Dr. Alberto Esteban Fernández
The management of cancer patients who develop cardiotoxicity is an important issue for oncologists and cardiologists; however, there is an important lack of evidence about its management and treatment. The current management is based on small studies, clinical experience and extrapolation from the treatment of other similar diseases (i.e., dilated cardiomyopathy). The recent publication of the European Society of Cardiology Position Paper about cancer treatments and cardiovascular toxicity is the starting point for developing clinical trials and cardio-oncology teams to manage this increasing problem, which also has an important prognosis impact.
Cancer patients are commonly treated with chemotherapy, which is related to myocardial injury and also left ventricular dysfunction (LVD). Classically, we consider two different chemotherapy drugs depending on the reversibility of the myocardial injury, although this classification is not very useful in clinical practice. Cancer patients habitually receive a “cocktail” of chemotherapy drugs, so it is sometimes difficult to determine the specific mechanism. Moreover, we know that cancer itself and also radiotherapy may induce myocardial injury.
Furthermore, there is a huge lack of evidence on cardiotoxicity management, and the clinical practice is based on small studies and medical experience. To help in cardiotoxicity management, in 2012 the European Society for Medical Oncology published the Clinical Practice Guidelines for cardiotoxicity management . Additionally, in 2016, the European Society of Cardiology published a position paper about cancer treatments and cardiovascular toxicity .
Therefore, cardiotoxicity is a very relevant issue in clinical practice, with little evidence based on clinical trials. The publication of these important clinical guidelines, supported by important scientific societies, is the starting point for developing strong evidence in cardiotoxicity management, with a close relationship between cardiologists and oncologists.
Cardiotoxicity is defined as any heart injury (functional or structural) related to cancer treatment, taking into consideration chemotherapy, radiotherapy and cancer itself. Although we are going to focus on myocardial damage, cardiotoxicity also produces other heart affections, including pericardial, valvular or coronary artery diseases. It is usual that cardiotoxicity equates with LVD, but not always, and there is not always an exact agreement between oncologists and cardiologists.
The ESC European Guidelines for the diagnosis and treatment of heart failure (HF), published in 2016, classified HF into: reduced ejection fraction (left ventricular ejection fraction [LVEF]<40%), mid-range ejection fraction (LVEF 40-49%) and preserved ejection fraction (LVEF >50%) . Although we do not have large trials and registries, most cardiotoxicity patients have HF with reduced ejection fraction, but it is not infrequent to encounter patients with mid-range ejection fraction due to early detection. Probably, there are also patients with myocardial injury, HF and preserved ejection fraction, but there are no data about its incidence.
There is also a fourth group of patients with recovered ejection fraction due to treatment and/or chemotherapy discontinuation. Although recent studies have shown lower mortality and hospitalisation in patients with recovered ejection fraction (not specifically in cancer patients) compared with those with HF and reduced or preserved ejection fraction, it seems reasonable to manage and follow them up differently from patients without LVD .
For oncologists, the current classification for cardiotoxicity is defined by theCardiac Review and Evaluation Committee on trastuzumab-associated cardiotoxicity and the ESMO Clinical Practice Guidelines. Cardiotoxicity is defined as “a decrease of LVEF by 5% or more to less than 55% in the presence of symptoms of HF or an asymptomatic decrease in LVEF by 10% or more to less than 55%” .
Because of the slightly different classifications, it is important to homogenise protocols in each hospital [] and also create a cardio-oncology team to improve patient management.
The different classification and knowledge of the multiple factors involved in cardiotoxicity explain the variability in cardiotoxicity incidence in different studies []. The estimated incidence of cardiotoxicity is between <1% with paclitaxel and 48% with 700 mg/m² of doxorubicin. Also, in some cases, such as with trastuzumab, it varies between 1.7% and 20.1%, rising to 28% when anthracyclines are used in combination with cyclophosphamide.
Myocardial injury due to the direct or indirect action of chemotherapy has a particular mechanism for each drug. Also, there are other factors involved, such as drug accumulative dose, presentation, preparation and route of administration of the drug, and the effect of the sequential and/or concomitant use of drugs. The mechanism of each drug producing cardiotoxicity is reviewed in Table 1. Also, these have been widely reviewed in multiple papers [2,7].
Table 1. Mechanism of action of some drugs used in cancer treatment.
