nonniekaye Shelburne, Bishow Adhikari, Joanna Brell, Myrtle Davis, Patrice Desvigne-Nickens, Andrew Freedman, Lori Minasian, Thomas Force, Scot C. Remick

Manuscript received May 6, 2014; revised June 27, 2014; accepted July 1, 2014.

Correspondence to: Nonniekaye Shelburne, CRNP, MS, AOCN, Clinical and Translational Epidemiology Branch, DCCPS, 9609 Medical Center Drive, Rm 4E110, Rockville, MD 20850 (e-mail: [email protected]).

Cardiotoxicity resulting from direct myocyte damage has been a known complication of cancer treatment for decades. More recently, the emergence of hypertension as a clinically significant side effect of several new agents has been recognized as adversely affecting cancer treatment outcomes. With cancer patients living longer, in part because of treatment advances, these adverse events have become increasingly important to address. However, little is known about the cardiovascular path- ogenic mechanisms associated with cancer treatment and even less about how to optimally prevent and manage short- and long-term cardiovascular complications, leading to improved patient safety and clinical outcomes. To identify research pri- orities, allocate resources, and establish infrastructure required to address cardiotoxicity associated with cancer treatment, the National Cancer Institute (NCI) and National Heart, Lung and Blood Institute (NHLBI) sponsored a two-day workshop, “Cancer treatment–related cardiotoxicity: Understanding the current state of knowledge and future research priorities,” in March 2013 in Bethesda, MD. Participants included leading oncology and cardiology researchers and health professionals, patient advocates and industry representatives, with expertise ranging from basic to clinical science. Attendees were charged with identifying research opportunities to advance the understanding of cancer treatment–related cardiotoxicity across basic and clinical science. This commentary highlights the key discussion points and overarching recommendations from that workshop.

JNCI J Natl Cancer Inst (2014) 106(9): dju232 doi:10.1093/jnci/dju232

The primary goal of cancer treatment is to eradicate and prevent recurrence of cancer, thereby prolonging life. Improvements in the efficacy of treatment are demonstrated by the approximately 13.7 million cancer survivors alive today in the United States (1). However, cancer survival gains have revealed unintended conse- quences of therapy, such as increased incidence of cardiovascular injury. Studies have shown that cancer treatment–related cardio- toxicity is the third leading cause of treatment-associated mortality in survivors of pediatric and adolescent cancers, with recurrence and second malignancies being the two leading causes (2). The incidence of treatment-induced heart damage in pediatric survivors of cancer increases over time, even after 30 years post-therapy (3). In adult patients, cardiotoxicity is agent-dependent, and incidence can be as high as 50%, depending on the type of cardiac condi- tion (4). Five-to-ten year male survivors of adult cancer self-report heart problems to be the most common post-treatment issue, while in women survivors it is the second most reported problem fol- lowing arthritis or osteoporosis (5). Although the exact proportion of cancer survivors who develop treatment-related cardiotoxicity is unknown, the emergence of cardio-oncology subspecialty clinics suggests the prevalence and impact are clinically significant. The reasons behind the increase may be due to the increasing number of cancer survivors, use of targeted therapies, multimodality and multidrug regimens, and longer courses of cancer therapy.

Cancer patients who cannot receive optimal therapy because of concern about potential cardiotoxicity may experience com- promised treatment outcomes. Thus, the overall management strategies for cardiotoxicity during and following cancer treat- ment are critical aspects of cancer care, potentially influenc- ing overall prognosis, survival, and quality of life. Research is needed to better understand the acute and long-term effects of all cancer treatment modalities—including cytotoxic chemo- therapy, targeted therapy, and radiation—on cardiovascular function and ways to mitigate these effects (6–8). Clinicians need information to screen high-risk patients and prevent cancer treatment–related cardiotoxicity, to balance effective cancer treatment and cardiac risk assessment in treatment deci- sion making, and to manage long-term cardiac risks and effects of treatment in cancer survivors. A  fiscal year 2012 National Institutes of Health (NIH) cardiotoxicity portfolio analy- sis (Supplementary Materials, available online) indicates that while NIH research support is continuing, it may be insuffi- cient to address the range and depth of the scientific questions in this area.

The National Cancer Institute (NCI) and the National Heart, Lung, and Blood Institute (NHLBI) convened a workshop in March 2013 entitled “Cancer treatment–related cardiotoxicity: Understanding the current state of knowledge and developing

September 10 dju232

JNCI | Commentary 2 of 9jnci.oxfordjournals.org

future research priorities” in Bethesda, MD (9). The goal of this workshop was to identify research opportunities and determine scientific priorities for issues related to cancer treatment–related cardiotoxicity. Although cardiotoxicity encompasses a wide trajec- tory of disorders (Box 1), this workshop focused on hypertension and heart failure and included all cancer treatment modalities. Over 60 participants with expertise in oncology, cardiology, epide- miology, patient advocacy, research and clinical care participated in person; in addition, over 200 online participants viewed the real- time and archived meeting videocast (10,11). This report provides an overview of workshop recommendations that highlight gaps in knowledge about cancer treatment–related cardiotoxicity, research questions that will address these gaps, and resources and collabora- tions required to move this field forward. Box 2 lists recommenda- tions for each of the workshop findings discussed herein.

Workshop Findings Developing Standards for Cancer Treatment–Related Cardiotoxicity One finding the workshop participants repeatedly noted is the nonuniform collection of cardiovascular outcomes in oncology research and clinical care. A  lack of nomenclature, assessment of baseline status and toxicity grading, and existence of dis- parities and/or omission of cardiovascular side effects present challenges in the medical literature for evaluating cancer treat- ment–related cardiotoxicity. The American Heart Association’s

Box 1. Common terminology criteria for adverse events v4.0 (CTCAE) cardiotoxicity events (9).

