Risk and prevention of anthracycline cardiotoxicity

INTRODUCTION — The anthracyclines (doxorubicin [liposomal and unencapsulated], daunorubicin, idarubicin, epirubicin) and the anthraquinone mitoxantrone (referred to hereafter as "anthracyclines") are chemotherapeutic agents that cause cardiotoxicity.

Most commonly, anthracycline toxicity manifests as a temporary or permanent reduction in left ventricular (LV) function. Among cancer survivors who develop symptoms of heart failure (HF) after anthracycline administration, cardiovascular mortality is a leading cause of death [1]. While anthracyclines are associated with an increased risk of cardiomyopathy, they are also important components of many chemotherapy regimens.

The underlying mechanisms, risk factors, and preventive management of anthracycline-induced cardiotoxicity will be reviewed here. The clinical manifestations, diagnosis, and treatment of anthracycline-induced cardiotoxicity and cardiovascular complications of other systemic anticancer agents are discussed separately:

●(See "Clinical manifestations, diagnosis, and treatment of anthracycline-induced cardiotoxicity".)

●(See "Cardiotoxicity of cancer chemotherapy agents other than anthracyclines, HER2-targeted agents, and fluoropyrimidines".)

●(See "Cardiotoxicity of trastuzumab and other HER2-targeted agents".)

●(See "Fluoropyrimidine-associated cardiotoxicity: Incidence, clinical manifestations, mechanisms, and management".)

●(See "Toxicities associated with immune checkpoint inhibitors", section on 'Cardiovascular toxicity'.)

GOALS OF PREVENTIVE MANAGEMENT — In defined clinical scenarios, the benefits of anthracyclines outweigh the risks of cardiotoxicity in published regimens for potentially curative cancer treatment. As an example, among patients receiving up to six cycles of doxorubicin plus bleomycin, vinblastine, and dacarbazine for Hodgkin lymphoma (which typically contains <400 mg/m² doxorubicin), the risk of cardiac dysfunction is ≤2 percent in the absence of additional risk factors. (See "Initial treatment of advanced (stage III-IV) classic Hodgkin lymphoma", section on 'Cardiac and pulmonary toxicity'.)

Thus, in patients at average risk for anthracycline-induced cardiotoxicity who will undergo anthracycline-based chemotherapy, the goals of preventive management are to mitigate the risk of cardiac injury during therapy and to detect any early signs of cardiotoxicity. In a potentially curative setting, the thresholds for toxicity are well known and the testing for toxicity is readily available.

In patients who have a high risk of anthracycline-induced cardiotoxicity or who have preexisting LV dysfunction, an additional goal of preventive management is to evaluate the risks and benefits of anthracycline-based therapy and the availability of equally effective non-anthracycline-based regimens to inform the choice between anthracycline-based and alternative chemotherapy regimens, when available.

MECHANISMS AND TIME COURSE OF CARDIOTOXICITY — Anthracyclines may affect cardiac function through mechanisms that include reactive oxygen species formation, induction of apoptosis, deoxyribonucleic acid (DNA) damage through interaction with topoisomerase II, and inhibition of protein synthesis [2]. Other potentially contributory mechanisms are mitochondrial iron accumulation [3] and dysregulation of cardiomyocyte autophagy [4]. Mitoxantrone has traditionally been classified as an anthracycline, but its cardiotoxic effects may occur through a distinct mechanism compared with other anthracycline-type drugs [5].

●Oxidative stress – Anthracycline-mediated myocyte damage was previously attributed to the production of toxic reactive oxygen species (ROS) and an increase in oxidative stress that lead to lipid peroxidation of cardiomyocyte membranes, vacuolization, and finally, myocyte replacement by fibrous tissue [6-10]. However, oxidative stress is unlikely to be the sole mediator of cardiomyocyte damage; treatment with ROS scavengers does not consistently prevent doxorubicin-related cardiotoxicity [11,12].

●Topoisomerase-II mediated cell death – Anthracycline cardiotoxicity may also be mediated by the topoisomerase-II (Top2) enzyme. In cancer cells, doxorubicin targets the enzyme Top2 and binds to DNA to form the ternary Top2-doxorubicin-DNA cleavage complex, which triggers cell death [13,14]. Adult mammalian cardiomyocytes express Top2-beta [15]. In murine studies, cardiomyocyte-specific deletion of the Top2b gene (which encodes the Top2-beta enzyme) protects cardiomyocytes from doxorubicin-induced DNA double-strand breaks and transcriptome changes associated with the formation of reactive oxygen species, and protects mice from the development of doxorubicin-induced HF [16].

●Time course of cardiotoxicity – Most evidence suggests that anthracyclines cause an acute cardiac injury that may either progress or resolve over time. As examples:

•Large studies have documented a significant early increase in serum troponin in 30 to 35 percent of anthracycline-treated patients and have linked such early increases in cardiac troponin to traditional measures of cardiotoxicity such as LV dysfunction [17,18].

•In a prospective study of 2625 patients receiving anthracycline-containing chemotherapy, LV ejection fraction (LVEF) was assessed at baseline, every three months during chemotherapy and for the following year, every six months for the following four years, and yearly thereafter (median follow-up of 5.2 years; interquartile range 2.6 to 8 years) [19]. Cardiotoxicity was defined as an LVEF <50 percent with a decrease in LVEF of >10 percent from baseline. The overall incidence of cardiotoxicity was 9 percent, and the median time between the end of chemotherapy and detection of cardiotoxicity was 3.5 months (interquartile range 3 to 6 months). Ninety-eight percent of cases occurred during the first year. During the study period, 82 percent recovered from cardiotoxicity, 11 percent had full echocardiographic recovery, and 71 percent had partial recovery.

The clinical manifestations of anthracycline cardiotoxicity and the approach to its evaluation and diagnosis are discussed separately. (See "Clinical manifestations, diagnosis, and treatment of anthracycline-induced cardiotoxicity".)

EPIDEMIOLOGY — The overall prevalence of late symptomatic anthracycline-induced cardiotoxicity varies widely depending on the age of the population treated; the dose of anthracycline; hematologic cancer; presence of cardiovascular risk factors; history of heart disease, including low or impaired LVEF before chemotherapy; and the length of follow-up and the definitions used [20-22]. In adults, symptomatic HF appears to occur predominantly within two to three years of anthracycline exposure [23,24]. In a study of 135 patients with a median age of 59 with non-Hodgkin lymphoma who were treated with anthracyclines, 20 percent developed a significant cardiac event within one year of treatment [25].

●Adults – The prevalence of asymptomatic LV dysfunction depends on the population, methods, and definition [2,19]; however, the prevalence of LV dysfunction is much higher than symptomatic disease, with reported rates varying from 7 percent using LVEF to 45 percent using cardiac strain [26].

•In a randomized trial that assessed the effect of statin therapy in patients treated with doxorubicin, the baseline LVEF was 63±6 percent and LVEF after 24 months of observation was 57±6 percent [27]. Further details on the results of this trial are discussed below. (See 'Primary prevention with cardiovascular drugs' below.)

•In an echocardiographic study of 1853 adult survivors of childhood cancer with a median age of 31 years, an asymptomatic reduction in LVEF to <50 percent was noted in 7 percent of participants, and nearly all of these patients were asymptomatic [28].

•Similarly, in a cardiovascular magnetic resonance (CMR) imaging study among 114 adult survivors of childhood cancer who were free of cardiovascular symptoms, the prevalence was 14 percent [29].

•In a CMR study of adults treated with low to moderate doses of anthracycline-based chemotherapy, 26 percent of subjects had an asymptomatic reduction to LVEF <50 percent at six months after anthracycline exposure [30].

•A large Kaiser patient population study showed that anthracycline therapy for breast cancer was associated with a 1.84-fold risk of cardiomyopathy/HF and a 2.91-fold risk of mortality from any primary or secondary cardiovascular event compared with matched controls [31].

●Children – Identification of cardiotoxicity late after exposure has been described mainly in survivors of childhood cancers. The latent period between exposure and identification of a cardiomyopathy could be as long as 30 years [32,33]. Compared with sibling controls, survivors of childhood cancer have a nearly sixfold greater risk of developing HF (largely attributed to anthracycline exposure) [28].