Doxorubicin (Adriamycin), epirubicin
Breast, lymphoma, leukaemia, lung
Myocardial injury due to oxidative stress producing irreversible damage of the cardiomyocyte membrane.
Myocardial ischaemia, endothelial cell injury, thrombosis.
Arterial vasospasm, endothelial cell injury, thrombosis, myocardial ischaemia.
Trastuzumab (HER2 inhibitor)
Changes in contractile proteins and mitochondria.
Bevacizumab (anti-VEGF, anti-angiogenesis agent)
Breast, colon, lung
Relevant arterial hypertension, endothelial cell injury, venous or arterial thrombosis.
Lymphoma, leukaemia, myeloma
Myocardial ischaemia, endothelial cell injury.
Toxicity related to the volume of administration, relevant arterial hypertension.
Vinca alkaloids (vincristine, vinblastine)
Myocardial ischaemia, relevant arterial hypertension.
Myocardial ischaemia, bradycardia.
Small molecule tyrosine kinase inhibitors
Relevant arterial hypertension.
Sunitinib (anti-angiogenesis agent)
Relevant arterial hypertension, venous thrombosis.
In 2005, Ewer and Lippman published a classification considering the presence of structural abnormalities and the potential reversibility of myocardial injury due to chemotherapy: type 1 (irreversible injury or structural abnormalities, i.e., anthracyclines) and type 2 (reversible injury and no structural abnormalities, i.e.,trastuzumab). Although this is a very useful classification for each individual drug, it is really difficult to determine the exact mechanism of cardiotoxicity in clinical practice with patients taking multiple drugs.
The time for cardiotoxicity onset is different for each drug and patient, depending on personal susceptibility and drug characteristics. It sometimes appears immediately after exposure, but in other cases clinical manifestations may appear even several years after cancer treatment. Although it is necessary to prove the relationship between treatment exposure and LVD to make the diagnosis, something that is not always easy, sometimes the diagnosis comes after ruling out other LVD causes.
The presence of cancer has been seen to produce myocardial injury independent of the treatment. For instance, in patients with breast cancer, before starting the treatment, a significant increase in cardiovascular peptides has been seen, and some of them, such as copeptin, have been related to all-cause mortality .
Radiotherapy also produces cardiac damage, including myocardial injury and affection of other structures, such as valves, which may also progress to HF. The damage depends on dose and thorax radiation, although it is sometimes difficult to evaluate due to the synergistic effect with chemotherapy and the long delay between radiotherapy exposition and cardiotoxicity in some cases .
Finally, the existence of cardiovascular risk factors in cancer patients can accelerate the development of cardiovascular disease, including HF. Moreover, cancer survivors have a greater risk for cardiovascular disease than the general population, due to the cancer and also the chemotherapy, especially with some drugs such as anthracyclines . For these reasons, previous cancer and treatments should be registered as risk factors in the clinical history.
Although patients with cardiotoxicity are sometimes asymptomatic and the diagnosis is made on a routine echo, classic symptoms and signs of HF can appear, and they should be explored routinely in all cancer patients for an early diagnosis. However, clinical diagnosis is sometimes difficult, because it is not uncommon for cancer patients to refer to tiredness, asthenia, fragility or even dyspnoea for non-cardiologic reasons.
In patients with HF, using the NYHA functional class (I-IV) to evaluate the clinical status is recommended, as is the AHA/ACC classification (A-D) .
As in other causes of HF, a completely normal ECG habitually suggests normal LVEF, so it is important to assess an ECG before and periodically during the treatment. The ECG abnormalities in cardiotoxicity patients may include arrhythmias, QT prolongation, bundle branch block, ischaemia or resting tachycardia.
Nowadays, due to its availability and reproducibility, echocardiography is the gold standard for heart function assessment in cancer patients before, during and after the treatment. The correct assessment of heart function is contained in the most recent recommendations for cardiac chamber quantification by echocardiography in adults published in 2015 by the American Society of Echocardiography and the European Association of Cardiovascular Imaging, and also in the ESMO Clinical Practice Guidelines and the ESC Position Paperabout cancer treatments and cardiovascular toxicity. In these documents, not only is LVEF determination recommended to evaluate heart function, but also other parameters such as diastolic function, left ventricular filling pressures, right ventricular function or ventricular diameters .
Moreover, global systolic longitudinal myocardial strains, with the limitation of reproducibility and availability, have recently been shown to predict LVD . For this reason, a systolic myocardial strain should be performed in echo studies (if available) to detect cardiotoxicity early.
