Cardiac Toxicity


CHEMOTHERAPEUTICS


Anthracyclines


Reversible acute cardiotoxicity and delayed irreversible dilated cardiomyopathy (CM) are uncommonly seen with the use of anthracyclines. The reversible acute toxicity presents as a myocarditis and occasionally with evidence of pericarditis, and may lead to nonischemic CM (NICM) with or without concomitant arrhythmias. This acute toxicity is rarely a fatal complication of anthracycline therapy.


The delayed CM presents as fatigue, dyspnea on exertion, orthopnea with sinus tachycardia, S3 gallop, pedal edema, pleural effusions, and elevated jugular venous distention. Nevertheless, these classic clinical features of NICM are often late manifestations. In order to reduce the incidence and/or severity of delayed CM, keen clinical recognition of factors that may lead toward a NICM include monitoring cumulative dose (CD) of agent used, early detection strategies, limiting total cumulative dose, and the timely use of adjunctive cardioprotective agents or modified infusional or pegylated regimens/drugs are important.


The risk of anthracycline CM depends on CD received. Historically, a 5% risk is seen at 400 to 450 mg/m2 for doxorubicin, 900 mg/m2 for daunorubicin, 800 to 935 mg/m2 for epirubicin, and 223 mg/m2 for idarubicin.3 Cofactors for cardiotoxic risk include mediastinal irradiation, which includes the heart, older (particularly >70 years) or younger (<15 years) age, known coronary artery disease, antecedent valvular disease, and hypertension. Concurrent monoclonal antibodies (i.e., HER2 antibodies) appear to potentiate anthracycline-induced CM, whereas other chemotherapeutic agents may have independent or synergistic cardiac effects when used in combination.


The diagnosis of an NICM in the setting of a history of anthracycline exposure is generally made by comparing serial left ventricular function studies. Several modalities can be employed including multigated radionuclide imaging, two-dimensional transthoracic echocardiography, and cardiac magnetic resonance imaging (cMRI).46 Nevertheless, it is imperative that an ischemic CM be excluded. Regardless of the modality employed, a left ventricular ejection fraction (LVEF) of ≥50% is considered within normal range. A threshold CD has not been identified below which left ventricular dysfunction is not seen.7,8 A low LVEF regardless of symptom burden remains a contraindication for anthracycline use.


Echocardiograms and cMRI can also evaluate other structural changes of the myocardium in the setting of CM. Typical progression on serial imaging includes left ventricular diastolic dysfunction and, later, left ventricular systolic dysfunction, particularly affecting septal wall motion. Initially, the left ventricle is not enlarged or only moderately enlarged; with complete development of CM, there is multisegment hypokinesis and evidence of muscle wall thinning. The electrocardiogram (ECG) findings associated with anthracycline-induced CM include sinus tachycardia, low voltage, poor R-wave progression, and nonspecific T-wave changes. Even sinus tachycardia alone is a relatively late finding, such that serial ECGs are of little value in early detection.


Early detection of anthracycline-induced CM may be important in preventing overt CM. As described above measurement of LVEF by multigated radionuclide imaging, transthoracic echocardiogram, cMRI can be utilized. In two small prospective studies of transthoracic echocardiogram and cMRI, both modalities reported that it was possible to distinguish patients who had developed CM from others by LVEF measures at baseline and at 200 to 300 mg/m2 of doxorubicin.9,10 A fall of ≥10% at 200 mg/m2 had 72% specificity and 90% sensitivity in detecting later CD. At a dose of 300 mg/m2, 50% had developed a change in LVEF >10% with 25% of patients with evidence of myocardial structure changes. Long-term follow-up data is lacking in the later publication. Others have examined the strategy of serial LVEF retrospectively and demonstrated that the overall costs of early detection were less than the medical costs of overt CM management.11 Therefore, it is generally accepted that all patients prior to undergoing anthracycline-based regimens should have a baseline measure of LVEF; the modality employed is often left to the physician. Subsequent follow-up evaluation of the myocardium is often driven by symptoms as no large prospective studies have demonstrated an overall survival advantage to surveillance and are unlikely to given the low incidence of CM and high number needed to screen.


Additional serologic methods of CM detection include measurement of levels of brain natriuretic peptide, a peptide synthesized by the myocardium that reflects the degree of heart failure, and cardiac troponin T levels, a measure of active myocardial damage.12 Brain natriuretic peptide and to a lesser extent troponin T levels are encouraging serologic markers and may correlate with subclinical left ventricular diastolic dysfunction; however, prospective sequential serologic and radiographic studies are lacking. Percutaneous endomyocardial biopsy of the right ventricle, the historic gold standard for anthracycline-induced CM, may demonstrate characteristic histologic features that include loss of myofibrils, distention of the sarcoplasmic reticulum, and vacuolization of the cytoplasm; however, with more conventional technologies, this modality is rarely used.