RISK FACTORS

Patient characteristics — Prospective studies of anthracycline cardiotoxicity are relatively uncommon. Among the patient characteristics that increase the risk of cardiotoxicity, the best-described are:

●Age – In patients >65 years old or who are <4 years old, the risk of anthracycline toxicity increases [21,34,35].

●Obesity – A body mass index ≥30 kg/m² is associated with an increased risk of anthracycline-associated cardiotoxicity (odds ratio [OR] 1.7, 95% CI 1.1-2.6) [36]. One potential cause of toxicity may be overestimation of body surface area in persons with obesity, which may lead to drug overexposure [37].

●Diabetes mellitus – The presence of diabetes mellitus is associated with a higher risk of anthracycline-induced cardiotoxicity (OR 1.7, 95% CI 1.1-2.7) [38].

●Hypertension – Preexisting hypertension increases the risk of anthracycline cardiotoxicity (OR 2.0, 95% CI 1.4-2.8) [36]. In addition, those patients treated with anthracyclines who subsequently develop hypertension had a 12-fold (95% CI 7.5-20.1) higher risk for HF compared with those without hypertension [38].

There is controversy regarding whether female sex increases the risk of anthracycline-induced cardiotoxicity. In pediatric cancer patients, some studies support female sex as a risk factor for anthracycline cardiotoxicity, while others do not [39]. In contrast, there is anecdotal evidence that female sex reduces the risk of anthracycline-induced cardiotoxicity in adults treated for cancer [34,35,40,41].

Type of anthracycline — Most studies describing anthracycline cardiotoxicity have focused on unencapsulated doxorubicin, which is the most commonly used agent. In general, cardiotoxicity rates are relatively higher with mitoxantrone and idarubicin than with doxorubicin; they are both considered to be four to five times as cardiotoxic as doxorubicin, on a mg per mg basis, while epirubicin is considered roughly similar to doxorubicin [42,43], and daunorubicin has a lower risk of long-term cardiotoxicity, at least in children [44]. These cardiotoxicity rates relative to doxorubicin form the basis for recommendations about cumulative lifetime doses that should not be exceeded to avoid clinically significant cardiotoxicity.

Cumulative dose — Regardless of the type of anthracycline, the cardiotoxic effects of the anthracyclines are dose-dependent. The cumulative anthracycline exposure is a consistent risk factor for cardiotoxicity [19-21,45], and the risk of toxicity increases substantially above the upper cumulative lifetime limit of each agent:

●Doxorubicin – The risk for doxorubicin-related clinical HF ranges from 0.2 to 100 percent for cumulative doses ranging from 150 to 850 mg/m² [21,30,46,47]. When cardiotoxicity is defined to include development of asymptomatic LV systolic dysfunction, rates of cardiotoxicity are higher (table 1). In one study, the rate of doxorubicin-induced cardiac dysfunction (either symptomatic or asymptomatic but with a decline in LVEF) was 7, 9, 18, and 32 percent at cumulative doses of 150, 250, 350, and 400 mg/m² [21].

Patients can be susceptible to cardiotoxicity at lower cumulative doses. For example, detailed echocardiographic data from adult patients with breast cancer suggest that with anthracyclines (doxorubicin 240 mg/m2) there are modest but persistent declines in LVEF (approximately 4 percent) [48], as well as persistent worsening of diastolic function [49].

●Mitoxantrone – The lifetime cumulative dose of mitoxantrone is approximately 140 mg/m² [50].

●Epirubicin – For epirubicin, the risk of toxicity for a cumulative dose of 550 mg/m² is 0.9 percent and for a cumulative lifetime dose of 900 mg/m² is 3.3 percent.

●Idarubicin – There is disagreement about the upper limit for the cumulative lifetime dose of idarubicin [44,45,51-54]. Some suggest limiting cumulative doses to <150 mg/m² [51,54], others suggest <225 mg/m² [45], and still others state that, given the rarity of cardiotoxicity, there is no definable cumulative dose beyond which rates of cardiotoxicity are higher than for lower doses [44,52,53].

●Daunorubicin – The incidence of myocardial toxicity increases after a total cumulative dose of 400 to 550 mg/m² in adults, 300 mg/m² in children over age 2, and 10 mg/kg for younger children.

●Treatment with multiple agents – For patients who may be treated with multiple anthracycline-type agents over the course of their disease, the contribution of each agent (ie, percent of lifetime dose received) must be estimated and added together to determine whether the patient has approached the allowable limit of lifetime anthracycline dose.

More information about the risk of cardiotoxicity for the various anthracycline agents is provided below. (See 'Other factors' below.)

Methods of anthracycline administration — The risk of anthracycline toxicity may be decreased by using lower per cycle doses, infusional versus bolus dosing, a slower rate of infusion, and liposomal rather than unencapsulated preparations. The benefits of these approaches in terms of reducing cardiotoxicity, and the impact on anticancer efficacy, are discussed below. (See 'Modification of dose, specific agent, and schedule of administration' below.)

Combined treatment with other cancer therapies — Concomitant administration of anthracyclines with other cancer treatments may increase the risk of anthracycline-induced cardiotoxicity; these include:

●Radiation therapy involving the cardiac silhouette [55]. (See "Cardiotoxicity of radiation therapy for breast cancer and other malignancies".)

●HER2-targeted therapy [56]. (See "Cardiotoxicity of trastuzumab and other HER2-targeted agents".)

●Cyclophosphamide therapy. (See "Cardiotoxicity of cancer chemotherapy agents other than anthracyclines, HER2-targeted agents, and fluoropyrimidines".)

Other factors — Factors that may influence the cardiotoxicity of anthracyclines but which are not routinely used to assess risk during the pretreatment assessment include:

●Genetic predisposition – Genetic variation also contributes to chemotherapy-related cardiotoxicity and may modify the relationship between anthracycline dose and cardiotoxicity risk. However, genetic tests are not currently available for routine clinical use.

Systematic reviews and case control studies suggest an association between inherited polymorphisms in specific genes and anthracycline-mediated cardiotoxicity across multiple patient populations, including children and adults [57]. As examples:

•In a case-control study of 170 cancer survivors with cardiomyopathy compared with 317 survivors without cardiomyopathy, the risk of cardiomyopathy was increased in patients with the myocardial cytosolic carbonyl reductase 3 (CBR3) homozygous GG genotype (OR 1.8) [58]. Furthermore, in patients with GG versus the GA or AA genotype who were treated with a low-dose anthracycline (1 to 250 mg/m²), the risk of cardiomyopathy was 3.3 times higher. At doses greater than 250 mg/m², there was no significant interaction by genotype; the risk of LV dysfunction increased irrespective of the CBR3 genotype status.

•Other studies have identified additional pathogenic variants that might modulate the risk of anthracycline-related cardiomyopathy [57,59-63]. Examples include key genes associated with drug biotransformation (SLC28A3, SLC22A17, UGT1A6), specifically with anthracycline elimination (eg, glutathione S-transferase 1 [GSTM1]), antioxidant pathways (HAS3), topoisomerase-2B-mediated DNA damage (RARG), truncating variants in the titin gene (eg, CMD1G truncating variants [TTNtvs]), and genetic variants in the profibrotic Slit-Robo signalling pathway (ROBO2).

●Risk scores – Risk scores that predict the incidence of anthracycline cardiotoxicity are available but are not typically used in clinical practice [35,64,65].

PRETREATMENT ASSESSMENT — For all patients with a plan to receive anthracycline-containing chemotherapy, a pretreatment assessment is performed to identify risk factors for anthracycline toxicity, which are then used to determine the risk for cardiotoxicity.

Initial evaluation — All patients should undergo a history, physical examination, and electrocardiography (ECG). The history and physical examination are focused on the detection of signs or symptoms of cardiovascular disease (CVD) including HF, arrhythmias, valve disease, and coronary artery disease. (See "Heart failure: Clinical manifestations and diagnosis in adults" and "Evaluation of palpitations in adults" and "Auscultation of cardiac murmurs in adults" and "Outpatient evaluation of the adult with chest pain".)