In order to avoid inter- and intra-observer variability, which can interfere with clinical decisions, it is necessary to choose the same cardiologists and equipment in echo studies.
Nuclear cardiac imaging and cardiac magnetic resonance (CMR) are excellent techniques to assess cardiac function. CMR is especially indicated to study the cause of LVD, right ventricular dysfunction and myocardial fibrosis with late gadolinium enhancement. Also, CMR is useful to determine the structural abnormalities present in type 1 cardiotoxicity.
The main limitation of these techniques is their low availability, although they can be as good as echo in the follow-up .
Natriuretic peptides arethe most usual cardiac biomarkers used in HF, and also in cardiotoxicity []. There is no evidence to guide cancer therapy according to NT-proBNP, including starting or stopping chemotherapy. Nevertheless, the ESC Position Paper on cancer treatments and cardiovascular toxicity recommends NT-proBNP to detect early cardiac injury and as a marker added to the rest of the tests in the follow-up.
On the other hand, troponin has been demonstrated to be a good marker of early myocardial injury, especially in patients treated with trastuzumab and/or anthracyclines, with a preference for high-sensitivity troponin [14,17].
Cardiac biomarkers should be checked before and during the follow-up, although the best schedule to repeat the tests periodically is unknown, and this should be adapted to each hospital and cancer treatment.
The treatment of LVD related to chemotherapy is the same as for any LVD, as developed in the ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure published in 2016 []. In cardiotoxicity, it seems reasonable to start treatment “the earlier the better”, determine the target doses and plan titration. However, in these patients, we may have controversial information due to the patient’s poor general status, asthenia or chemotherapy side effects, which can make titration more difficult.
There are only small studies evaluating the standard treatment in LVD related to chemotherapy, with a beneficial effect of traditional treatment in recovering LVEF. At the Spanish Congress of Cardiology in 2015, our group presented a small study on 50 patients with LVD related to chemotherapy and treated with specific cardiovascular drugs, with 62% of patients recovering ejection fraction in a mean time of 14 (3-41) months.
In symptomatic patients with LVD, an ACE-I (or ARB if there is a contraindication or intolerance) and a beta-blocker are recommended, and also an MRA if symptoms persist. Also, we know that an angiotensin receptor-neprilysin inhibitor (ARNI) (sacubitril/valsartan) should be considered instead of an ACE-I (or ARB) if symptoms persist and also ivabradine in symptomatic patients with sinus rhythm and heart rate above 70 bpm. In addition, diuretics should be considered in case of congestion or risk of it, and regular aerobic exercise should be recommended according to the general condition of the patient.
In asymptomatic patients, in the SOLVD trial only enalapril demonstrated a specific benefit slowing or reversing left ventricular dilatation, with no specific evidence for the rest of the drugs. Despite the lack of evidence, the ESC Position Paper about cancer treatments and cardiovascular toxicity recommended treatment with an ACE-I (or ARB if there was a contraindication or intolerance) and a beta-blocker in patients with asymptomatic LVD, or in case LVEF decreases >10% of the lower normal limit to prevent HF or LVD development because of the high risk of cardiotoxicity.
The necessity of devices, especially an implantable cardioverter defibrillator (ICD) and/or cardiac resynchronisation therapy (CRT) should be considered according to the criteria in the guidelines, although we have to consider the global prognosis and cancer cure probability, and also the reversibility of LVD. In patients with advanced HF, a cancer cure and life expectancy of at least more than one year, heart transplant and ventricular assist devices should be considered as in other HF patients. The recent results of the DANISH trial may also change the perspective of ICD in LVD as it relates to chemotherapy patients.
It is difficult to decide whether is necessary to stop chemotherapy completely in patients who develop cardiotoxicity. The prognostic impact of changing, reducing or stopping cancer treatment should be carefully considered by the cardio-oncology team, also taking into account the prognostic impact of LVD. In these cases, the decision must be taken considering not only the oncologic parameters (prognosis, alternatives, doses, chemotherapy schedule, etc.), but also the cardiologic ones (HF status, cardiovascular risk factors, LVEF, cardiovascular drugs, etc.).
Finally, there is a lack of evidence about the management of patients with recovered ejection fraction. The main problems are not only the importance of continuing the cardiovascular treatment, but also the cancer status or further necessity of new chemotherapy in the follow-up. It seems reasonable to maintain the treatment while the patient may need more oncological treatments. After a complete cure (although cancer is sometimes a chronic disease and the aim is not to cure), some experts recommend maintaining the treatment for at least one year after normalisation and then discontinuing it, although others think that the treatment should be prolonged indefinitely if there is no contraindication.