Aside from early detection, other strategies to reduce the incidence of CM are the use of low-dose or infusional drug schedules in an attempt to reduce peak drug dose delivery, and the use of liposomal formulations.13,14 Liposomal formulations have been shown to permit higher CD and lower CM at the same dose. Data remains limited, however, regarding tumor response and survival. The iron-chelating cardioprotectant dexrazoxane decreases the risk of clinical CM in patients who have received doxorubicin doses of ≥300 mg/m2.15 The American Society of Clinical Oncology recommends the use of dexrazoxane in this setting.16 However, American Society of Clinical Oncology guidelines do not advocate dexrazoxane in the several important situations including adjuvant regimens, CD <300 mg/m2, pediatric patients, or in high-risk disease patients. Clinical data are also insufficient to recommend dexrazoxane with other anthracyclines, except epirubicin in metastatic breast cancer. A pivotal study in pediatric acute lymphoblastic leukemia may serve to change these recommendations. With a median follow-up of 5.7 years, this prospective trial revealed no significant impact on 5-year event-free survival, whereas dexrazoxane reduced serial troponins during therapy.17


The management of anthracycline-induced CM is similar to other causes of NICM.1 Angiotensin-converting enzyme inhibitors, β-blockers, and diuretics are commonly used. As pointed out in long-term pediatric studies, these agents do not cure or permanently control the CM. Rather, the CM may become progressive despite these agents after more than 5 years. The only curative therapy remains cardiac transplantation.


The mechanism of anthracycline CD is not fully elucidated, but there is emerging evidence that cardiomyocyte apoptosis may play a major role.18 The production of free radicals generated during cardiomyocyte metabolism of the anthracycline results in membrane lipid peroxidation, with the consequent activation of the extrinsic and intrinsic apoptotic pathways. It is thought that free radicals are generated by enzymatic reduction of the anthracycline quinone ring and by formation of iron-anthracycline complexes. The intrinsic antioxidant defense of the cardiomyocyte is more limited than other organs, leading to its apparent selective toxicity profile. In a knockout mouse model suggesting that a deficiency of the HFE gene (associated with hereditary hemochromatosis) confers increased susceptibility to doxorubicin CD.19 They studied wild-type, HFE (+/−), and HFE (−/−) mice that were chronically treated with doxorubicin. Survival was significantly decreased in the HFE-negative mice, and cardiac iron concentration was significantly elevated. Moreover, cardiac ultrastructure changes demonstrated iron-associated mitochondrial damage. Interestingly, in rats the site of cardiac oxidative toxicity was also the mitochondria of the cardiac cell by gene expression profiling, and using cMRI in an anthracycline-exposed mouse model, these lesions demonstrate gadolinium enhancement.20,21


Cyclophosphamide


The classic cardiac toxicity of cyclophosphamide (CTX) is an acute myopericarditis associated with high-dose therapy. In the era of stem cell transplantation and vigorous hydration, irreversible hemorrhagic myonecrosis is rare. More commonly, patients develop an acute or subacute congestive heart failure that is generally reversible with medical management.


Prospective monitoring of 16 patients during CTX high-dose administration (7 g/m2) and serial enzymes (CPK, CPK-MB, and troponin I) did not become elevated using a fractionated drug administration schedule during 13 hours.22 The only positive findings were four cases of transient, mild diastolic, and systolic left ventricular dysfunction. Moreover, others have reported on early and late cardiac toxicity in a combination regimen using a CTX used in doxorubicin-naïve patients.23 A total of 6 of 100 patients developed transient NICM. There were no acute cardiotoxic fatalities; late cardiotoxicity was not seen except in those patients requiring anthracycline salvage. Thus, with current administration guidelines used with high-dose therapy, CTX cardiotoxicity is uncommon (<10% of patients treated), generally transient, and reversible.24


Ifosfamide


Ifosfamide is an alkylating agent with similar properties to CTX. As a result of possible chemical cystitis, it is always administered with mesna. However, there is no evidence that mesna has any cardioprotective effect for CTX or ifosfamide. Similar to CTX, ifosfamide may cause a dose-related NICM, which is generally transient and reversible. In a small report, no episodes of NICM were seen at 10 g/m2 (0 of 6 patients), but at ≥16 g/m2, a NICM was more seen in 40% (6 of 15 patients).25 Clinical symptoms were delayed and presented subacutely (mean, 12 days; range, 6 to 23 days). They often resolved with medical management. These doses are rarely encountered in current clinical practice. A larger study prospectively studied high-dose CTX or ifosfamide-based regimens in 211 patients with metastatic breast cancer.26 The cardiac effects were similar with the two agents: troponin I became positive in 70 patients (33%) and remained negative in 141. LVEF fell significantly and continuously after the first month among the troponin I–positive patients, and the degree of dysfunction correlated with the troponin I level. Those who developed NICM had prior or concurrent exposure to anthracycline.