Left ventricular function assessment — Prior to administration of an anthracycline, the contributors to this topic routinely obtain an assessment of LV function, regardless of the planned number of cycles or expected cumulative dose of anthracycline.

●Initial testing with echocardiography – For patients who undergo pretreatment echocardiography, we obtain a full echocardiographic study with appropriate use of echo contrast. (See "Contrast echocardiography: Contrast agents, safety, and imaging technique", section on 'Microbubble contrast agents'.)

•Ejection fraction – The LVEF should be quantified using two-dimensional (2D) images. If the echocardiography laboratory performing the echocardiogram has experience with three-dimensional (3D) LVEF calculation or global longitudinal strain (GLS), those measures should be obtained. In patients who undergo echocardiography, abnormal LV function is typically defined by a 2D or 3D echocardiographic LVEF below the institutional lower limit of normal.

Though what constitutes an abnormal LVEF varies between centers and between professional societies, the 2015 American Society of Echocardiography and European Association of Cardiovascular Imaging guidelines classify a 2D echocardiographic LVEF of <52 percent in men and <54 percent in women as abnormal (ie, more than two standard deviations below the mean LVEF for adults without CVD) [66]. In practice, other institutions use the lower limit of normal of 50 percent suggested by cancer and cardio-oncology societies [67-69].

•Global longitudinal strain – The contributors to this topic rely primarily on LVEF to determine the risk of toxicity and use GLS as another echocardiographic measure of LV function, especially in patients with low normal LVEF (within 5 points of the lower limit of normal). In practice, an absolute value of GLS <16 percent is of concern, and a GLS of 16 to 18 percent is considered borderline [70,71].

An absolute value of GLS less than 19 percent has been found to predict subsequent LV dysfunction in women with breast cancer [26,46,72]. (See 'Monitoring during and after treatment' below.)

•Other echocardiographic findings – In addition to the LVEF, the echocardiogram should be reviewed for abnormalities that corroborate the overall assessment of LV function (eg, chamber dilation, right ventricular systolic pressure estimation, central venous pressure) or that suggest the presence of another cardiovascular disorder that may affect the approach to therapy (eg, valve disease, pericardial disease). Patients with abnormal LV systolic function or another severe cardiac abnormality should undergo additional evaluation prior to determining whether the benefits of anthracycline administration outweigh the risks. (See 'Additional assessment for patients with evidence of CVD' below.)

●Patients with a nondiagnostic echocardiogram – In patients who have a nondiagnostic echocardiogram, a cardiology consultation for review of the images and recommendations for additional imaging may be appropriate. In patients whose echocardiographic images are nondiagnostic, we typically obtain a CMR examination for quantitation of LVEF. If CMR is not available, radionuclide ventriculography (RVG, also known as multiple gated cardiac blood pool imaging) can be used to assess LVEF. (See "Tests to evaluate left ventricular systolic function" and "Tests to evaluate left ventricular systolic function", section on 'Radionuclide ventriculography'.)

●Rationale – The rationale for LVEF assessment before treatment is to detect preexisting LV dysfunction prior to anthracycline-based chemotherapy that may alter the plan to administer anthracycline, and to provide a baseline study for comparison during monitoring. In the general population, asymptomatic LV dysfunction is detected in 3 percent, increases with age, and may be higher in patients with cancer [73]. Our preference to perform a pretreatment assessment with echocardiography is consistent with professional guidelines [46,74].

However, limitations to this approach include the possibility that a falsely low LVEF could result in withholding of effective cancer treatment and the ongoing uncertainty around the overall efficacy of serial testing to prevent cardiotoxicity [46]. There are few data to guide this approach.

The evidence for specific tests includes the following:

•Echocardiography – We prefer echocardiography among the available modalities given its wide availability and the evidence base for its use for LVEF (as well as more sensitive measures such as strain) [2]. In addition, echocardiography enables detection and assessment of preexisting, alternate, or concurrent cardiac conditions such as pericardial disease, valve disease, and coronary artery disease [75].

-Prognostic value – There are limited data on the prognostic value of a low LVEF on the occurrence of subsequent HF and/or cardiac mortality among patients receiving anthracycline-containing chemotherapy. In a large, single-center study with 2285 subjects, 45 had a baseline LVEF at or below the lower limits of normal (defined as an LVEF of 52 percent in males and 54 percent in females) and 112 patients had an LVEF within 5 percent of the lower limits of normal [22]. Having an abnormal baseline LVEF or a baseline LVEF within 5 percent of the lower limits of normal was predictive of HF and cardiac death; each 5 percent decrement in LVEF predicted a 40 percent greater risk of developing HF or cardiac death.

-Issues related to measurement error – The wide variability in the confidence intervals for standard 2D measurement of LVEF is significant and approaches approximately 8 to 10 percent in research studies [76,77]. The use of 3D echocardiography or GLS may provide more accurate assessment of LVEF, but these measures are not widely available. A full discussion of the properties of various echocardiographic measures of LV function can be found separately. (See "Tests to evaluate left ventricular systolic function", section on 'Echocardiography'.)

The rationale for sequential monitoring of LV function with echocardiography during treatment with anthracyclines and the relative value of 2D LVEF, 3D LVEF, and GLS are discussed separately in this topic. (See 'Monitoring during and after treatment' below.)

•Cardiovascular magnetic resonance imaging – CMR is the clinical gold standard imaging technique for measurement of LV volumes and LVEF, provides information on other cardiac conditions (eg, valve disease, infiltration), and does not require ionizing radiation or iodinated contrast. CMR imaging is a more sensitive test for LV dysfunction than echocardiography [29] and can be used to assess for a decline in LV mass [78-80], which is associated with a higher likelihood of developing HF symptoms [79,80].

The disadvantages of CMR imaging include its restricted availability to specialized centers and the inability to perform CMR imaging in patients with retained metal. (See "Tests to evaluate left ventricular systolic function", section on 'Cardiovascular magnetic resonance imaging'.)

•Nuclear imaging – RVG imaging has a high degree of reproducibility but requires radiation exposure. Nuclear imaging is not the first choice for LVEF assessment; it does not provide additional assessment of cardiac structure or function and requires radiation exposure. (See "Tests to evaluate left ventricular systolic function", section on 'Radionuclide ventriculography' and "Tests to evaluate left ventricular systolic function", section on 'Single-photon emission computed tomography myocardial perfusion imaging (SPECT-MPI)'.)

Additional assessment for patients with evidence of CVD — In patients who have preexisting cardiovascular risk factors for anthracycline-induced cardiotoxicity identified during the pretreatment assessment, we suggest the following:

●The prognosis for the patient’s cancer with and without anthracycline-containing therapy should be assessed, and the availability of effective non-anthracycline-containing regimens should be determined. This information is vital to guide the estimation of the risk to benefit ratio for anthracycline administration. (See 'Preventive management of anthracycline therapy' below.)

●A cardiology consultation is suggested for most patients with history of CVD or preexisting evidence of LV dysfunction. In this setting, the role of the cardiologist is to ascertain the accuracy and prognosis of any CVD detected. If present, the cardiologist should determine the cause and severity of HF as well as appropriate HF therapy.

Identification of patients at increased risk — As described in the American Society of Clinical Oncology expert consensus guideline on the prevention and monitoring of cardiac dysfunction in survivors of adult cancers [81], the presence of any one of the following criteria identifies patients with an increased risk for anthracycline-related cardiac dysfunction:

●Any patient receiving high-dose anthracycline (eg, cumulative doxorubicin ≥250 mg/m², epirubicin ≥600 mg/m²).

●Patients receiving lower-dose anthracycline (eg, cumulative doxorubicin <250 mg/m², epirubicin <600 mg/m²) who have at least one of the following factors:

•Concurrent low-dose radiotherapy (<30 Gy) where the heart is in the treatment field. (See "Cardiotoxicity of radiation therapy for breast cancer and other malignancies".)

•Two or more CVD risk factors including smoking, hypertension, diabetes, dyslipidemia, or obesity.

•Older age (≥60 years) at initial cancer treatment.