The prognosis of cardiotoxicity patients depends on cancer mortality, LVD and HF severity, and general clinical status. HF is known to be a disease with a poor prognosis, with 50% of patients dying in the five years following the diagnosis, while cancer prognosis depends on each particular cancer. When a decision is taken to stop chemotherapy due to cardiotoxicity, it is important to consider the whole prognostic situation and the alternatives.
As has been previously mentioned, the prognosis of patients with recovered ejection fraction is better than those with HF and reduced or preserved ejection fraction, although there are other factors that should be taken into account, such as cancer healing and the necessity of new cancer or cardiovascular treatments during the follow-up.
First, cardiotoxicity prevention starts with a good assessment of cardiovascular risk and predisposing factors (Table 2). We do not have specific scores to determine global cardiovascular risk, including past cancer or previous chemotherapy (especially anthracyclines) , but they should be considered as risk factors for cardiotoxicity and cardiovascular disease. Therefore, patients with several risk factors, especially with a poor control, and/or previous cardiovascular disease, should be managed in the same way as high-risk patients.
Table 2. Predisposing and risk factors for cardiovascular disease and cardiotoxicity.
Risk Factors or Predisposing factors
Current myocardial disease
Previous heart failure and/or LVD
Any cardiomyopathy (i.e., hypertrophic) or significant cardiopathy (i.e., hypertensive)
Coronary artery disease
Previous cardiotoxic cancer treatment
Previous anthracycline use
Previous thoracic radiotherapy
Cardiovascular and lifestyle risk factors
Smoking, arterial hypertension, obesity, etc.
Second, it is important to assess basal heart function before starting the treatment, considering not only LVEF (as is mentioned in “Diagnosis”). If some relevant findings are detected in basal heart structure or function, such as LVEF in the lower limit of normal values, the risk of cardiotoxicity is potentially higher.
The role of cardiac biomarkers is more controversial. It is really unknown whether patients with elevated basal cardiac biomarkers (NT-proBNP and/or troponin) have more risk of cardiotoxicity, although they probably have a worse prognosis. Only one trial showed that cardioprotective treatment is beneficial in patients treated with anthracyclines when troponin increases during the treatment .
The most difficult point in clinical practice is to decide what to do if any of these determinations is abnormal. In all cases, it is necessary to carry out cardio-oncology protocols to determine the periodicity of repeating the tests during the follow-up and the management of these abnormalities. In any case, they should be adapted according to each cancer and treatment, patient characteristics and hospital peculiarity. Also, some publications suggest several good algorithms and protocols to manage this situation .
The main problems are the poor evidence in cardiotoxicity prevention and the controversial results in small trials. In the OVERCOME trial, treatment with an ACE-I and a beta-blocker prevented cardiotoxicity in patients with haematologic cancer. However, neither the MANTICORE-101 trial nor the PRADA trial showed a preventive benefit of ACE-I (or ARB) or a beta-blocker in anthracycline cardiotoxicity. In the recent ESC Congress in Rome, Tomescu et al presented a small trial with 60 patients with breast cancer treated with anthracyclines for six months, showing a potential benefit of nebivolol vs. placebo in preventing LVD.
In conclusion, in high-risk patients, here are some recommendations to prevent cardiotoxicity:
Due to the complex management of these patients, it seems important to give a real boost to cardio-oncology teams, with specialists with specific skills and dedication to both cardiology and oncology. The cardio-oncology specialist should look after patients before, during and after the treatment, and also for several years after a cancer cure.
The specific training for the cardio-oncology specialist makes them able to face some different patient situations during the cancer pathway:
Finally, here are some key points to be studied in the future:
Notes to Editor
Dr Alberto Esteban Fernández, MD
Heart Failure Unit, Cardiology Department, San Carlos University Hospital, Madrid, Spain
Address for correspondence:
Dr Alberto Esteban Fernández, Heart Failure Unit, Cardiology Department, San Carlos University Hospital, Calle Profesor Martín Lagos s/n, 28040 Madrid, Spain.
The author has no conflicts of interest to declare.
Our mission: To reduce the burden of cardiovascular disease
© 2018 European Society of Cardiology. All rights reserved