Cardiac toxicity in cancer survivors


  • European-American Dialogues on Cancer Survivorship: Current Perspectives and Emerging Issues

  • This supplement was guest edited by Vittorio Mattioli, MD (NCRC, Bari, Italy) and Kevin Stein, PhD (American Cancer Society, Atlanta, Georgia) and was produced with the authoritative contribution of 58 authors from the European Union and the United States. The primary aims are to highlight the potential differences between European and American approaches to cancer survivors' issues, increase coordination among oncologists and other primary care providers, and aid the development of a shared care model that can improve the quality of cancer care.

  • The opinions or views expressed in this supplement are those of the authors and do not necessarily reflect the opinions or recommendations of the journal editors, the American Cancer Society, John Wiley & Sons, Inc., or the National Cancer Research Centre Istituto Tumori “Giovanni Paolo II” Bari.

  • All 4 authors contributed equally in this work.

  • We greatly appreciate the expert article preparation by Lauren Morrow.


Cardiac disease is a major concern for cancer survivors, and this can be manifested as cardiac dysfunction from myocardial damage, valvular disease, atherosclerosis, and/or pericardial disease. In this “dialogue” between select European and American investigators, present current perspectives (both similarities and differences) are presented regarding the diagnosis and management of cardiac disease among cancer survivors, with a focus on left ventricular (LV) systolic dysfunction and heart failure. The authors conclude that comprehensive cardiac risk assessment is necessary to optimally manage all cancer survivors, and the integration of common definitions is necessary. Ongoing and future research will need to incorporate cardiovascular management principles in the long-term assessment of cancer survivors. Cardiac biomarkers, troponin, and the natriuretic peptides are becoming essential in the management of cardiac disease in cancer survivors coupled with periodic use of sophisticated imaging tools. Recognition of specific cancer therapy and the increased cardiovascular risk is an ongoing task that will remain of paramount importance for optimal outcomes among cancer survivors. Cancer 2013;119(11 suppl):2131-42. © 2013 American Cancer Society.


Over the past 10 to 15 years, cardiac complications resulting from cancer therapy have been recognized increasingly as major contributors to morbidity and, ultimately, mortality in cancer survivors.[1-6] However, despite this increased recognition, there are major limitations in our collective understanding of the proper tools necessary for the identification, treatment, or prevention of these complications. Although many systematic reviews have been conducted with regard to various topics related to cardiac disease in this population (and are referenced in the current article), this “dialogue” between European and American investigators will provide insight to some of the current perspectives in Europe and the United States on viewing and managing this burgeoning issue. Because of the success of cancer therapy and the complexity of the treatment regimens that are being increasingly used to control cancer, there are myriads of cardiac conditions that must be considered in cancer survivors. Although the term “cardiotoxicity” generally refers to heart damage as a result of treatment, specifically, this has been used most commonly to describe left ventricular (LV) systolic dysfunction and heart failure (HF) related to certain types of chemotherapy.[7-9] Although LV systolic dysfunction and HF are the focus of this report, it is important to recognize that other components of the cardiovascular system (eg valves, vasculature, pericardium) also may be affected by both chemotherapy and radiotherapy, with subsequent effects on heart function.[5, 6, 10-12] Featuring select European and American perspectives, 5 topics are discussed below: 1) definitions of “cardiotoxicity” and current limitations, 2) epidemiology and risk factors, 3) early detection, 4) treatment, and 5) possible prevention strategies. Finally, in conclusion, suggested areas for research and improved clinical practice are outlined for development internationally.

Definitions of “Cardiotoxicity” and Current Limitations

American perspective

Historically, HF resulting from cancer treatment was considered a synonym for “cardiotoxicity” largely based on the description of adverse cardiac events that resulted from anthracycline administration. The Common Terminology Criteria for Adverse Events (CTCAE) formulated by the US National Cancer Institute for clinical research reporting has defined “cardiotoxicity” solely as symptomatic LV systolic dysfunction (an LV ejection fraction [LVEF] of <50%) or congestive HF.[13] This method of reporting is a very insensitive assessment of HF as a clinical condition that results from chemotherapy, especially because nearly half of patients admitted in the United States for HF actually have a normal or nearly normal LVEF.[14, 15] In addition, HF was not a codeable diagnosis in the first 3 versions of the CTCAE. Furthermore, in any trial that was not a phase 1 or 2 study, only grade 3 toxicity was reportable (symptomatic severe reduction of systolic dysfunction with an LVEF <20%) and, thus, milder cases of LV systolic dysfunction were not ascertained. Finally, rates of HF and LV systolic dysfunction generally are described best in the minority of patients who participate in clinical trials. Patients who are cancer survivors uniformly would be excluded from cardiology research trials, and cancer survivors who dropped out of an oncology trial because of toxicity generally would not have any systematic follow-up. Even for patients on clinical trials, ascertainment of late cardiotoxicity may be limited by the sometimes long interval between potential cardiotoxic exposure and the development of clinical or even subclinical findings.