Cardiac disorders Acute coronary syndrome Aortic valve disease Asystole Atrial fibrillation / flutter Atrioventricular block Cardiac Arrest / sudden cardiac death Chest pain and palpitations Conduction disorder Constrictive pericarditis Heart failure Hypertension Left ventricular dysfunction Mitral valve disease Myocardial infarction Myocarditis Paroxysmal atrial tachycardia Pericardial effusion Pericardial tamponade Pericarditis Restrictive cardiomyopathy Right ventricular dysfunction Sick sinus syndrome Sinus bradycardia and tachycardia Supraventricular tachycardia Thromboembolic events Tricuspid valve disease Ventricular arrhythmia Ventricular fibrillation Ventricular tachycardia Vascular disorders

Box 2. Workshop recommendations.

Developing standards

1. Consider an optimal classification system to characterize and grade cardiovascular safety parameters.

2. Develop standard definitions, common data elements, and valid measures of baseline cardiac status for can- cer treatment–related cardiotoxicity in clinical trials and clinical practice to enable comparison of data elements over time.

3. Develop a national database/repository consisting of the common data elements to conduct epidemiologic studies for determination of cardiotoxicity incidence, severity, natu- ral history, and phenotypic characteristics

Mechanisms

4. Utilize cutting-edge technology and biomarker approaches to identify new agents of potential concern and elucidate mechanisms of cardiac toxicity.

5. Utilize metabolomics to assess mitochondrial activity, myocardial energy and efficiency, signaling pathways and response to injury to enhance understanding of cardiotoxicity.

6. Examine the utility of –omics approaches in identifying car- diotoxicity of cancer therapies.

7. Engage systems biology approaches to better understand how anticancer drugs might interact with various bio- logic pathways and targets and potentially worsen cardiac damage.

8. Identify on- and off-target effects of drugs at the gene, pathway, organelle, cellular, organ, and organismal lev- els to inform cardiac toxicity screening and prediction approaches.

9. Explore the pathogenesis of cancer treatment–related car- diotoxicity using ex vivo models.

10. Identify the role of cardiac fibroblasts, endothelial cells, and the vascular bed in cancer treatment–related cardiotoxicity.

11. Investigate the role of cardiac progenitor and stem cells in cardiac repair and remodeling during and after cancer treatment.

12. Elucidate the ways that dysregulation of fundamental path- ways (eg, inflammation, angiogenesis) in cancer patients interacts with cardiac conditions.

Preclincial and animal studies

13. Enhance preclinical toxicology assessments for new cancer agents with potential cardiac toxicity.

14. Support cellular and molecular studies by enhancing access to human cardiac system tissues.

15. Modify treatment-dosing schedules in animal studies and look at quantitative damage to the cardiovascu- lar and other systems to potentially minimize toxicity clinically.

16. Use human data to validate existing vertebrate models of human age-related disease and obesity, for use in mecha- nistic studies of anticancer agents.

17. Develop preclinical models and validation criteria that can be used to identify risk of cardiac toxicity, investigate mecha- nisms of cardiac toxicity of novel compounds, and evaluate potential of cardioprotective therapies.

18. Utilize reverse translation of known treatment-induced cardiotoxic agents and primary cardiac disease pathways in the approach to cancer drug development that avoid or minimize cardiotoxicity.

Vol. 106, Issue 9 | dju232 | September 10, 20143 of 9 Commentary | JNCI

(AHA) statements and guidelines (12) on cardiac events may not be entirely pertinent for patients receiving cancer therapies. The current version of Common Terminology Criteria for Adverse Events v4.0 (CTCAE) developed by NCI classifies the undesired effects of the novel agent(s) with criteria for grading the severity of each event (13), but, by design, it does not utilize standards for assessing the event severity. Nonetheless, cardiotoxicity criteria (CTCAE v4.0 or AHA) are not consistently utilized and assess- ment measures for toxicities are often developed for individual studies. These challenges clearly preclude concise and consistent definitions to optimally reconcile and compare patient outcomes and cardiovascular side effects across studies and in clinical prac- tice. Consensus nomenclature for definitions of cardiac events and utilization of common data elements in cancer clinical trials could ensure that key information is uniformly collected during each study.

To address this barrier, workshop participants recommended the need for standard definition development, including but not limited to: a mechanism-based classification of hypertension and heart failure events, the creation of common data elements and a national data- base. Having a data repository with defined common elements could enhance the accuracy of the diagnosis of cardiac toxicity. Furthermore, such a population resource could provide researchers a way to inves- tigate the natural history of cardiotoxicity. It could also facilitate association studies of phenotype characteristics, genotype and phar- macogenomic information, and cardiac outcomes that could better predict at-risk patients and provide personalized cardiotoxicity care.

Workshop participants raised the challenges of creating such an infrastructure. Not only are there technical difficulties—stand- ardization of data, incomplete or missing data—but some unique data elements may potentially be required for some anticancer

Early phase therapeutic studies

19. Develop a phase I clinical trial database of in-depth cardiac information early in drug development.

20. Standardize oncology clinical trial entry eligibility for preex- isting cardiovascular disease and comorbidities.

Minimally invasive methods for diagnosing and monitoring

21. Develop cardiotoxicity risk stratification tools utilizing prom- ising clinical, genomic, and imaging data.

22. Compare promising cardiotoxicity screening and detection modalities via large retrospective and prospective cancer cohorts that include genomic, biomarker, and imaging data.

23. Identify and confirm genetic markers linked to cardiac myo- cyte and endothelial cell damage and cardiomyopathy risk during and after cancer treatments through cardiomyopa- thy gene panels, whole genome or exome sequencing, can- didate gene studies, and single nucleotide polymorphism arrays.

24. Conduct biomarker studies with the goal of discovering and validating mechanistic markers that can be utilized as robust screening tools for early identification of cardiotoxic- ity with greater sensitivity and specificity.