Taxanes (Paclitaxel and Docetaxel)


The taxanes (paclitaxel and docetaxel) are important antimicrotubule agents derived from the yew tree (paclitaxel, Taxus brevifolia; docetaxel, Taxus baccata). The yew tree is known to be poisonous, and the taxine alkaloid byproduct can affect cardiac conduction and automaticity. Although not fully proven, the mechanism of paclitaxel cardiotoxicity appears to be related to its taxane ring structural similarities to yew taxine.


The cardiovascular effects of paclitaxel are multiple: asymptomatic bradycardia may be documented in almost one-third of patients, hypersensitivity reactions associated with the Cremophor EL diluent (which can be ameliorated with corticosteroids, histamine H1, and H2 receptor antagonists), and most importantly, life-threatening atrial and/or ventricular rhythm disturbances and/or conduction abnormalities in approximately 0.5% of patients. Rare ischemic events have also been reported.


Life-threatening arrhythmias may occur acutely during infusion, or delayed up to 14 days after treatment. These tend to occur after two or more treatment exposures, rather than with initial therapy. There does not appear to be a cumulative dosing threshold or limit, unlike that with anthracyclines. The grade 4 and 5 adverse cardiac events, atrial arrhythmias (tachycardia, flutter, and/or fibrillation) occurred in 0.24%, ventricular arrhythmias (tachycardia/fibrillation) in 0.26%, heart block in 0.11%, and ischemia in 0.29%.27 Paclitaxel does not appear to cause an NICM.


In metastatic breast cancer, the incidence of NICM in patients treated with a taxane and anthracycline in combination has been studied. Eighty-two doxorubicin/taxane-naïve patients received doxorubicin 60 mg/m2 followed by a 1- or 3-hour paclitaxel 200 mg/m2 infusion (AT) administered every 21 days for six to seven cycles.28 The change in LVEF was a median of 10% (to 52.5%) at a CD of doxorubicin 310 to 360 mg/m2. A later phase 3 trial compared AT, doxorubicin (60 mg/m2) followed by paclitaxel (175 mg/m2 in 3-hour infusion) versus doxorubicin/CTX (60/600 mg/m2; AC) in 375 patients.29 Similarly, treatment was repeated every 3 weeks for six cycles (maximum doxorubicin, 360 mg/m2). Overt NICM was not statistically different in the two study arms (3% versus 1%; p = 0.62). However, a fall in LVEF (≥10% fall from baseline or ≥5% drop below the normal range) was significantly more frequent for AT (27%) versus AC (14%). Moreover, the risk of LVEF decrease was significantly greater for AT than AC at every cumulative doxorubicin dose level >180 mg/m2. Importantly, paclitaxel was administered within 30 minutes of preceding doxorubicin. It is now apparent that the enhanced cardiac toxicity of the doxorubicin-paclitaxel combinations is related to prolonged doxorubicin elimination, resulting in a plasma level of up to 30% greater when paclitaxel immediately precedes doxorubicin or follows it by <1 hour. This risk has been shown to be mitigated when the interval between doxorubicin and taxanes was 4 to 24 hours or longer and has remained minimal at 7-years of follow-up.3032


It is known that paclitaxel appears to facilitate the metabolic conversion of doxorubicin to the toxic metabolite doxorubicinol in a human heart in vitro model.33 The taxane docetaxel has a similar effect on doxorubicinol generation and both are attributed to allosteric interactions with cytoplasmic aldehyde reductases.


Docetaxel has not been generally associated with clinical cardiac toxicity, NICM, or enhancement of doxorubicin cardiac dysfunction. However, data suggests this is due to a lower cumulative anthracycline level rather than a different mechanism of drug interaction than paclitaxel. Both docetaxel and paclitaxel increase toxic cardiomyocyte doxorubicinol production in vitro. Therefore, the lack of docetaxel effect on doxorubicin clearance and the relatively lower doses of the drug used when combined with doxorubicin are now thought to be the explanation for the prior differences.