Because of these reporting limitations, outside of some special populations, it has been extremely difficult to know the incidence and prevalence of LV systolic dysfunction or HF that results from chemotherapy, even anthracycline-based therapy, which is the most commonly recognized. Although not without limitations, Surveillance, Epidemiology, an End Results-Medicare linkages provide some population-based insight into the epidemiology among older cancer patients,[16, 17] and large cohort studies like the Childhood Survivor Study provide important data on childhood cancer survivors.[18] The current CTCAE (version 4) is more detailed than previous versions in its reporting of LV dysfunction (reduced LVEF <50%) and HF. However, a more careful description of cardiac toxicity is represented in data from Italy that examines cardiac toxicity in terms of HF, asymptomatic LV dysfunction, arrhythmia requiring treatment, and sudden cardiac death.[19] These data form a broader and more complete description of cardiac toxicity that should serve as a model for future research efforts.[7]

European perspective

Given the potentially long interval (up to 2 or 3 decades in some instances) between cancer therapy exposure (both chemotherapy and radiotherapy) and subsequent adverse, measurable effects on cardiac health, such adverse effects can have an important, albeit delayed effect on the prognosis of long-term survivors.[2, 4, 9, 19, 20] It has been difficult even to develop a consensus for a more specific outcome, such as LV dysfunction, including identifying the most sensitive markers and the most appropriate threshold for action. LV dysfunction may result from many anticancer drugs through different mechanisms and may manifest as declined LVEF as a final common pathway.[21] However, although LVEF remains one of the most commonly used indexes of LV systolic myocardial performance, a reduction in LVEF is considered a very late finding.[22] Currently, cardiotoxicity is commonly defined either as an LVEF reduction of greater than 10 percentage points with a final LVEF <50% or as an LVEF reduction greater than 15 percentage points with a final LVEF >50%.[23] It is unclear whether a minimal reduction of 5 percentage points with accompanying HF symptoms is enough for a diagnosis of cardiotoxicity.[24] Nevertheless, currently, a change in LVEF remains the basis for all definitions of cardiotoxicity issued by scientific societies in both Europe and the United States.[10, 25, 26] Table 1 provides examples of different definitions of cardiotoxicity used by selected studies,[34] and one of the most commonly used severity grading systems of chemotherapy-related LV dysfunction is detailed Table 2.[13, 32, 35-38]

Table 1. Previous Definitions of Cardiotoxicity Used by Selected Clinical Trialsa
ReferenceDefinition of CardiotoxicityDrugResults and Cardiotoxicity
  1. Abbreviations: AC-T, doxorubicin and cyclophosphamide followed by docetaxel; AC-TH, doxorubicin and cyclophosphamide followed by docetaxel and trastuzumab; CHF, congestive heart failure; CI, confidence interval; HR, hazard ratio; LLN, lower limit of normal; LVEF, left ventricular ejection fraction.

  2. a

    Adapted from Sawaya 2011.[34]