25. Support replication studies to validate cardiac biomarkers in the cancer clinical trial population by enhancing access to biorepositories.

26. Examine the role of kinetic biomarkers, microRNA, and etcin in the prediction and detection of cardiotoxicity.

27. Develop minimally invasive MRI methods (eg, late-gla- donium enhancement, T1 and T2 mapping, and MR spec- troscopy) to assess and monitor early in vivo changes in myocardial tissue, including apoptosis, fibrosis, oxidative stress and other metabolic perturbations because of cancer treatments.

28. Identify promising imaging modalities and test their effec- tiveness to predict or stratify cancer therapeutics–induced cardiac toxicity risk in humans.

29. Conduct comparative effectiveness studies for noninvasive imaging techniques that may detect early onset of cancer treatment–related cardiotoxicity.

Prevention

30. Identify the most effective pharmacologic cardio-protective interventions that do not interfere with anticancer treat- ment efficacy and that mitigate or reverse cardiac dam- age, specifically pertaining to dose, type, and duration of medication.

31. Examine and measure the impact of nonpharmacologic car- dio-protective interventions, including diet, physical activity, and comorbidity management, in preventing cardiotoxicity and determining effects on cancer recurrence, progression, and survival.

32. Identify the optimal strategy for cardio-protection approaches for hypertension and heart muscle damage dur- ing cancer treatment and patient subgroups that receive the greatest benefit from prophylactic cardio-protection.

Treatment

33. Develop treatment guidelines for determining when to utilize pharmacologic interventions for cancer treatment– related cardiotoxicity.

34. Determine the optimal medications most effective for the treatment of cardiotoxicity and treatment targets.

35. Examine and measure the impact of nonpharmacologic pro- tective interventions, including diet and physical activity, in

treating cardiotoxicity and the effect on cancer recurrence, progression, and survival.

36. Develop cancer treatment–related, cardiotoxicity evidence– based practice guidelines for risk prediction, screening, prevention, diagnostics, treatment, and follow-up utilizing a trans-disciplinary working group.

Survivorship

37. Conduct epidemiologic studies to determine incidence, severity, and natural history of various cardiotoxicities and identify the specific clinical and epidemiologic risk profile that is associated with susceptibility to the development of cardiotoxicity.

38. Identify the major determinants of cardiac reserve via large cohort studies with direct assessment and longi- tudinal follow-up, including pediatric and adult cancer survivors.

39. Identify the impact of early detection and treatment of car- diotoxicity on cardiac and cancer outcomes in the cancer survivor.

40. Investigate the long-term effects of newer cancer therapies and combination therapy in cancer survivors.

41. Develop care coordination and communication methods, implementation standards and evaluation measures among multidisciplinary teams for cardiotoxicity surveillance and potential interventions for prevention and/or treatment in cancer survivors.

JNCI | Commentary 4 of 9jnci.oxfordjournals.org

therapies. These problems are compounded by the fact that not all clinics and hospitals have access to the same equipment or labo- ratories for identifying cardiac conditions. Despite these scientific and logistical issues, the science of cancer treatment–related car- diotoxicity will be advanced by a consensus classification system.

research opportunities: Basic Science Mechanisms of Cancer Treatment–Related Cardiotoxicity Understanding fundamental mechanisms underlying cancer treat- ment–related cardiotoxicity is essential to the development of new methods to monitor, treat, and prevent these toxicities. For exam- ple, mechanistic studies contributed to the approval of dexrazoxane by the Food and Drug Administration (FDA) as a pretreatment for prevention of cardiac injury in adults receiving anthracycline ther- apy (14,15). In addition, recent studies suggest that topoisomer- ase IIB inhibition and engineered neuregulin ligands may offer novel ways to protect the heart from anthracycline injury without reducing its therapeutic efficacy (4,16–18). Additional mechanistic studies may lead to novel cardioprotective strategies, not only for anthracyclines but for other targeted agents as well.

Compared with research that has revealed mechanisms of anthracycline-induced cardiotoxicity, very little is known about the molecular pathways that underlie cardiotoxicities associated with emerging cancer therapeutics. This is particularly true for the small molecule tyrosine kinase inhibitors developed for various forms of cancer treatments, which at present include 19 agents, with many more in development (18). The majority of these agents have off- tumor target cardiovascular effects (19) and there is poor under- standing of the role the various kinases and defined drug targets play in the heart and peripheral vasculature.

As our mechanistic understanding of cardiotoxicity progresses, it is becoming increasingly apparent that the key signaling path- ways involve less understood and vulnerable cellular signaling pathways that affect the cardiovascular mitochondria, cardiomyo- cytes, endothelial cells, fibroblasts, vascular pericytes, and cells involved in inflammation, angiogenesis, cardiac repair and regen- eration. Future studies are needed to not only characterize how these signaling pathways are affected by the therapeutic agents, but also to develop strategies to minimize their injury and reduce the associated cardiotoxicity. This approach requires defining cardio- vascular risk phenotypes so injury can be accurately connected to translational findings.

Preclinical and Animal Studies in Cancer Treatment–Related Cardiotoxicity The application of more suitable disease models and more effec- tive methods for toxicity screening will serve to advance our understanding of cardiotoxicity. Animal models have been suc- cessfully employed to predict potential problems in some cases (18), but more highly predictive models are needed. For instance, in rodent models, adding an additional stressor, such as hyperten- sion or thoracic aortic constriction, may be needed to inform risk for cancer patients with these comorbidities. Large animal mod- els, although believed to be superior in some cases, could also be improved by more advanced monitoring and application of clini- cal biomarkers.