•Decreased LV function (eg, borderline low LVEF [50 to 55 percent], history of myocardial infarction, moderate or greater valvular heart disease) at any time before or during treatment. (See "Echocardiographic evaluation of the mitral valve" and "Echocardiographic evaluation of the aortic valve".)

•Sequential treatment with a HER2-targeted agent (eg, trastuzumab). (See "Cardiotoxicity of trastuzumab and other HER2-targeted agents", section on 'Risk factors'.)

●High-dose radiotherapy (≥30 Gy) where the heart is in the treatment field.

PREVENTIVE MANAGEMENT OF ANTHRACYCLINE THERAPY

Using non-anthracycline-containing regimens when feasible — An important issue to consider when addressing preventive management of anthracycline cardiotoxicity is the availability of a non-anthracycline-containing regimen with efficacy similar to the anthracycline-containing regimen.

For patients with advanced cancers who are being treated with palliative intent, a non-anthracycline containing regimen is usually available, and for selected patients with earlier-stage, potentially curable cancers, it might also be feasible. As an example, in the adjuvant treatment of breast cancer (both HER2-positive and HER2-negative), docetaxel plus cyclophosphamide effectively reduces the risk of recurrence to a similar magnitude as an anthracycline-containing regimen. In this setting, patients who are at high risk for anthracycline-based cardiotoxicity could reasonably be offered a non-anthracycline-containing regimen. Among patients with HER2-positive disease, these non-anthracycline-containing regimens are actually preferred because of the higher risk of cardiotoxicity when doxorubicin is combined with trastuzumab [56]. (See "Selection and administration of adjuvant chemotherapy for HER2-negative breast cancer", section on 'Choosing a regimen' and "Adjuvant systemic therapy for HER2-positive breast cancer", section on 'Tumors >2 cm and/or node positive'.)

Patients with preexisting left ventricular systolic dysfunction — In patients with preexisting LV systolic dysfunction, the decision to administer or withhold an anthracycline should be based on an individualized assessment of the risk and benefit of anthracycline therapy relative to any alternative regimen; the assessment of benefit may differ according to the disease setting (ie, potentially curative versus palliative). However, the degree of LV systolic dysfunction influences the decision to proceed with anthracycline-based therapy as follows:

●Moderate to severe LV systolic dysfunction – In most patients with an LVEF ≤40 percent, anthracycline chemotherapy should generally be avoided and, when available and feasible, alternative non-anthracycline-containing regimens should be used.

If non-anthracycline-containing regimens are not feasible, and the decision is made to proceed with an anthracycline after a shared decision-making process, we suggest the following:

•Management of LV dysfunction – Regardless of the decision to treat with anthracycline, patients with newly diagnosed LV dysfunction require optimal management, which includes a diagnostic assessment and initiation of treatment with appropriate medications and devices. The approach to management depends on the type of LV dysfunction and the severity of symptoms, as described separately. (See "Overview of the management of heart failure with reduced ejection fraction in adults" and "Management and prognosis of asymptomatic left ventricular systolic dysfunction".)

•Mitigation of cardiotoxicity – The use of alternative, potentially less-cardiotoxic anthracyclines, lower per cycle doses, infusional rather than bolus schedule of administration, avoidance of concomitant cardiotoxic agents, if possible, or use of upfront dexrazoxane may be appropriate in this population, although the use of dexrazoxane in this setting would be off-label. (See 'Use of dexrazoxane' below.)

For this group of patients, the use of neurohormonal therapy for HF with reduced ejection fraction (HFrEF) is indicated for treatment regardless of any potential cardioprotective effects, and treatment should be started prior to chemotherapy. (See 'Primary prevention with cardiovascular drugs' below.)

In addition, careful and frequent monitoring for cardiotoxicity is required during therapy, as discussed below. (See 'Monitoring during and after treatment' below.)

These approaches to management are based on our experience and the desire to avoid further injury to the myocardium as the result of anthracycline administration; there are no studies that report the results of anthracycline-based regimens in this group of patients.

●Mild LV systolic dysfunction – In patients with an LVEF of 41 to the lower limit of normal and who have mildly reduced global longitudinal strain (absolute value <16 percent), alternatives to anthracycline-containing therapy are preferred, if available and feasible.

If alternatives to anthracycline-containing therapy are not feasible, and the decision is made to proceed with an anthracycline after a shared decision-making process, we suggest the following:

•Management of mild LV systolic dysfunction – Asymptomatic patients with mild LV systolic dysfunction should undergo an evaluation for the cause of reduced LVEF. The optimal approach to neurohormonal therapy in these populations is unclear and is discussed separately. (See "Approach to diagnosis of asymptomatic left ventricular systolic dysfunction" and "Management and prognosis of asymptomatic left ventricular systolic dysfunction" and "Asymptomatic left ventricular diastolic dysfunction".)

Patients with HF symptoms and mild LV systolic or diastolic dysfunction may benefit from therapies for HFrEF or HF with preserved ejection fraction (HFpEF). This issue is discussed separately. (See "Treatment and prognosis of heart failure with preserved ejection fraction" and "Management and prognosis of asymptomatic left ventricular systolic dysfunction".)

•Mitigation of cardiotoxicity – Cardiotoxicity mitigation in this population includes serial monitoring of LVEF. The use of alternative, potentially less-cardiotoxic anthracyclines, lower per cycle doses, infusional rather than bolus schedule of administration, avoidance of concomitant cardiotoxic agents, if possible, or use of upfront dexrazoxane may be appropriate in this population, although the use of dexrazoxane in this setting would be off-label. (See 'Approaches to cardiotoxicity mitigation' below.)

The use of neurohormonal treatment strategies depends on the presence or absence of symptoms:

-Asymptomatic patients – In asymptomatic patients with mildly reduced LV systolic function who do not otherwise have an indication for neurohormonal HF therapy but who have an LVEF between 41 and 49 percent, we typically administer an angiotensin converting enzyme (ACE) inhibitor (or an angiotensin II receptor blocker if patient has angioedema or cough with ACE inhibitor) and a beta blocker (eg, carvedilol) prior to the administration of anthracyclines. This approach treats patients who may not have an indication for treatment based on the presence of reduced LVEF alone. (See 'Primary prevention with cardiovascular drugs' below.)

This approach is based on our experience and extrapolation of the favorable effects of neurohormonal HF therapy on cardiac remodeling among patients with HFrEF; there are no prospective studies to support benefit of neurohormonal HF therapy in asymptomatic patients with mildly reduced LVEF who will undergo treatment with an anthracycline.

-Symptomatic patients – Patients with mild LV dysfunction and symptomatic HF may benefit from a variety of mitigation strategies listed separately in this topic (see 'Approaches to cardiotoxicity mitigation' below).

In patients with mild LV dysfunction (systolic) and symptoms of HF, the treatment of HF typically consists of therapies appropriate for HFpEF. Therapy for HFpEF should be optimized before adding other agents used to treat HF for the purpose of preventing anthracycline cardiotoxicity. There are no data to guide practice in this group of patients.

Patients with other forms of cardiovascular disease — In patients with a high risk of or preexisting coronary artery disease, diastolic dysfunction, arrythmias, valve disease, or other forms of cardiac disease (eg, pericardial disease), the approach to therapy must be individualized based on the possible risks and benefits of anthracycline therapy as determined by collaboration between the treating oncologist and a cardiologist. At a minimum, appropriate monitoring for cardiotoxicity is required during and after treatment. (See 'Monitoring during and after treatment' below.)

Patients without cardiovascular disease — For patients with no history of HF with baseline LVEF ≥50 percent, the benefits of anthracycline typically outweigh the risks, especially for early-stage cancers that are potentially curable. Patients who have no signs of CVD should undergo serial monitoring for early detection of cardiotoxicity, as described below. (See 'Monitoring during and after treatment' below.)

There is no clear role for neurohormonal therapy for the primary prevention of cardiotoxicity ("cardioprotective therapy") in this population. (See 'Primary prevention with cardiovascular drugs' below.)