Tan-Chiu 2005[27]Decline LVEF by 10% to <55%TrastuzumabCardiac events at 3 y: 4.1% (AC-TH) vs 0.8% (AC-T)
Perez 2004[28]LVEF decline ≥15% compared with baseline to below the LLN (toxicity grade 2)Doxorubicin and cyclophosphamideGrade 2 toxicity, 6.6%
Suter 2004[29]Decline of LVEF ≥15 points to <50%TrastuzumabReceived trastuzumab, 6.5%; did not receive trastuzumab, 0.7% (preliminary data from 6 trials)
O'Brien 2004[30]Decline in LVEF of 20 points to >50% or at least 10 points to <50% or clinical CHFDoxorubicinDecline of LVEF, 18.8% (of which 21% had clinical CHF); clinical CHF without LVEF decline, 0.8%
Smith 2007[31]Decline in LVEF of ≥10 points from baseline to <50%Trastuzumab after adjuvant or neoadjuvant chemotherapyUnadjusted HR for the risk of an event with trastuzumab compared with observation alone: 0.64 (95% CI, 0.54-0.76; P < .0001)
Romond 2005[32]Decline of LVEF ≥16 points or <LLNDoxorubicin and cyclophosphamide followed by trastuzumabDiscontinued trastuzumab because of toxicity: 31.4%
Ryberg 2008[33]Decline of LVEF <45% or 15 points from baselineEpirubicinDeveloped cardiotoxicity: 11.4%
Table 2. Severity Grades of Chemotherapy-Related Left Ventricular Systolic Dysfunction
  1. Abbreviations: LVEF, left ventricular ejection fraction.

  2. aAdapted from National Cancer Institute Common Terminology Criteria for Adverse Events,[13] Martin 2009,[24] Eschenhagen 2011,[25] Bovelli 2010,[26] Swaya 2011,[34] and Tarantini 2012.[35]

Grade IAsymptomatic decline in LVEF of >10% from baseline evaluation
Grade IIAsymptomatic decrease in LVEF of <50% or ≥20% compared with baseline value
Grade IIIHeart failure responsive to treatment
Grade IVSevere or refractory heart failure or requiring intensive medical therapy and/or intubation
Grade VDeath related to cardiac toxicity

Epidemiology and Risk Factors

The dose relation between anthracyclines, radiation, and subsequent cardiotoxicity has been extensively reviewed by others.[20],[39-41] The exact mechanism by which anthracyclines cause cardiac injury still is not well understood but likely involves the production of free radicals, which accentuate mitochondrial and cardiomyocyte damage.[41] Separate effects mediated by cardiomyocyte topoisomerase also may be important.[42] The relation between anthracycline exposure and cardiac injury was appreciated early on[43]; and, as dosing practices have evolved, acute toxicity has become rare, but anthracycline-treated patients continue to experience an increased risk of chronic, typically dilated cardiomyopathy that may not be detected until years or even decades after exposure, especially with doses greater than 250 or 300 mg/m2.[16, 18, 39]

In a European prospective, longitudinal study of childhood cancer survivors, the most important predictor of worsening cardiac performance was total anthracycline dose.[44] Similarly, a large, retrospective Swedish experience in young patients, including children, who received chemotherapy and radiotherapy for Hodgkin lymphoma indicated that age at diagnosis (<40 years) and family history of HF predicted the development of HF and stroke at follow-up after 20 years.[45] If young age is a risk for childhood patients, then, similarly, older age is a notable risk for older women with breast cancer who are currently receiving trastuzumab in an adjuvant setting[35] and for all patients who have received anthracyclines.[46, 47] In addition to age, it also has been demonstrated that female childhood cancer survivors are at increased risk of cardiomyopathy in some studies, but that finding had borderline significance[44, 48] or was not supported in other studies.[9]

Similarly, radiotherapy-related late effects usually take years to manifest in the form of accelerated (compared with expected population rates) coronary artery and other vascular diseases, valvular dysfunction, pericardial disease, and sometimes LV dysfunction and restrictive cardiomyopathy.[5, 18, 40, 49] Although there also exists a clear dose-response relation, especially in patients with left-sided breast cancer, some reports suggest that the risk may be increased even with doses <5 Gy.[20]

Although cardiac complications associated with anthracyclines and radiotherapy are the best studied to date, other agents also have been associated with cardiotoxicity. Anthracycline-related derivatives, such as mitoxantrone, may be less likely to cause cardiomyopathy, but they are not risk-free either.[50] There are different anthracycline derivates that may reduce cardiotoxicity in cancer patients.[50] Other chemotherapy, such as high-dose alkylator therapy given as part of hematopoietic cell transplantation conditioning regimens, also have been linked to an increased risk of HF.[51] Finally, newer agents, such as trastuzumab, a monoclonal antibody to the human epidermal growth factor receptor-2 (HER2) (also called ErbB2), have been associated with LV dysfunction, including an increased risk of clinical HF, when received by patients who previously received or are currently receiving anthracycline-based regimens.[52] The mechanism of this toxicity may be related to additive stress on repair mechanisms after anthracycline administration, like a response to the negative stress of chemotherapy.[53] Increased rates of myocardial ischemia and LV dysfunction also have been observed after treatments with tyrosine kinase inhibitors, such as sorafenib[54] and sunitinib,[55] respectively.