Given the urgency to identify risk of cardiotoxicity earlier, strat- egies that can identify problematic agents and elucidate mechanisms of toxicity are needed. In the preclinical drug discovery setting, however, animal models are typically not as amenable to compound screening intended for comparing several compounds. Screening alternatives such as zebrafish, human induced pluripotent (hiPS)- derived cardiomyocytes, and human cardiac microtissue arrays to test for toxicity were suggested as opportunities (20–22). All of these approaches may be adaptable to high-throughput screen- ing. It is also possible to derive information on metabolism using metabolomic approaches in cell culture and zebrafish models, with possibilities of identifying novel biomarkers of injury that could then be tested and validated in patients.

research opportunities: Clinical Care Early Phase Therapeutic Studies in Cancer Treatment–Related Cardiotoxicity In the absence of adverse preclinical toxicology and pharmacologic observations, novel anticancer agents are first clinically evaluated in cancer patients. Not uncommonly, however, when agents are administered for the first time in humans, unanticipated toxicities emerge. It is important to identify potential cardiotoxicity using validated measures early in drug development without premorbid cardiovascular risks compounding the clinical evaluation for car- diac safety and to assure patient safety. While this is the ideal sce- nario, this approach is not entirely pragmatic, as the majority of patients ineligible for Phase I  trials are often older with multiple pretrial comorbid cardiovascular risks or overt atherosclerotic car- diovascular disease. Workshop participants suggested that it may be possible to include cancer patients with premorbid cardiovascu- lar disease in early phase trials, similar to the approach taken by the NCI Organ Dysfunction Working Group, which evaluates patients with hepatic and renal dysfunction (23).

Once an agent’s cardiotoxic risk has been established, other obstacles to early phase testing remain. Only a minority of Phase I agents advance to phase II testing, usually due to intolerable tox- icity. While cardiac assessment of subsequently abandoned drugs has scientific merit, resources are not available to sort out the challenging cardiovascular safety profiles for a given agent. Even if the requirements for cardiovascular system–focused preclinical and toxicology testing are satisfied, drug sponsors may not want to engage in further toxicity assessments, especially if the results invite restrictions to the drug label at time of FDA approval. Given the high incidence of baseline cardiovascular disease and the shared risk factors with drug development, cancer treatment–related car- diotoxicity should be well-described and managed for patient safety (24,25).

Minimally-Invasive Methods for Diagnosis and Monitoring Biopsy-based approaches have been used to diagnose and moni- tor (anthracycline-induced) cardiotoxicity, but more noninvasive or minimally invasive approaches to diagnosing and monitoring cancer treatment–related cardiotoxicity are needed. The following section outlines areas of research opportunities identified by work- shop attendees, categorized into biomarkers and imaging.

Vol. 106, Issue 9 | dju232 | September 10, 20145 of 9 Commentary | JNCI

Biomarkers. Biomarkers may represent one of the most cost- effective and minimally invasive means for diagnosing and moni- toring cardiac injury in an oncology setting. Workshop attendees discussed the similarities of cancer treatment–induced cardiac dam- age to the traditional process of cardiac damage because of aging and environmental factors, with an exception around timing, as the injury process is accelerated and compressed after cancer treat- ment. As a result, studies are needed to investigate the potential of known cardiac biomarkers to detect asymptomatic cancer treat- ment–related cardiac damage and to predict cumulative effects of initial injury or loss of cardiac myocytes. High sensitivity troponin is, at present, considered the biomarker of choice for the detection of cardiac injury (26–28). Although the biochemical characteris- tics and utility of troponins for the diagnosis of cardiac injury and acute myocardial infarction have been extensively reviewed (29), it is clear that application of troponins in a cardio-oncology set- ting will require understanding the basic science of these proteins, validation against true clinical phenotypes, and if and how they are influenced by other treatments. There is also considerable debate about whether biomarkers such as troponins or creatinine kinase MB fraction are released with reversible as well as irreversible car- diac injury, warranting further study.

Additional research to investigate whether these biomarkers may fulfill the role as an indicator of acute risk as well as serving as a risk factor in prediction of the long-term adverse effects of myocardial injury and hypertension is warranted. Studies of cardiac remodeling in response to certain agents appear promising. Further investiga- tion of biomarkers (eg, galactin-3, N-terminal brain natriuretic pep- tide, titin, neuregulin-1, hypoxia-inducible factor-1, Topoisomerase II beta) that measure changes in pathways that drive cardiac and cancer disease processes and therapeutic response is needed (27,30–32). Of paramount importance, investigators will need to consider the meaningfulness of these assay results in cancer treat- ment–related cardiotoxicity and how clinicians should respond to biomarker changes. The ultimate goal would be to identify patients at an asymptomatic stage and intervene. Investigators may need to identify multiple markers to more accurately measure risk and make decisions about early intervention.

There is a need to improve available biomarker data, includ- ing access to biospecimens. Workshop participants suggested using existing cohort studies, where possible, and encouraging collabora- tions between cooperative oncology groups. For example, the reor- ganized NCI’s Clinical Trials Network could be leveraged where blood and tissue from prospectively studied cancer patients are collected and stored. Although cost, consent for older studies, and patient management issues might be a barrier to developing such a resource, it would allow pretreatment genetic assessment as well as longitudinal follow-up.

Imaging. Minimally invasive imaging methods are indispensable tools to monitor cardiac function and cardiac injury. Unfortunately, current clinical cardiac imaging modalities lack the resolution (or sensitivity) and specificity to assess early, subclinical changes in myocardial tissue (33). Imaging methods that allow tissue analy- sis in vivo will be important because there is strong experimental evidence supporting cell- and tissue-level changes preceding func- tional alterations at the organ level (33–35). Early and non-invasive

detection of cancer therapy–induced cardiac tissue injury would clearly not only help optimize cancer treatment but could also pro- vide insights leading to novel cardio-protective interventions.