APPROACHES TO CARDIOTOXICITY MITIGATION

Modification of dose, specific agent, and schedule of administration — The risk of anthracycline cardiotoxicity may be mitigated by altering various aspects of the chemotherapeutic regimen, though none of these measures are routinely employed to reduce the risk of toxicity. In patients who are at high risk of toxicity or who show signs of toxicity and for whom continued anthracycline treatment is the preferred therapy, individualized use of one or more of the following strategies may reduce the risk of toxicity:

●Limit cumulative dose – As noted above, the risk of cardiotoxicity has long been recognized to correlate with the cumulative anthracycline dose. For patients receiving doxorubicin, the lifetime dose is generally limited to 450 mg/m², although higher doses may be delivered on an individualized basis, particularly for those with limited other options who are being treated with curative intent and who have not developed HF or a significant decline in LVEF. Cumulative lifetime dose limits for other anthracycline drugs are provided above. (See 'Type of anthracycline' above and 'Cumulative dose' above.)

●Infusional rather than bolus dosing – A cardioprotective strategy that has been employed in hematologic malignancies and in certain sarcoma protocols is infusional rather than bolus dosing of anthracycline.

The rationale for using infusional rather than bolus regimens to reduce cardiotoxicity is as follows:

•In adult patients, slower infusion of anthracyclines lowers the peak plasma level (C_max), which correlates with the propensity to develop cardiotoxicity, but it does not affect the area under the curve, which correlates with antitumor activity [82].

•Early studies demonstrated that infusion times over 48 to 96 hours resulted in less injury to the myocardium on endomyocardial biopsy, less clinical HF, and preserved antitumor responses [83].

•In patients with breast cancer, infusional rather than bolus doxorubicin dosing has been associated with lower rates of HF [84,85].

•Evidence for the cardioprotective effect of infusional rather than bolus dosing comes from a meta-analysis including seven studies with different infusion durations. Among a total of 803 participants, the risk of HF (risk ratio [RR] 0.27, 95% CI 0.09-0.81) and subclinical evidence of cardiotoxicity (RR 0.36, 95% CI 0.15-0.90) was reduced with longer-duration infusion compared with more rapid infusion [86]. No significant difference in cancer response rate was observed on analysis of the two studies that included response rate data.

The downsides of infusional versus bolus anthracyclines are the longer duration of administration (increases cost) and the need for administration through a central venous line because of the vesicant properties. (See "Extravasation injury from cytotoxic and other noncytotoxic vesicants in adults".)

●Use of liposomal formulations – For patients with a metastatic, anthracycline-responsive cancer who are anticipated to receive cumulative doses of doxorubicin ≥300 mg/m² or epirubicin ≥600 mg/m², we suggest a switch to pegylated liposomal doxorubicin (PLD) rather than continuation of the nonencapsulated anthracycline alone. Another option in this setting is to add dexrazoxane to the unencapsulated anthracycline. (See 'Use of dexrazoxane' below.)

Pegylated liposome-encapsulated of doxorubicin may preferentially enter and accumulate in malignant tissues as a result of a tumor's leaky microvasculature and impaired lymphatics, whereas drug entry into cardiomyocytes in the heart is limited [87]. Several clinical trials and at least two meta-analyses have confirmed comparable or nearly comparable antitumor efficacy and significantly reduced cardiotoxicity observed with liposomal compared with unencapsulated anthracycline formulations [42,88-91]. However, the bulk of the available data are in patients with advanced breast cancer. Furthermore, there may be a trade-off in terms of noncardiovascular toxicity. Compared with unencapsulated anthracyclines, the use of PLD has been associated with an increased risk of mucositis, infusion reactions, and hand-foot syndrome [92]. (See "Endocrine therapy resistant, hormone receptor-positive, HER2-negative advanced breast cancer", section on 'Comparing anthracyclines' and "Overview of the initial treatment of metastatic soft tissue sarcoma", section on 'Pegylated liposomal doxorubicin'.)

Use of dexrazoxane — Dexrazoxane is an effective drug for primary prevention of anthracycline-related cardiotoxicity. Mechanistically, dexrazoxane is believed to chelate intracellular iron, block iron-assisted oxidative radical production, and inhibit the topoisomerase II-beta isoenzyme, which has been implicated in anthracycline cardiotoxicity [3,14,16,93].

●Indications – There is no widespread agreement on which patients receiving anthracyclines for cancer treatment are appropriate to initiate dexrazoxane. For most adults with advanced cancer who are anticipated to exceed a cumulative dose beyond 300 mg/m² of doxorubicin (or 600 mg/m² epirubicin), a switch to a different non-anthracycline-containing regimen is a preferred strategy, if feasible. For the rare patient in whom this approach is not feasible, and in whom there is an ongoing indication to continue doxorubicin-based chemotherapy, we suggest dexrazoxane rather than continuation of doxorubicin alone. Another option is to switch to a liposomal preparation of doxorubicin.

This recommendation is consistent with the 2017 American Society of Clinical Oncology (ASCO) guidelines for prevention and monitoring of cardiac dysfunction in survivors of adult cancers, which indicate that initiation of dexrazoxane or a switch to a liposomal encapsulated formation may be considered to prevent cardiotoxicity in patients planning to receive additional treatment with either doxorubicin ≥250 mg/m² or epirubicin ≥600 mg/m² [81].

Children receiving anthracycline-based chemotherapy may also be considered for dexrazoxane. However, most children with cancer are treated on standard chemotherapy protocols as defined by major cooperative groups, with the use of dexrazoxane usually dictated by the protocol, and specific to the diagnosis and disease stage. An example of this approach is described separately. (See "Overview of hepatoblastoma", section on 'Approaches to minimizing long-term treatment-related toxicity'.)

●Dosing – The recommended dose ratio of dexrazoxane to doxorubicin is 10:1 (eg, 500 mg/m² of dexrazoxane per each 50 mg/m² of doxorubicin). Dexrazoxane is administered as an IV infusion over 15 minutes prior to doxorubicin dose; the interval from the completion of the dexrazoxane infusion to the initiation of doxorubicin should not exceed 30 minutes.

●Evidence for efficacy and concerns about adverse oncologic outcomes – Although much of the evidence supports the safety and efficacy of dexrazoxane as a cardioprotectant in individuals receiving high doses of an anthracycline, its effect on long-term cancer-related outcomes remains controversial. Several trials and meta-analyses provide evidence of efficacy for dexrazoxane as a cardioprotectant [42,94-103]. Although the majority of data are in adults with advanced breast cancer, benefits have been observed in children, and in other malignancies as well:

•A network meta-analysis of randomized controlled trials of primary prevention of anthracycline cardiotoxicity included eight studies with a total of 666 patients treated with dexrazoxane (66 percent had breast cancer as the primary indication for the anthracycline; the remainder had other solid tumors). Dexrazoxane therapy significantly reduced the risk of HF (odds ratio [OR] 0.12, 95% CI 0.06-0.23) as well as a composite of HF or LV systolic dysfunction (OR 0.26, 95% CI 0.11-0.74) [101]. There was no impact of dexrazoxane on response rate or risk of malignancy-associated death.

•At least three meta-analyses have been conducted [42,102,103] using the same four randomized trials of anthracycline with or without dexrazoxane [104-107]. All concluded that dexrazoxane significantly decreased the risk of acute clinical or subclinical HF (relative risk 0.21-0.31) without compromising antitumor efficacy. Notably, the vast majority of patients in all four trials were adults with advanced breast cancer, the anthracycline used was either epirubicin or doxorubicin, and two of the four trials were conducted in patients who had previously received anthracyclines in the setting of adjuvant therapy [106,107]. As an example, the 2011 Cochrane review in children and adults (10 trials, 1619 patients, 9 of 10 involving patients with breast cancer) concluded that compared with no dexrazoxane, the use of dexrazoxane reduced the occurrence of HF (RR 0.29, 95% CI 0.20-0.41) [102]. There was no difference in response rate or survival between the dexrazoxane and control patients.

•Among children, long-term follow up of six anthracycline-containing pediatric trials, five of which randomized patients to treatment with or without dexrazoxane, found that the risk of serious cardiovascular outcomes (ie, cardiomyopathy, ischemic heart disease, and stroke) occurred less commonly in those treated with dexrazoxane than without it (5.6 versus 17.6 percent), although the incidences of cardiomyopathy were similar (4.4 versus 8.1 percent) [95]. Use of dexrazoxane did not appear to adversely affect long-term mortality, event-free survival, or second cancer risk.