Finally, although cancer therapy-related cardiotoxic exposures are important, many studies have indicated that conventional risk factors, such as smoking,[38] hypertension, and diabetes,[26],[42-44] remain important independent risk factors. Variation in individual risk may also be explained by underlying genetic variation. Recent studies have reported several candidate single nucleotide polymorphisms in drug metabolism pathway genes that may be associated with anthracycline-related cardiomyopathy among childhood cancer survivors,[48, 56] supplementing previous work done using in vitro assays.[57, 58] Investigators also have begun to examine the role of genetic polymorphisms in the inflammatory pathway genes in relation to heart disease after radiotherapy.[59] Because heterogeneity in phenotype has been a major barrier to replication in genetic epidemiology research,[60] a consistent, internationally accepted definition of key cardiotoxic outcomes will be important in facilitating future work in this area, including the validation of these preliminary findings.

Early Detection of Cardiac Toxicity

Given the significant impact of cardiotoxicity on prognosis for patients with cancer, earlier detection has become a primary goal for both cardiologists and oncologists. Many providers do not recognize that anthracycline-based chemotherapy is such a powerful risk factor for the development of LV dysfunction, although it is identified as a high-risk clinical indicator for the development of HF in the American College of Cardiology/American Heart Association (ACC/AHA) guidelines for the detection and treatment of HF.[61] Currently, the most frequently used modality for detecting cardiotoxicity is the periodic measurement of LV systolic function (typically the ejection fraction and sometimes fractional shortening) by either 2-dimensional echocardiography (2D ECHO) or multigated acquisition scanning (MUGA).[62] To date, however, there are no evidence-based adult guidelines for cardiotoxicity monitoring during and after anticancer therapies (Table 3).[25, 61] Evidence-based and consensus-based guidelines in pediatric oncology have been published by different national groups, although they differ in their recommendations.[64-66]

Table 3. Summary of Guidelines With Recommended Approaches for Cardiotoxicity Monitoring
Professional OrganizationReferenceRecommendationsa
  1. Abbreviations: ACC/AHA, American College of Cardiology/American Heart Association; ASCO, American Society of Clinical Oncology; CCSG, Children's Cancer Study Group (United Kingdom); COG, Children's Oncology Group (primarily United States); DCOG, Dutch Children's Oncology Group; ESC, European Society of Cardiology; ESMO, European Society of Medical Oncology; HFSA, Heart Failure Society of America.

  2. a

    The numbered guidelines with recommended approaches for cardiotoxicity monitoring are as follows: 1) All patients who receive cardiotoxic therapy are considered to be at high risk for the development of heart failure; 2) left ventricular systolic function assessment should be performed at baseline and at some subsequent interval (no specific time point); 3) angiotensin-converting enzyme inhibitors and beta blockers are recommended for treatment; 4) angiotensin-converting enzyme inhibitors are recommended for prevention; 5) adult survivors; and 6) pediatric cancer survivors.

ACC/AHAHunt 2009[61]1, 2, 5
HFSALindenfeld 2010[63]1, 2, 5, 6
ASCOCarver 2007[10]2, 5, 6
ESCEschenhagen 2011[2][5]2, 5, 6
ESMOCurigliano 2012[21]1, 2, 3, 4, 5, 6
COGLandier 2004[64]1, 2, 6
CCSGSkinner 2006[65]1, 2, 6
DCOGSieswerda 2012[66]1, 2, 3, 6

European perspective

Recently published European Society of Medical Oncology clinical practice guidelines specify more details, although these are not necessarily based on high levels of evidence.[21] For example, these guidelines recommend assessing cardiac function 4 years and 10 years after anthracycline therapy in patients who were treated at age <15 years (evidence level, III [with level I considered the most rigorous]; recommendation grade, B [with A considered superior]), or even at age >15 years but with a cumulative dose of doxorubicin >240 mg/m2 or epirubicin >360 mg/m2 (evidence level III B). If LV dysfunction is detected by imaging or cardiac biomarkers (evidence level III C), then the receipt of angiotensin-converting enzyme inhibitors (ACE-Is), angiotensin receptor blockers (ARBs), or perhaps beta-blockers may limit progression to symptomatic HF. In fact, in the European guidelines, the aggressive medical treatment of those patients, even if they are asymptomatic, who demonstrate LV dysfunction after anthracycline therapy “is mandatory,” especially if long-term survival is expected.