Current approaches to assess cardiac function primarily utilize ultrasound-based techniques, such as transthoracic echocardiog- raphy and 2D echocardiography. Compared with other available methods, these techniques are inexpensive, easy to use, noninvasive and highly portable and therefore preferable. However, most com- monly used means to assess cardiac function by echocardiography suffers from poor reproducibility (36). Scintigraphic methods, such as multigated acquisition scan or gated blood pool scan, are much more reproducible at measuring left ventricular ejection fraction but the radiation dose with repeated examinations dampens enthu- siasm for using this approach long term (36). Newer ultrasound approaches, such as 3D echocardiography, strain rate imaging, and tissue Doppler imaging, continue to improve, and further innova- tion is needed to make these methods more reproducible for serial monitoring in the assessment of toxicity progression over time.

Workshop attendees stated that cardiac magnetic resonance imaging (MRI) methods should be further developed and tested for cardiac toxicity assessment. For example, recent advances in late- gadolinium enhancement and T1 and T2 mapping approaches, together with the development of molecularly targeted imaging probes, offer unique opportunities to monitor tissue fibrosis and apoptosis in vivo (37). Continuing advances in multimodal imaging that combine complementary imaging modalities may also offer unique advantages and need to be explored. In summary, while substantial advances have been made in noninvasive imaging, many questions remain as to the utility of these strategies, specifically in cancer patients with possible cardiotoxicity.

Prevention and Treatment of Cancer Treatment–Related Cardiotoxicity The aging of cancer patients, application of combination therapy, and emerging novel agents offer both challenges and opportunities. Older patients are more likely to have comorbidities including pre- existing cardiovascular disease, which in turn may exacerbate car- diotoxicity, complicating cancer therapy. The workshop attendees noted the need to examine both the prevention and treatment of cardiotoxicity, noting that as cancer therapies continue to improve, the prevalence of cancer treatment–related cardiovascular dysfunc- tion will likely increase.

The need for clinical assessment of cardiovascular risk prior to initiation of cancer treatment is evident and would likely have implications for treatment planning, including weighing cancer and cardiac risk/benefit, which may change in early vs late-stage can- cer. Standard surveillance assessments to determine pretreatment risk, as well as serial measures of cardiovascular integrity during and in follow-up of cancer therapy are needed. These assessments will inform patient management and are needed to better under- stand individual risk (eg, age, comorbidities, genetic susceptibility), inform drug and radiation dose, predict response to cancer therapy, and develop effective prevention of cardiotoxicity. Evidence-based nonpharmacologic cardio-protective interventions, including diet, physical activity, and comorbidity management, need to be stud- ied within various at-risk cancer populations in order to deter- mine efficacy. Existing infrastructure such as the NHLBI funded

JNCI | Commentary 6 of 9jnci.oxfordjournals.org

Cardiovascular Research Network and NCI funded Clinical Trials Network are potential resources to establish, implement, and eval- uate standard measures in clinical trials, identify imaging and bio- markers sensitive to subclinical changes, and identify reversible and long-term adverse cardiovascular outcomes.

Further identification of clinical, genetic, biomarker, and imag- ing cardiac risk factors might allow for better stratification of patients at low and high risk for cancer treatment–induced cardio- toxicity and lead to improved treatment and monitoring options and increased safety of cancer therapy without compromising sur- vival. Several investigators have attempted to develop risk predic- tion models that include a number of these risk factors, although they exclude radiation therapy as a risk factor and their use in clini- cal care has been limited (26–28). Large retrospective and prospec- tive cancer cohorts that include detailed genomic, biomarker, and imaging data linked to cardiovascular outcomes will ultimately be required to allow the discovery of factors that accurately predict the development of cardiotoxicity for specific anticancer agents and also permit the direct comparison of different risk factor algorithms.

Clinical practice guidelines for controlling cardiovascular complications of cancer treatment across disciplines are essential. Systematic surveillance data and standard metrics from clinical tri- als and patient management studies are lacking and are needed to improve evidence. Further, the current large number of recom- mendations for managing cardiovascular and oncology diseases are cumbersome, making consistent application a challenge, especially given the limited recommendation on the intersection of the two

diseases. (38). As noted earlier, recognition and reporting of cardio- toxicity suffers from lack of standards and long-term surveillance. The utility of conventional cardiovascular treatments in cancer treatment–induced cardiotoxicity management offers promise and is supported by preclinical, observational, and small trials, but lacks rigorous randomized trial evidence, standard reporting of adverse outcomes, and infrastructure for long term surveillance (4,25,39).

Survivorship and Cancer Treatment–Related Cardiotoxicity As the number of pediatric and adult cancer survivors continues to increase, surveillance and monitoring for cancer treatment–related cardiotoxicity is an essential care component for at-risk persons. While the extent to which cardiac status is affected in short- and long-term cancer survivors exposed to various cardiotoxic treat- ment regimens remains unclear, evidence suggests that the risk of toxicity continues to increase many years after treatment (3). The timing and types of cardiosafety surveillance lack universal defini- tion and implementation across cancer survivors. Likewise, existing health screening recommendations, based on patient history and potential cardiovascular effects of predisposing therapy, lack con- sistent diagnostic definitions and interventions (12,40).

There is a critical need to conduct well-designed epidemio- logic studies to determine incidence, severity, and natural his- tory of various cardiotoxicities. Previous studies have varied widely (0% to 57%) in reporting the incidence rate of cancer therapy–induced cardiotoxicity, which can be attributed to dif- ferences in study design (41–45). Previous studies have focused

Table 1. Established and recommended resources for cardiotoxicity research*

Category Established resources Resource needs

Practice standards The FDA Global Data Elements Initiative—defining cardiovascular endpoints and imaging

International Cardio-oncology Society

A joint NHLBI, NCI, professional society task force to develop common, feasible goals and guideline topics by identifying small pre-, during, and post-therapy topics and addressing effectiveness as well as other domains of quality

Mechanisms The National Council on Radiation Protection and Measurements, National Library of Medicine, National Center of Biomedical Oncology, and others—linking terminologies

for a revised ontology FDA is collaborating with academic institutions to

build ontology in the adverse event–reported terms, in a way that maps back to biology