The routine use of this agent had been tempered by concerns that use of dexrazoxane might result in inferior long-term outcomes, either by reducing the antitumor activity of the anthracycline-containing regimen or increasing the rate of secondary malignancies; however, data have challenged this hypothesis:

•In an early trial of patients with advanced breast cancer, a decreased objective response rate was observed in patients treated with dexrazoxane versus those receiving placebo (47 versus 61 percent) [108]. However, time to tumor progression and survival were not adversely impacted, and patients treated with placebo had a much higher rate of LVEF decline or HF compared with those treated with dexrazoxane (hazard ratio 2.63, 95% CI 1.61-4.27).

•Another early study raised concern that dexrazoxane may be associated with a higher risk of secondary malignancies in survivors of Hodgkin lymphoma [109].

•In contrast, a meta-analysis and network meta-analysis of subsequent trials failed to confirm impaired antitumor efficacy with the use of dexrazoxane, with the data primarily derived in adults with advanced breast cancer [110].

•Furthermore, subsequent reports of childhood survivors of a variety of hematologic malignancies [111-116], and of adults with breast cancer or hematologic malignancies receiving dexrazoxane as a cardioprotectant strategy [117,118], did not confirm adverse oncologic outcomes.

Monitoring during and after treatment — Another approach to mitigation of cardiotoxicity caused by anthracyclines is to closely monitor for toxicity during anthracycline-based treatment. The contributors to this topic have different approaches to on-treatment and posttreatment monitoring that include differences in the ideal measurement of LV function, the threshold for testing, and the interval for testing:

●Monitoring with echocardiography – In patients who undergo monitoring for anthracycline toxicity, we typically perform serial echocardiography with calculation of 2D LVEF and, if available, calculation of 3D LVEF and global longitudinal strain (GLS). Some contributors to this topic rely on the calculated LVEF, while others prefer to use GLS for surveillance. If the patient has a history of suboptimal images or cannot undergo echocardiography due to chest wall radiation or surgery, serial CMR imaging or radionuclide ventriculography can be used for monitoring. (See 'Left ventricular function assessment' above.)

There is no consensus on the optimal frequency of monitoring with echocardiography, and the approach is variable among the contributors to this topic:

•Patients without risk factors for toxicity – In patients undergoing anthracycline-based chemotherapy who have no risk factors for toxicity, the contributors to this topic differ in their approach:

The contributors to this topic agree that echocardiography should be obtained before anthracycline therapy in all patients. This approach is consistent with the 2014 expert consensus statement of the American Society of Echocardiography and the European Association of Cardiovascular Imaging [46].

Some of the contributors also obtain a follow-up echocardiogram after six months to one year of anthracycline treatment even if the cumulative doxorubicin dose is ≤240 mg/m². This approach is generally consistent with the National Comprehensive Cancer Network (NCCN) guidelines, which states that surveillance at one year should be considered [119].

Aside from any routine assessment, an echocardiogram should be obtained if routine clinical assessment identifies signs or symptoms of cardiac dysfunction. (See 'When can monitoring be discontinued?' below.)

•Patients who received a high dose of anthracycline – In patients who have received a high dose of anthracycline, our authors differ in their approaches to surveillance:

Some contributors change their surveillance approach once the cumulative doxorubicin dose exceeds 240 mg/m²; these contributors evaluate with echocardiography when the patient’s cumulative dose of anthracycline exceeds 240 mg/m² and after each additional cumulative dose of 50 mg/m². This approach is consistent with the 2014 expert consensus statement of the American Society of Echocardiography and the European Association of Cardiovascular Imaging [46].

Some contributors to this topic begin surveillance after the cumulative dose of doxorubicin exceeds ≥360 mg/m². Upon exceeding this dose, an echocardiogram is obtained, and further echocardiograms are obtained after two subsequent cycles until the completion of chemotherapy.

Beyond these assessments, echocardiography is obtained if routine clinical assessment identifies signs or symptoms of cardiac dysfunction. (See 'When can monitoring be discontinued?' below.)

•Patients with preexisting or acute LV dysfunction – In patients with preexisting LV dysfunction or acute cardiotoxicity who will undergo further anthracycline exposure, we typically obtain an echocardiogram after each subsequent cycle of chemotherapy during therapy and at least annually after the completion of chemotherapy.

●Other monitoring tests – We do not monitor for toxicity with serial assessment of troponin or B-type natriuretic peptide (BNP); the role of these biomarkers to identify subclinical cardiotoxicity is uncertain [67]. Our practice is consistent with guidelines from ASCO [81], Cancer Care Ontario, and the European Medicines Agency. The most recent consensus-based guidelines for management of invasive breast cancer from the NCCN did not address this issue [120,121], although an earlier version suggested not using serial biomarkers to assess subclinical cardiotoxicity during chemotherapy for breast cancer [122].

However, other professional organizations recommend biomarker monitoring, and practice seems to be different in Europe. Joint guidelines from the European Society of Cardiology, the European Society for Therapeutic Radiology and Oncology, and the International Cardio-Oncologic Society recommend that patients receiving anthracyclines who are at high or very high risk for cancer therapy-related cardiac disease (age ≥80, prior history of HF, severe valvular heart disease, myocardial infarction or coronary revascularization, stable angina, baseline LVEF <50 percent) undergo baseline measurement of biomarkers (natriuretic peptides and/or troponin), before every new cycle of anthracycline therapy, and 3 and 12 months after therapy is completed [123-125]. Consensus recommendations for management of cardiac disease in cancer patients from the European Society of Medical Oncology noted that the evidence to support biomarker monitoring is not clear but include a weak recommendation for monitoring biomarkers such as troponin I and BNP as a potential means of identifying patients who may require further assessment [67]. They suggest that all asymptomatic patients with normal LVEF receiving anthracycline treatment should undergo periodic (every three to six weeks or before each new cycle of therapy) measurement of troponin, BNP, or N-terminal pro-BNP (NT-proBNP) [67].

Other groups endorsing the use of cardiac biomarkers during anthracycline therapy include the Canadian Cardiovascular Society [126], the Italian Society of Cardiology working group on Drug Cardiotoxicity and Cardioprotection [127], and the American Society of Echocardiography/European Association of Cardiovascular Imaging Expert Consensus Panel [46]. Another group proposes the use of biomarkers to monitor only patients at higher risk of cardiotoxicity (eg, doxorubicin ≥250 mg/m², age ≥60, mediastinal radiotherapy, previous heart disease, two or more cardiac risk factors [smoking, hypertension, diabetes, dyslipidemia, chronic kidney injury, obesity], baseline elevation of cardiac troponin) [128]. (See 'Goals of preventive management' above.)

●Rationale and evidence – Our approach to screening is intended to detect cardiotoxicity at a time when its effects can be mitigated. The efficacy of this approach is largely based on the ability of monitoring to identify severe signs of acute toxicity that lead to alterations of chemotherapy or treatment of HF.

•Echocardiography – There are limited data on the long-term benefits of serial echocardiographic monitoring on survival. Our approach is based on the relatively high frequency with which echocardiography detects LV systolic dysfunction, which ranges from 3 to 10 percent of monitored patients. Data support the hypothesis that early detection of cardiac toxicity improves outcomes.

In one study published in 2015 that included patients with breast cancer undergoing anthracycline-based therapy, screening with standard echocardiography was performed at baseline, every three months during chemotherapy, and every six months after the conclusion of therapy for four years [19]. Cardiotoxicity was defined as a decrease in LVEF of >10 percent to less than 50 percent. The overall incidence of cardiotoxicity was 9 percent, the median time between the end of chemotherapy and detection of cardiotoxicity was 3.5 months (interquartile range 3 to 6 months), and 98 percent of cases occurred during the first year. During the study period, 11 percent of patients had full echocardiographic recovery while 71 percent had partial recovery.

-Strain imaging - GLS measures myocardial deformation and may enable early detection of subclinical cardiotoxicity. Due to measurement variability, the routine use of strain rate imaging is less common in clinical practice and therefore most studies have focused on the use of strain imaging.