Recommendations also exist for trastuzumab-treated patients to receive serial evaluation of LVEF every 3 months.[24, 67] However, not all patients who receive trastuzumab have LVEF monitoring as frequently as suggested by the guidelines; conversely, a considerable percentage of patients do not have favorable outcomes even with close cardiac monitoring.[68, 69] This reflects the reality that LVEF measurement is a relatively insensitive tool for detecting cardiotoxicity at an early stage, largely because no considerable change in LVEF occurs until a critical amount of myocardial damage has taken place, and it only comes to the forefront after compensatory mechanisms are exhausted. Serial measurements of troponin I can provide complimentary information in this setting but are not routinely done or recommended.[22, 70] Therefore, evidence of a decrease in LVEF precludes any chance of preventing the development of cardiotoxicity.[71] Conversely, a normal LVEF does not exclude the possibility of later cardiac deterioration. In addition, the measurement of LVEF presents several challenges related to image quality, assumption of LV geometry, load dependency, and expertise. Novel echo imaging techniques, like contrast echocardiography and real-time, 3-dimensional echocardiography, have emerged that allow for an improvement in the accuracy of calculating LVEF.[72] Small studies examining tissue Doppler and strain rate imaging appear promising for detecting early subclinical changes in cardiac performance that anticipate a decrease in conventional LVEF, even if long-term data on large populations confirming the clinical relevance of such changes are not yet available.[73, 74] The advantages are that these techniques do not require a separate examination and that the technology is available on most current machines. The disadvantage is that data analysis is currently off-line, very time-consuming, and still depends on the quality of the acoustic windows. Currently, although they are promising, these new echo imaging techniques cannot yet be recommended as part of routine assessment of cardiac function among cancer survivors.[75] Alternative imaging modalities and serum-based biomarkers are discussed further below.

American perspective

Currently, the measurement of cardiac biomarkers (troponin I and troponin T, B-type natriuretic peptide [BNP], and N-terminal pro-BNP) are becoming more widely used to detect cardiac toxicity among patients actively receiving cancer therapy as well as short-term survivors.[77-79] However, there is not a clear, consistent recommendation, because there are many variations in the techniques to measure each assay, and the optimal timing of measurement and interpretation in relation to chemotherapy have not been established. Furthermore, less established data exist to guide choices on how to respond to an abnormal value when a patient is receiving anthracyclines[79] and perhaps trastuzumab. Data for other drug regimens are even less well studied. Notwithstanding these limitations, the early detection of cardiac toxicity is possible and can facilitate earlier intervention, which may attenuate cardiac injury even if it is associated with anthracyclines.[22] Again, this is a critical difference from the long-held belief that the trajectory of anthracycline-related cardiac injury is not modifiable.

Cardiac imaging with 2D ECHO, MUGA, or magnetic resonance imaging (MRI) has been used for many years to detect cardiac toxicity. There is considerable evidence suggesting that 2D ECHO or MUGA are adequate screening tests for clinically detectable LV dysfunction, but they are not very sensitive for detecting earlier subclinical injury.[22] There have been no recent developments regarding MUGA and the utility of MUGA-based results for early detection. MRI has gained importance recently, primarily because of the accuracy of its LV measurements and its ability to characterize the myocardium as either normal, recently injured, or permanently damaged, which is superior using MRI compared with 2D ECHO.[80] Compared with 2D ECHO, MRI has disadvantages in terms of more limited availability and radiologic expertise as well as a potential for its accompanying contrast to worsen any renal insufficiency. MRI also is contraindicated in patients who have implanted mechanical devices. Finally, young children undergoing MRI often require anesthesia. However, these issues aside, MRI likely will be a valuable resource for monitoring and detecting cardiac damage for years to come.