Gene and tissue arrays Cell cultures or ‘organs on chips’ New in vitro and in vivo models Systematic database to organize the

network across levels in computable and quantifiable ways

Data linked by phenotype Transdisciplinary approach to otology

Markers of risk and injury Prospective, longitudinal cohorts with common data elements

Prevention, treatment, and survivorship

Models for patient-centered medical home (PCMH) and survivor care plans exist to assist patients and providers coordinate post-treatment care between disciplines

Tools for stratification of samples Prospective, longitudinal studies Centralized databases Common data elements Larger sample size studies with common

data elements and repeated measures General NCI DCCPS EGRP Knowledge Integration website

on cancer treatment–related cardiotoxicity http://epi.grants.cancer.gov/ cardiotoxicity/

* DCCPS = Division of Cancer Control and Population Sciences; EGRP = Epidemiology and Genomics Research Program; FDA = US Food and Drug Administration; NCI = National Cancer Institute; NHLBI = National Heart, Lung and Blood Institute.

Vol. 106, Issue 9 | dju232 | September 10, 20147 of 9 Commentary | JNCI

on the effects of many different drug and radiation therapies, all with different incidence rates (46). In addition, these studies used different classifications, such as symptomatic, subclinical, acute, early (within one year) or late effects, for varying con- ditions across the cardiotoxicity spectrum. Other major factors affecting the incidence of cancer therapy–induced cardiotoxic- ity in the various populations studied include age, cancer types, stage, follow-up length, sample size, health status of the patients, and cardiovascular risk factors.

Workshop participants also stated that large longitudinal stud- ies of cardiotoxicity in cancer survivors need to be conducted. Collecting detailed cancer and concomitant treatment, biomarker, and imaging data in a standardized manner is required to allow direct comparison of screening modalities and novel identification of cardiovascular injury. In addition, data on host factors, comorbid conditions, health behaviors (eg, diet, physical activity), and genetic factors need to be collected. Follow-up over many years is also rec- ommended for determining long-term cardiotoxic outcomes, as well as effectiveness of cardioprotective interventions among new and combination therapies. Longitudinal studies of childhood can- cer survivors could provide insights into whether cancer treatment accelerates the aging process in relation to cardiovascular damage.

Workshop participants recommended mining cancer registries and electronic health records for retrospective data, as well as modeling future cohort studies after the Childhood Cancer Survivorship Study and Framingham Study, emphasizing the need for biospeci- men collection. Other possible sources to mine include the FDA and industry data.

Strategies are needed to implement patient-centered and col- laborative approaches to caring for cancer survivors. This type of approach might help resolve certain barriers to assessing, pre- venting, and treating cardiotoxicity. The patient-centered medical home, survivorship clinics, and dynamic survivorship care plans are potential resources to identify, monitor, and treat long-term cardiac effects with the involvement of oncology, cardiology, and primary care providers. This team-oriented approach provides coordinated and managed care across specialties and includes care planning and shared decision making (47). More research is needed to determine when to transition survivors from the oncology team to a multidisciplinary survivorship care team and the optimal approach to developing a cancer survivorship plan. As evidence is being developed, consensus recommendations and guidelines by oncology and cardiology professional societies and think tanks are needed.

Summary This report highlights the broad scope of basic and clinical car- diotoxicity research across all cancer treatment regimens needed to inform the field of cardio-oncology and improve patient out- comes. This is an area of great interest and the conceptualized and developed approaches above have the potential to contribute to progress in understanding the mechanisms of cancer treatment– related cardiotoxicity and translating findings to improve risk stratification, screening, prevention, and treatment. The NCI- NHLBI workshop on cancer treatment–related cardiotoxicity was a first step in bringing together a transdisciplinary team to identify the research priorities, resources and infrastructure needed (9). Outcomes of this workshop to date include the archived meeting videocast available online (10,11), over 40 research recommen- dations across the cardiotoxicity spectrum (Box 2), identification of established and recommended resources (Table  1) and infra- structure (Box 3) needed to move cardiotoxicity research forward, development of the NCI Community Oncology Cardiotoxicity Task Force, and several new collaborations spanning basic and clinical science. The workshop recommendations are being used to develop and shape NCI’s and NHLBI’s research priorities around cancer treatment–related cardiotoxicity, as well as gener- ate new and expanded research collaborations within the cardio- oncology community.

references 1. Siegel R, DeSantis C, Virgo K, et  al. Cancer treatment and survivorship

statistics, 2012. CA Cancer J Clin. 2012;62(4):220–241. 2. Mertens AC, Yasui Y, Neglia JP, et  al. Late mortality experience in five-

year survivors of childhood and adolescent cancer: the Childhood Cancer Survivor Study. J Clin Oncol. 2001;19(13):3163–3172.

3. Meacham LR, Chow EJ, Ness KK, et  al. Cardiovascular risk factors in adult survivors of pediatric cancer—a report from the childhood cancer survivor study. Cancer Epidemiol Biomarkers Prev. 2010;19(1):170–181.

Box 3. Infrastructure needed to advance cancer treatment–related cardiotoxicity research.

1. Interprofessional education and cardio-oncology training programs.

2. Requests for applications and proposals from professional societies representing primary care providers, oncologists, and cardiologists to collaboratively develop both research and clinical guidelines.

3. Create mechanisms for investigators to leverage exist- ing infrastructure within the National Cancer Institute and National Heart Lung and Blood Institute, such as the Clinical and Translational Science Awards, oncology and cardiology trial networks, the Childhood Cancer Survivorship Study, existing funded (primarily clinical) studies, genetic data- bases, existent clinical data and resources, and informatics resources.