A number of studies have explored the utility of strain and strain rate imaging in the detection of cancer therapy-induced cardiac toxicity and in the prediction of subsequent decreases in LVEF [26,72,129-131]. In one trial that included 331 patients treated with an anthracycline and that assigned patients to monitoring with GLS (toxicity defined by a ≥12 percent reduction in GLS) or LVEF (≥10 percent decrease), the change in 3D LVEF (ie, the primary endpoint) was similar in the GLS group and the LVEF group (-3 versus -2.7 percent, respectively, p = 0.69); this finding was of unclear statistical significance [132,133]. Chemotherapy was discontinued in 6 percent of patients in the GLS group and 3 percent in the LVEF group.

•Biomarkers – The value of serial testing with cardiac biomarkers to detect subclinical anthracycline cardiotoxicity is controversial, and recommendations of national and international organizations vary widely [134]. In our opinion, the available evidence does not suggest that serial measurement of troponin and BNP levels is an effective monitoring strategy. (See 'Goals of preventive management' above.)

-Troponin – Among patients with acute cardiotoxicity from anthracycline exposure, serum levels of cardiac troponin may be acutely elevated [17,26,129,135-138]. However, the overall effectiveness of troponin monitoring is difficult to assess; studies of troponin monitoring included only small numbers of heterogeneous patients, have different timing of troponin assays, have different thresholds for a positive, and arrived at different conclusions.

In a prospective single-center study with 703 subjects, cardiac troponin I concentrations were assessed immediately following and one month after chemotherapy. An increase in troponin I concentrations at one time-point was associated with a 37 percent increase in the risk of a subsequent decline in LVEF, while elevation of troponin I at two separate time points was associated with an 84 percent increase in risk [17]. Similarly, in a multicenter cohort of 78 patients, for every one standard deviation increase in ultrasensitive troponin I, there was a 38 percent increase in the risk of a subsequent decline in LVEF [135].

In contrast, a meta-analysis of cardiac biomarkers for the detection of cardiotoxicity in childhood cancer concluded that while anthracycline chemotherapy was associated with an increased frequency of elevated troponin (OR 3.7, 95% CI 2.1-6.5), the available evidence on the association between elevated troponin and LV dysfunction was insufficient [138]. In five of the studies included in this analysis (179 patients), the frequency of LV dysfunction was not significantly increased in patients with elevated troponin (OR 2.5, 95% CI 0.5-13.2).

-Natriuretic peptides – There are conflicting data on the role of natriuretic peptides (BNP and NT-proBNP) in predicting a decline in LVEF or subsequent HF after anthracycline therapy [137-144]. In the meta-analysis described above, there were higher BNP levels posttreatment compared with controls (ie, those not receiving cardioprotectant therapies) or pretreatment values (standardized mean difference 1.0, 95% CI 0.6-1.4), and the risk for LV dysfunction was significantly higher in patients with elevated BNP levels (OR 7.1, 95% CI 2.0-25.5). However, the sensitivity of BNP for the detection of LV dysfunction was only 33 percent, while the specificity was 92 percent. Sensitivity increased when selecting for studies that assessed patients within five years after anthracycline exposure and for studies including high cumulative anthracycline doses.

When can monitoring be discontinued? — Monitoring should occur immediately after the completion of anthracycline therapy and annually for life and should include a review of systems and physical examination. In the absence of symptoms or physical findings to suggest cardiomyopathy, there are no clear data to support routine imaging or blood work.

Consensus-based guidelines for posttreatment cardiomyopathy surveillance are available for both adult and pediatric cancer survivors, but they do not address when to discontinue monitoring:

●ASCO guidelines support periodic history and physical examination after treatment in patients at risk for cardiac dysfunction, and specifically recommend an echocardiogram at 6 to 12 months after completion of therapy in patients considered at high risk for cardiac toxicity [18]. (See 'Risk factors' above.)

Beyond early posttreatment screening, they state that no recommendations can be made regarding the frequency or duration of surveillance for patients who are at high risk or asymptomatic and have no evidence of cardiac dysfunction on the immediate posttreatment echocardiogram.

●The long-term follow-up for survivors of childhood cancer is discussed in detail separately. (See "Cancer survivorship: Cardiovascular and respiratory issues".)

Therapies of unclear benefit — Strategies and therapies of unclear benefit include the following:

Alternative anthracycline-type drugs — In some cases, such as metastatic breast cancer, patients who are responding to a doxorubicin-based regimen and who are approaching lifetime cumulative doses (>400 mg/m²) may be switched to an alternative, less cardiotoxic preparation, such as pegylated liposomal doxorubicin.

Otherwise, for patients with advanced, anthracycline-sensitive cancers, we do not substitute other anthracycline-type drugs, such as epirubicin, for doxorubicin in an attempt to minimize cardiotoxicity. (See 'Modification of dose, specific agent, and schedule of administration' above.)

While it has long been thought that epirubicin confers a lower risk of cardiotoxicity compared with doxorubicin, this is only true when the drugs are compared on a mg-for-mg basis [42]. Most epirubicin-based regimens (which are typically used for the adjuvant treatment of breast cancer) employ a higher dose of epirubicin per cycle (eg, 100 to 120 mg/m² compared with 50 to 60 mg/m² of doxorubicin) and there are no differences between epirubicin and doxorubicin with regard to the development of clinical HF [88].

Primary prevention with cardiovascular drugs — For patients undergoing therapy with an anthracycline-type drug, drugs used to treat CVD such as beta blockers, angiotensin converting enzyme (ACE) inhibitors, and statins are not routinely indicated as preventive measures in the absence of cardiac risk factors or preexisting LV dysfunction; however, they may be appropriate for select patients who have other indications for such therapy, such as borderline or untreated hypertension. (See 'Patients with preexisting left ventricular systolic dysfunction' above.)

The evidence supporting use of agents such as beta blockers, ACE inhibitors, or angiotensin II receptor blockers for the treatment of HF with reduced ejection fraction and myocardial infarction has prompted study of the potential efficacy of such agents for the primary prevention of drug-related cardiotoxicity. However, the data for primary prevention of anthracycline cardiotoxicity with neurohormonal agents used to treat HF have conflicting results, and these agents have not demonstrated a compelling clinical benefit in trials designed to test their ability to prevent anthracycline cardiotoxicity [145]. (See 'Patients with preexisting left ventricular systolic dysfunction' above.)

Randomized trials do not suggest a large benefit of neurohormonal or statin therapy on changes in LVEF or prevention of HF:

●Beta blockers – Initial trials that included small groups of patients suggested a benefit of carvedilol [146] and nebivolol [147]. However, larger trials of metoprolol and carvedilol did not show an effect [148-150].

●Angiotensin system inhibitors – In a network meta-analysis, low-quality data suggested that angiotensin antagonists significantly reduced the risk of HF (OR 0.18, 95% CI 0.05-0.55) but not the composite of HF or LV dysfunction (OR 0.53, 95% CI 0.12-2.30) [101].

In the PRADA trial (PRevention of cArdiac Dysfunction during Adjuvant breast cancer therapy) that included 130 patients treated with epirubicin with or without trastuzumab, the differences in the overall decline of LV function between patients assigned to candesartan, metoprolol, and placebo were all less than 2 percent [149].

●Combined neurohormonal therapy – The role of combined therapy to prevent anthracycline-induced LV dysfunction is uncertain. In the OVERCOME trial, 90 patients with leukemia or who were undergoing stem-cell transplantation for a hematologic malignancy were randomly assigned to treatment with enalapril and carvedilol or to usual care (table 2) [151]. At six months, the incidence of death or HF was lower in the enalapril plus carvedilol group (7 versus 22 percent), and the average change in LVEF was greater in the control group (between-group difference 3 percent). However, the majority of patients in both groups retained an LVEF greater than 50 percent by the end of the trial and the causes of death were unclear.