Which ever imaging technique is used, it is best to repeat serial measurements with the same tool to minimize intermeasurement variability, although interobserver variability remains a potential issue. However, currently, there is no international consensus on recommendations regarding the expected or required interval for testing, although efforts are being made to harmonize recommendations for pediatric survivors,[64-66] and some consensus guidelines have been issued for hematopoietic cell transplantation survivors.[81] This has great importance, because these all of imaging studies are expensive, much more so than troponins and/or BNPs, and cannot be routinely repeated without clear benefit and necessity. One recommendation by the US Food and Drug Administration is to follow patients who are receiving long-term trastuzumab with imaging studies every 3 months, although this is not necessarily practical or economically feasible.[82] Issues related to an optimal screening schedule and relative cost-utility remain an understudied area but may very well differ in different health care delivery systems.

Today, strong data indicate that troponin detects anticancer drug induced-cardiotoxicity in its earliest phase, long before any reduction in LVEF has occurred.[77] Troponin is now the gold-standard biomarker for myocardial injury from any cause.[83] Its evaluation during high-dose chemotherapy allows for the early identification of patients who are at risk of developing cardiac dysfunction, the stratification of risk for cardiac events after chemotherapy, and the opportunity for a preventive therapy in selected high-risk patients.[19, 84, 85] Indeed, it has been demonstrated that prophylactic treatment with enalapril in adult patients who have an early increase in troponin after chemotherapy prevents cardiac dysfunction and associated cardiac events in patients who receive high-dose anthracyclines (Fig. 1).[85] More recently, increases in troponin levels have been observed in patients who received standard anthracycline doses and in patients who received newer antitumor agents.[74, 77, 78, 86] Moreover, in trastuzumab-treated patients, by identifying myocardial cell necrosis, troponin may help us to distinguish between reversible and irreversible cardiac injury.[70]

Figure 1.

These charts illustrate the protective effect of angiotensin-converting enzyme inhibitors (ACE-I) on cardiac outcomes in patients undergoing chemotherapy (predominately anthracycline-based) who were randomized to receive ACE-I after an early increase in troponin I levels (ACE-I group, n = 56; controls, n = 58). LV dysfunction indicates left ventricular dysfunction (defined as a left ventricular ejection fraction <50% plus a decrease >10% from the starting value). Cardiac events include sudden death, any cardiac death, symptomatic heart failure, LV dysfunction, and serious rhythm disturbance requiring treatment. PM indicates pacemaker. (Adapted from Cardinale D, Colombo A, Sandri MT et al Prevention of high-dose chemotherapy-induced cardiotoxicity in high-risk patients by angiotensin-converting enzyme inhibition. Circulation. 2006;114:2474-2481.[85]).

In the published studies, there is wide variation in the sampling protocols for the measurement of troponins, with increased levels detected at various time intervals after chemotherapy, possibly because of diverse troponin release kinetics in response to cardiotoxic injury with different agents. Thus, currently, most research surveillance protocols have deemed it necessary to collect blood samples several times to document a potential increase in troponin levels.[77] This represents a possible limitation for using the marker in clinical practice; however, the measurement of troponin only immediately before and immediately after each cycle of cancer therapy may be sufficient, and such a protocol would be more easily transferable from the research setting to actual clinical practice.[70] This biomarker-based protocol is likely to be cost-effective when negative values allow for the exclusion of most patients from a long-term monitoring program based on more expensive imaging methods. However, the standardization of routine troponin measurement in the clinical setting to maximize single-time-point assay sensitivity and specificity is needed and should be an important focus for future research. Furthermore, additional vascular biomarkers, such as endothelial growth factors, may be important in identifying those at risk for vascular toxicity with newer antiangiogenic-based treatment, although, currently, those are speculative.[87]

Finally, although our focus is on chemotherapy-related HF, the long-term effects of radiation therapy on the heart can be quite profound.[88] Typically, there is a long period of latency and then the complexity is significant. Peripheral vascular and coronary artery disease are major issues that are classically asymptomatic until a major clinical event occurs. Consideration of appropriate screening tests in high-risk individuals is imperative. For example, carotid ultrasound screening is highly appropriate for patients who have received mantle or head and neck radiation, whereas cardiac stress testing is important in survivors who received radiation to the mediastinum. It also is important to factor in radiation scatter to these regions from radiotherapy primarily directed toward the spine and upper abdomen. Cardiac valvular structures also are affected by mediastinal radiation, and 2D ECHO is the most appropriate screening test for this condition.