4. Studies across institutions with centralized recruitment. 5. Examine the role of implementation science to improve

translation. 6. Supplements for cross collaborators to be added to large

grants. 7. Create novel mechanisms of funding and resource support

for exploratory studies or the creation of a new cross-disci- plinary research network.

8. Use special emphasis panels or other multidisciplinary review groups for cross-cutting science.

9. Large, longitudinal cohort studies and epidemiology. 10. Case-control studies will be useful for investigating rare out-

comes to examine mortality and provide insight into mecha- nisms for various cardiotoxic outcomes.

11. Mechanisms are needed to foster broad interdisciplinary science across the study of pharmacology, computational science, biostatistics, epidemiology, and other scientific disciplines.

12. Biobanks of blood and other biospecimens would facilitate the testing of biomarkers, genome analyses, and biology.

13. Develop cancer treatment–related cardiotoxicity research consortium or task force.

JNCI | Commentary 8 of 9jnci.oxfordjournals.org

4. Yeh ET, Bickford CL. Cardiovascular complications of cancer therapy: incidence, pathogenesis, diagnosis, and management. J Am Coll Cardiol. 2009;53(24):2231–2247.

5. Schultz PN, Beck ML, Stava C, et  al. Health profiles in 5836 long-term cancer survivors. Int J Cancer. 2003;104(4):488–495.

6. Curigliano G, Mayer EL, Burstein HJ, et  al. Cardiac toxicity from sys- temic cancer therapy: a comprehensive review. Prog Cardiovasc Dis. 2010;53(2):94–104.

7. Lipshultz SE, Adams MJ, Colan SD, et  al. Long-term cardiovascular toxicity in children, adolescents, and young adults who receive cancer therapy: pathophysiology, course, monitoring, management, prevention, and research directions: a scientific statement from the American Heart Association. Circulation. 2013;128(17):1927–1995.

8. Travis LB, Ng AK, Allan JM, et  al. Second malignant neoplasms and cardiovascular disease following radiotherapy. J Natl Cancer Inst. 2012;104(5):357–370.

9. National Cancer Institute. Cancer Treatment-Related Cardiotoxicity: Understanding the Current State of Knowledge and Developing Future Research Priorities. http://epi.grants.cancer.gov/workshops/cardiotoxicity/ Accessed August 11, 2014.

10. National Institutes of Health Videocasting and Podcasting. Cancer Treatment Related Cardiotoxicity Day 1. http://videocast.nih.gov/sum- mary.asp?live=11985&bhcp=1 Accessed August 11, 2014.

11. National Institutes of Health Videocasting and Podcasting. Cancer Treatment Related Cardiotoxicity Day 2. http://videocast.nih.gov/sum- mary.asp?live=11987&bhcp=1 Accessed August 11, 2014.

12. American Heart Association. Statements & Guidelines. http:// my.americanheart.org/professional/StatementsGuidelines/Statements- Guidelines_UCM_316885_SubHomePage.jsp Accessed August 11, 2014.

13. National Cancer Institute. Common terminology criteria for adverse events (CTCAE). Version 4.0 http://evs.nci.nih.gov/ftp1/CTCAE/CTCAE_ 4.03_2010-06-14_QuickReference_5x7.pdf. Accessed August 11, 2014.

14. Myers C. The role of iron in doxorubicin-induced cardiomyopathy. Semin Oncol. 1998;25(4 Suppl 10):10–14.

15. Xu X, Persson HL, Richardson DR. Molecular pharmacology of the inter- action of anthracyclines with iron. Mol Pharmacol. 2005;68(2):261–271.

16. Force T, Wang Y. Mechanism-based engineering against anthracycline car- diotoxicity. Circulation. 2013;128(2):98–100.

17. Jay SM, Murthy AC, Hawkins JF, et al. An engineered bivalent neuregulin protects against doxorubicin-induced cardiotoxicity with reduced proneo- plastic potential. Circulation. 2013;128(2):152–161.

18. Ky B, Vejpongsa P, Yeh ET, et al. Emerging paradigms in cardiomyopathies associated with cancer therapies. Circ Res. 2013;113(6):754–764.

19. Eschenhagen T, Force T, Ewer MS, et  al. Cardiovascular side effects of cancer therapies: a position statement from the Heart Failure Association of the European Society of Cardiology. Eur J Heart Fail. 2011;13(1):1–10.

20. Clements M, Thomas N. High-Throughput Multi-Parameter Profiling of Electrophysiological Drug Effects in Human Embryonic Stem Cell Derived Cardiomyocytes Using Multi-Electrode Arrays. Toxicol Sci. 2014; doi:10.1093/toxsci/kfu084.

21. Liang P, Lan F, Lee AS, et  al. Drug screening using a library of human induced pluripotent stem cell-derived cardiomyocytes reveals disease-spe- cific patterns of cardiotoxicity. Circulation. 2013;127(16):1677–1691.

22. Kitambi SS, Nilsson ES, Sekyrova P, et al. Small molecule screening plat- form for assessment of cardiovascular toxicity on adult zebrafish heart. BMC Physiol. 2012;12:3.

23. Takimoto CH, Mita AC. Design, conduct, and interpretation of organ impairment studies in oncology patients. J Clin Oncol. 2006;24(21):3509– 3510; author reply 3510–1.

24. Maitland ML, Bakris GL, Black HR, et  al. Initial assessment, surveil- lance, and management of blood pressure in patients receiving vascular endothelial growth factor signaling pathway inhibitors. J Natl Cancer Inst. 2010;102(9):596–604.

25. Steingart RM, Bakris GL, Chen HX, et al. Management of cardiac toxicity in patients receiving vascular endothelial growth factor signaling pathway inhibitors. Am Heart J. 2012;163(2):156–163.

26. Sawaya H, Sebag IA, Plana JC, et al. Assessment of echocardiography and biomarkers for the extended prediction of cardiotoxicity in patients treated

with anthracyclines, taxanes, and trastuzumab. Circ Cardiovasc Imaging. 2012;5(5):596–603.