●Statin therapy – We do not routinely treat all patients who will undergo anthracycline-based therapy with a statin to prevent cardiotoxicity. Some UpToDate experts do not use statins to prevent anthracycline toxicity in the absence of data showing improved survival or lower rates of HF with statin use. However, other UpToDate experts may use a statin after a discussion of the risks and benefits of statin therapy with select patients (eg, >52 years of age, presence of obesity, scheduled dose of more than 250 mg/m² doxorubicin equivalent) based on subgroup analyses of the STOP-CA trial [152]. Trials that investigated the efficacy of statins focused on the surrogate endpoint of a change in LVEF and were not designed to detect differences in incident HF or death:

•In a trial (STOP-CA) that included 300 patients with lymphoma scheduled to receive anthracycline-based therapy, patients randomly assigned to receive placebo were more likely to have a decrease in LVEF by ≥10 percent to a final LVEF <55 percent at 12-month follow-up than patients assigned to atorvastatin (22 versus 9 percent; OR 2.9, 95% CI 1.4-6.4) [152]. In an exploratory analysis, the placebo group had a decrease in LVEF of 5.4 percent, while the atorvastatin group had a decrease of 4.1 percent (difference of 1.3 percent). The trial used the gold standard of CMR imaging to quantify LVEF.

•In a trial (PREVENT) that included 279 patients who were treated with doxorubicin for either lymphoma or breast cancer and who had normal LVEF, patients randomly assigned to statin therapy or to placebo had similar decreases in LVEF after 24 months of observation (adjusted change in LVEF 3.3 versus 3.2 percent) [27]. The trial used the gold standard of CMR imaging to quantify LVEF but was limited by missing LVEF values in 36 percent of participants.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Heart failure in adults".)

SUMMARY AND RECOMMENDATIONS

●Goals of management – In patients at average risk for anthracycline-induced cardiotoxicity who will undergo anthracycline-based chemotherapy, the goals of preventive management are to mitigate the risk of cardiac injury during therapy and to detect any early signs of cardiotoxicity. (See 'Goals of preventive management' above.)

In patients who have a high risk of anthracycline-induced cardiotoxicity or who have preexisting LV dysfunction, an additional goal of preventive management is to evaluate the risks and benefits of anthracycline-based therapy to inform the choice between anthracycline-based and alternative chemotherapy regimens, when available.

●Risk factors

•Patient characteristics and underlying conditions – The best-described factors that increase the risk of cardiotoxicity are age, obesity, hypertension, and diabetes. (See 'Patient characteristics' above.)

•Type and cumulative anthracycline dose – The cardiotoxic effects of the anthracyclines are type- and dose-dependent. Cumulative anthracycline exposure is a consistent risk factor for cardiotoxicity, and the risk of toxicity increases substantially above the upper cumulative lifetime limit, which is different for each agent. (See 'Type of anthracycline' above and 'Cumulative dose' above.)

•Method of anthracycline administration – The risk of anthracycline toxicity may be decreased by infusional rather than bolus dosing, a slower rate of infusion, and liposomal rather than unencapsulated preparations. (See 'Methods of anthracycline administration' above and 'Modification of dose, specific agent, and schedule of administration' above.)

•Other cancer therapies and factors – Other factors that increase the risk of anthracycline toxicity include concomitant administration of other cancer treatments (eg, radiation, cyclophosphamide), and patient genotype. (See 'Combined treatment with other cancer therapies' above and 'Other factors' above.)

●Pretreatment assessment

•Initial assessment – All patients should undergo a history, physical examination, and ECG. The history and physical examination are focused on the detection of signs or symptoms of cardiovascular disease (CVD) including heart failure (HF), arrhythmias, valve disease, and coronary artery disease. (See 'Initial evaluation' above.)

•Left ventricular function assessment – Prior to administration of an anthracycline, the contributors to this topic routinely obtain an assessment of left ventricular (LV) function regardless of the planned number of cycles or expected cumulative anthracycline dose. (See 'Left ventricular function assessment' above.)

•Additional assessment for patients with evidence of CVD – In patients who have preexisting CVD or risk factors for anthracycline-induced cardiotoxicity identified during the pretreatment assessment (see 'Additional assessment for patients with evidence of CVD' above):

-The oncologist should assess the prognosis for the patient’s cancer with and without anthracycline-containing therapy and the availability of effective non-anthracycline-containing regimens.

-A cardiology consultation is suggested for most patients with history of CVD or preexisting evidence of LV dysfunction.

•Identification of patients at increased risk – Patients at high risk of anthracycline toxicity include (see 'Identification of patients at increased risk' above):

-Patients whose lifetime cumulative dose of anthracycline exceeds the upper limit.

-Patients who are unlikely to exceed the lifetime cumulative dose limit but who have other risk factors for toxicity (eg, low-dose radiation of the cardiac silhouette, two or more traditional CVD risk factors, age >60 years).

-Patients who will receive high-dose radiation to the cardiac silhouette.

●Preventive management of anthracycline therapy

•Using non-anthracycline-containing regimens – Where available and equally effective, a non-anthracycline regimen should be carefully considered. (See 'Using non-anthracycline-containing regimens when feasible' above.)

•Patients with preexisting left ventricular systolic dysfunction – In patients with preexisting LV systolic dysfunction, the decision to administer or withhold an anthracycline should be based on an individualized assessment of the risk and benefit of anthracycline therapy relative to any alternative non-anthracycline-containing regimen; the assessment of benefit may differ according to the disease setting (ie, potentially curative versus palliative). (See 'Patients with preexisting left ventricular systolic dysfunction' above.)

•Patients with other forms of CVD – In patients with a high risk of or preexisting coronary artery disease, diastolic dysfunction, arrythmias, valve disease, or other forms of cardiac disease (eg, pericardial disease), the approach to therapy must be individualized based on the possible risks and benefits of anthracycline therapy as determined by collaboration between the treating oncologist and a cardiologist. (See 'Patients with other forms of cardiovascular disease' above.)

•Patients without cardiovascular disease – For patients with no history of HF with baseline LV ejection fraction (LVEF) ≥50 percent, the benefits of anthracycline likely outweigh the risks, especially for early-stage cancers that are potentially curable. (See 'Patients without cardiovascular disease' above.)

●Approaches to cardiotoxicity mitigation

•Dose, agent, and schedule of administration – In patients who are at high risk of toxicity or who show signs of toxicity and for whom continued anthracycline treatment is the preferred therapy, individualized use of one or more of the following strategies may reduce the risk of toxicity (see 'Modification of dose, specific agent, and schedule of administration' above):

-Limit the cumulative dose

-Infusional rather than bolus dosing

-Use of liposomal formulations

•Use of dexrazoxane – In the setting of early stage, potentially curable cancer, most anthracycline regimens include doses that do not exceed accepted thresholds, and cardiotoxicity mitigation is not generally needed. For most adults with advanced cancer who are anticipated to exceed a cumulative dose beyond 300 mg/m² of doxorubicin (or 600 mg/m² epirubicin), a switch to a different non-anthracycline-containing regimen is a preferred strategy, if feasible. (See 'Using non-anthracycline-containing regimens when feasible' above.)

For the rare patient in whom this approach is not feasible, and in whom there is an ongoing indication to continue doxorubicin-based chemotherapy, we suggest initiation of dexrazoxane or a switch to liposomal encapsulated doxorubicin rather than continuation of unencapsulated doxorubicin alone (Grade 2B). (See 'Approaches to cardiotoxicity mitigation' above.)

Children receiving anthracycline-based chemotherapy may also be considered for dexrazoxane, though dexrazoxane use in children is typically protocolized. (See 'Use of dexrazoxane' above.)

•Monitoring during treatment – In patients who undergo monitoring for anthracycline toxicity, we typically perform serial echocardiography with calculation of two-dimensional (2D) LVEF and if available, calculation of three-dimensional (3D) LVEF and global longitudinal strain (GLS). We do not monitor for toxicity with serial assessment of troponin or B-type natriuretic peptide (BNP). (See 'Monitoring during and after treatment' above and 'When can monitoring be discontinued?' above.)

•Therapies of unclear benefit – Therapies of unclear benefit include the use of alternative anthracycline-type drugs and primary prevention with drugs typically used to treat CVD. (See 'Alternative anthracycline-type drugs' above and 'Primary prevention with cardiovascular drugs' above.)