Treatment of Cardiotoxicity

American perspective

The treatment of cardiac toxicity is greatly influenced by the comorbidities that exist in a given patient and the context in which that damage is detected. For example, if a patient is acutely ill from a hematologic-based malignancy, then there may be transient LV dysfunction from a variety of causes, including stress cardiomyopathy or sepsis; and, after a period of stabilization, the patient may be able to resume cardiotoxic chemotherapy if that is necessary for the optimal treatment of their cancer. Alternatively, a patient who had no prior cardiac disease, received anthracycline-based therapy 4 years ago, and now has severe LV dysfunction may not currently be considered for chemotherapy if cancer treatment is needed. The general principles that apply to the treatment of LV dysfunction in all patients are equally important in cancer survivors: 1) dietary adjustments, especially sodium limitation; 2) carefully monitored exercise and weight management; 3) maximally tolerated doses of renin-angiotensin system inhibitors (ACE-Is, ARBs, and beta-blockers); 4) selective use of aldosterone antagonism; 5) appropriate use of implantable cardiac defibrillators or biventricular pacing; and 6) other prevention-based therapies (aspirin, statins, and avoidance of alcohol/smoking).[89] Nevertheless, data supporting the efficacy of these interventions are limited, specifically among cancer survivors and especially among childhood cancer survivors.

European perspective

There are no well established recommendations for the treatment of cancer patients who develop HF as a result of anticancer treatment. Typically, these patients have been excluded from large randomized trials evaluating the effectiveness of novel HF therapies, and the use of ACE-I and beta-blockers in this particular clinical setting remains a matter of debate. One of the more challenging features of this form of cardiac dysfunction is that it usually remains asymptomatic for a very long time.[90] Many American and European authors have recommended screening programs to look for overt HF, as highlighted by Yoon et al,[91] because many cancer patients who develop cardiac dysfunction do not appear to be receiving optimal treatment and often are treated only if symptomatic. This is probably because there are special concerns related to the use of ACE-I and beta-blockers, even if these medications may be highly effective in treating therapy-related HF, possibly because cancer patients are considered frail, and the tendency is not to treat them aggressively. A recently published prospective study that included the largest population of anthracycline-related cardiomyopathy patients to date (N = 201 adult patients, including many still actively receiving anticancer therapy) demonstrated that the time elapsed from the end of chemotherapy to the start of HF therapy (the time to treatment) with ACE-I and with beta-blockers, when tolerated, is a crucial variable for the recovery of cardiac dysfunction.[92] Indeed, the likelihood of obtaining complete LVEF recovery was greater for patients who had treatment initiated within 2 months from the end of chemotherapy. After 2 months, this proportion progressively decreased, and no complete LVEF recovery was observed among those who had therapy initiated only after 6 months (Fig. 2). Notably, in that study, the clinical benefit was more evident in asymptomatic patients than in symptomatic patients. Therefore, monitoring of cardiotoxicity exclusively based on symptom evaluation may miss the opportunity to detect early cardiac injury that is still in a reversible stage. At least among adult cancer patients, this emphasizes the crucial importance of early detection of cardiotoxicity and suggests that an aggressive pharmacologic approach, based on ACE-Is, possibly in combination with beta-blockers, should always be considered, and attempted in all cases of anthracycline-related cardiomyopathy (Fig. 3).[93] Data on the efficacy of ACE-Is among childhood cancer survivors are much more limited, and no recommendations can be made.[90] The length of therapy required once a cancer patient has developed LV dysfunction and HF remains uncertain, but at least some data suggest that treatment should be long-term.[94]

Figure 2.

The reversibility of left ventricular dysfunction (both symptomatic and asymptomatic) in patients undergoing treatment with anthracyclines depends critically on the timing of the initiation of cardioprotection therapy with beta-blockers and angiotensin-converting enzyme inhibitors. If initiation of therapy is delayed longer than 6 months since the time of anthracycline exposure, then the likelihood that patients will respond to therapy is greatly reduced. (Adapted from Cardinale D, Colombo A, Lamantia G et al Anthracycline-induced cardiomyopathy: clinical relevance and response to pharmacologic therapy. J Am Coll Cardiol. 2010;55:213-220.[92]).

Figure 3.

A suggested algorithm for screening for cardiotoxicity during and after chemotherapy with anthracyclines is illustrated. ECHO indicates echocardiography; CT, chemotherapy; TnI, troponin I; LVD, left ventricular dysfunction. (Adapted from Cardinale D. Cardiac dysfunction after cancer treatment. In: Bonow RO, Zipes DP, Libby P, eds. Braunwald's Heart Disease: A Textbook of Cardiovascular Medicine. Ninth ed. Philadelphia, PA: Saunders Elsevier; 2012.[93]).