27. Sawaya H, Sebag IA, Plana JC, et  al. Early detection and predic- tion of cardiotoxicity in chemotherapy-treated patients. Am J Cardiol. 2011;107(9):1375–1380.

28. Cardinale D, Sandri MT, Colombo A, et al. Prognostic value of troponin I  in cardiac risk stratification of cancer patients undergoing high-dose chemotherapy. Circulation. 2004;109(22):2749–2754.

29. Alpert JS, Thygesen K, Antman E, et al. Myocardial infarction redefined- -a consensus document of The Joint European Society of Cardiology/ American College of Cardiology Committee for the redefinition of myo- cardial infarction. J Am Coll Cardiol. 2000;36(3):959–969.

30. Ky B, Putt M, Sawaya H, et  al. Early increases in multiple biomark- ers predict subsequent cardiotoxicity in patients with breast cancer treated with doxorubicin, taxanes, and trastuzumab. J Am Coll Cardiol. 2014;63(8):809–816.

31. Motiwala SR, Szymonifka J, Belcher A, et al. Measurement of novel bio- markers to predict chronic heart failure outcomes and left ventricular remodeling. J Cardiovasc Transl Res. 2014;7(2):250–261.

32. Vassiliadis E, Rasmussen LM, Byrjalsen I, et  al. Clinical evaluation of a matrix metalloproteinase-12 cleaved fragment of titin as a cardiovascular serological biomarker. J Transl Med. 2012;10:140.

33. Jiji RS, Kramer CM, Salerno M. Non-invasive imaging and moni- toring cardiotoxicity of cancer therapeutic drugs. J Nucl Cardiol. 2012;19(2):377–388.

34. Aissiou M, Perie D, Cheriet F, et al. Imaging of early modification in car- diomyopathy: the doxorubicin-induced model. Int J Cardiovasc Imaging. 2013;29(7):1459–1476.

35. Gillespie HS, McGann CJ, Wilson BD. Noninvasive diagnosis of chemo- therapy related cardiotoxicity. Curr Cardiol Rev. 2011;7(4):234–244.

36. de Geus-Oei LF, Mavinkurve-Groothuis AM, Bellersen L, et  al. Scintigraphic techniques for early detection of cancer treatment-induced cardiotoxicity. J Nucl Med Technol. 2013;41(3):170–181.

37. Karamitsos TD, Neubauer S. The prognostic value of late gadolinium enhancement CMR in nonischemic cardiomyopathies. Curr Cardiol Rep. 2013;15(1):326.

38. Lindenfeld J, Kelly PA. Developing a cardiology-oncology clinical practice guideline. Prog Cardiovasc Dis. 2010;53(2):173–179.

39. McMurray JJ, Adamopoulos S, Anker SD, et  al. ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2012: The Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure 2012 of the European Society of Cardiology. Developed in collabo- ration with the Heart Failure Association (HFA) of the ESC. Eur Heart J. 2012;33(14):1787–1847.

40. National Comprehensive Cancer Network. NCCN Practice Guidelines. h t t p : / / w w w. n c c n . o r g / p r o f e s s i o n a l s / p h y s i c i a n _ g l s / f _ g u i d e l i n e s . asp#detection Accessed August 11, 2014.

41. Diller L, Chow EJ, Gurney JG, et  al. Chronic disease in the Childhood Cancer Survivor Study cohort: a review of published findings. J Clin Oncol. 2009;27(14):2339–2355.

42. Kremer LC, van der Pal HJ, Offringa M, et al. Frequency and risk factors of subclinical cardiotoxicity after anthracycline therapy in children: a sys- tematic review. Ann Oncol. 2002;13(6):819–829.

43. Moja L, Tagliabue L, Balduzzi S, et al. Trastuzumab containing regimens for early breast cancer. Cochrane Database Syst Rev. 2012;4:CD006243.

44. Smith LA, Cornelius VR, Plummer CJ, et al. Cardiotoxicity of anthracy- cline agents for the treatment of cancer: systematic review and meta-anal- ysis of randomised controlled trials. BMC Cancer. 2010;10:337.

45. Swain SM, Whaley FS, Ewer MS. Congestive heart failure in patients treated with doxorubicin: a retrospective analysis of three trials. Cancer. 2003;97(11):2869–2879.

46. Schmitz KH, Prosnitz RG, Schwartz AL, et  al. Prospective surveillance and management of cardiac toxicity and health in breast cancer survivors. Cancer. 2012;118(8 Suppl):2270–2276.

47. Institute of Medicine. Patient-Centered Cancer Treatment Planning: Improving the Quality of Oncology Care: Workshop Summary. Washington, DC: The National Academies Press, 2011. http://www.nap.edu/openbook. php?record_id=13155. Accessed August 11, 2014.

Vol. 106, Issue 9 | dju232 | September 10, 20149 of 9 Commentary | JNCI

Funding This work was supported by The National Cancer Institute and National Heart, Lung and Blood Institute.

notes NCI and NHLBI employees, along with extramural investigators, wrote this commentary and chose to submit for publication.

The authors thank the workshop speakers (9), attendees, and Alice Mascette, MD, for her work on the planning team.

Affiliations of authors: Division of Cancer Control and Population Sciences (NS, AF) and Division of Cancer Prevention (JB, LM) and Division of Cancer Treatment and Diagnosis (MD), National Cancer Institute, Rockville, MD; Division of Cardiovascular Sciences, National Heart, Lung and Blood Institute, Bethesda, MD (BA, PDN); Cancer Center, MetroHealth Medical Center and Casewestern Reserve University, Cleveland, OH (JB); Vanderbilt Heart and Vascular Institute, Vanderbilt University School of Medicine, Nashville, TN (TF); Mary Babb Randolph Cancer Center, West Virginia University (SCR), Morgantown, WV.