American and European perspectives

Because cardiomyopathy may occur even many months or years later, it is crucial to consider cardiac issues for a long time, because it is known that these patients are at high risk for subsequent serious events. True prevention of late cardiotoxicity in cancer patients begins before chemotherapy administration: a baseline assessment of cardiovascular health and effective treatment of cardiovascular risk factors is needed to prevent most late cardiac toxicities. Aspirin, control of hypertension and dyslipidemia, and tobacco cessation all are interventions that should be aggressively pursued where appropriate.[16, 49, 95, 96]

Strategies for primary prevention also have been proposed, balancing the need to preserve therapeutic efficacy while minimizing adverse late effects. Refinements in treatment protocols and radiotherapy delivery have led to decreased rates of cardiovascular disease among select survivor cohorts over time.[97] Continued improvements in technology (eg the advent of intensity-modulated and proton radiotherapy) have the potential to reduce radiation scatter to critical organs even further for some patients.[40] Effective primary strategies also exist to reduce anthracycline-related cardiotoxicity (at least among adults), including the use of less cardiotoxic anthracycline derivatives, prolonging infusion time, and concurrent administration of a cardioprotectant, such as dexrazoxane.[27, 50, 98-101] Summary risk estimates based on randomized trial data (mostly adult patients with breast cancer) suggest that dexrazoxane is associated with a significantly decreased risk of both clinical and subclinical HF without affecting tumor response rates and without being associated with increased noncardiac side effects compared with conventional anthracyclines.[50, 100, 101] Currently, it is approved for use by the US Food and Drug Administration only among women with metastatic breast cancer who have received 300 mg/m2 of doxorubicin and who may benefit from further anthracycline-based therapy.[102] The American Society of Clinical Oncology similarly recommends consideration of dexrazoxane in adults (with any histology) who have already received 300 mg/m2 of anthracyclines.[103] Data among children are limited,[104-106] and concerns regarding the possible association of dexrazoxane with an increased risk of second cancers[101, 107, 108] have limited more widespread use among children, and the European Medicines Agency has specifically recommended that children not receive dexrazoxane.[109]


All patients with cancer who are treated with potentially cardiotoxic chemotherapy represent a high-risk group for the development of HF. Adults, at least, should be treated with ACE-Is and/or beta-blockers, in accordance with the 2009 AHA/ACC guidelines for the treatment of stage A congestive HF, especially when and if the LVEF is reduced to <55%.[61, 110] Increased collaboration between cardiologists and oncologists is needed to determine the best treatment combinations and the best preventive strategies that will improve the cardiac health of our patients. To foster this collaboration, new societies based in both Europe and the United States, such as the Italian Association of Cardio-Oncology (AICO) ( and the International CardiOncology Society (ICOS) ( [accessed January 22, 2013]), have been formed that bring together interested researchers and clinicians from both fields. Finally, childhood cancer survivors need dedicated research, because findings relevant to adult-onset cancer survivors may not necessarily apply to this population.

Key Gaps

Although much progress has been made in better understanding and treating cardiac disease and toxicity associated with cancer therapy, multiple issues remain. These include: 1) the need to develop internationally accepted and uniform definitions of cardiac toxicity to make results across studies more comparable; 2) further study of potential interactions between known treatment risk factors and novel agents as well as a better understanding of possible genetic influences on risk; 3) refinement of screening tests, both serum-based and imaging-based biomarkers, and the optimal timing and interval between tests; 4) additional clinical studies that determine the best treatment strategies both for survivors with asymptomatic LV dysfunction and for those with symptomatic HF; and 5) continued development of strategies that allow for delivery of effective cancer therapy while minimizing unintended toxicity to the heart and the cardiovascular system as a whole.


This supplement was sponsored by the National Cancer Research Centre Istituto Tumori “Giovanni Paolo II” Bari (Italy) through the Italian Ministry of Health-funded research project “Multidimensional assessment of long-term cancer survivors including discovery of genetic bases of susceptibility, depressive stage, prevention of affective disorders,” and through intramural funding of the American Cancer Society's Behavioral Research Center.


The authors made no disclosures.