Fluoropyrimidines


5-Fluorouracil (5-FU) is a synthetic pyrimidine antimetabolite and an important backbone agent in many common solid tumors regimens. Its cardiac toxicity is manifested as angina, atrial/ventricular arrhythmias, myocardial infarction (MI), and cardiogenic shock. A prior history of coronary artery disease significantly increases the risk of complication (15.1% versus 1.5%, with no coronary artery disease history). Other predisposing factors include prior mediastinal irradiation and prior/concurrent exposure to other cardiac toxic medications. The reported incidence of cardiac toxicity was low at 1.6% among 1,083 patients.34 Using a continuous infusion of 5-FU noted a higher incidence of cardiac toxicity (6%) as compared with other daily schedules of administration.35 The addition of leucovorin to the continuous infusion regimen appeared to further increase the cardiotoxicity risk, but this has not been conclusively reproduced.36


The oral fluoropyrimidine capecitabine also appears to have similar frequency of cardiotoxicity. In a large retrospective study of the cardiac toxicities (grade 3 to 4) reported in four large metastatic cancer trials, it was demonstrated that in 593 patients who received bolus 5-FU-leucovorin (Mayo Clinic regimen) or capecitabine (596 patients), the incidence of cardiotoxic events was 3% and 1%, respectively.37


The mechanism of fluoropyrimidine cardiotoxicity is not well understood. It is possibly related to vascular spasm in reaction to the parent drug and its catabolites (fluoro-beta-alanine and fluoroacetate). In an in vitro model, 5-FU causes vasoconstriction in smooth muscle rings, which is reversible with nitrates.38,39 On electron microscopy, the changes appear to be in the small arterial endothelium.39 Although coronary angiography following the 5-FU–induced angina did not reveal ongoing cardiac spasm, and in fact on autopsy evaluation, patients were found to have myocarditis.40 Brachial artery ultrasound was performed prospectively in 60 patients receiving regimens containing 5-FU (n = 30) versus regimens not containing 5-FU (n = 30).41 Fifty percent of the patients showed artery contraction with 5-FU and none with other therapy. Interestingly, arterial tone recovered within 30 minutes, but reoccurred with retreatment in a high proportion. Pretreatment with glyceroltrinitrate, a vasodilator, abrogated the contractions in five patients.


Nevertheless, there is no known prophylactic regimen that can be provided to prevent the cardiotoxicity, and despite evidence to the contrary vasodilators do not necessarily relieve the symptoms once apparent. As a result, the optimal treatment is discontinuation of the fluoropyrimidine infusion and supportive measures with keen attention toward likely myocardial damage, with available cardiac support.42


Monoclonal Antibodies


Trastuzumab and Lapatinib


Trastuzumab and lapatinib are humanized monoclonal antibodies that target p185HER2 (ErbB2 or HER2 receptor), a transmembrane receptor tyrosine kinase of the epidermal growth factor family. This receptor protein is overexpressed or amplified in 20% to 30% of breast cancer and is associated with a poor clinical outcome. These antibodies are important agents in the management of primary and metastatic breast cancer.43


Cardiac HER-2 is essential for normal embryonic and adult cardiac development and function, respectively. Using cardiac HER2–deficient mutant mice, a dilated CM begins in the second postnatal month and extends into adulthood of affected mice.44,45 Moreover, HER2-deficient cardiomyocytes were more susceptible to anthracycline toxicity. However, it was subsequently discovered that no cardiac toxicity was seen when trastuzumab preceded anthracycline-based adjuvant therapy in a prospective study.46


Currently, trastuzumab is generally given after an anthracycline-based regimen. Evidence that supported that conclusion came from an independent Cardiac Review and Evaluation Committee that retrospectively evaluated data from seven phase 2/3 clinical trials.47 They identified 112 patients with a CM (as defined by the Cardiac Review and Evaluation Committee) among 1,219 treated patients. Three studies used trastuzumab alone: 383 patients, of whom 17 (4%) had a CM. Among 114 patients who received single-agent trastuzumab as first-line therapy for metastatic disease, 3% developed a CM. In contrast, concurrent trastuzumab with anthracycline-based regimens or paclitaxel had an increased incidence of CM (27% versus 13%, respectively) as compared to chemotherapy alone (11% versus 1%, respectively) or antibody alone (3%).


The degree of cardiac impairment (New York Heart Association class III or IV) among those with CM was greatest for patients receiving trastuzumab plus concurrent anthracycline (64%) as compared with 20% receiving trastuzumab plus paclitaxel. Moreover, all trastuzumab plus paclitaxel patients had functional cardiac recovery with treatment as compared with only half the trastuzumab plus concurrent anthracycline patients. Risk factors for CM included older age and cumulative doxorubicin 300 mg/m2 or more. Concurrent anthracycline (doxorubicin or epirubicin) appeared to be more hazardous than temporally separated trastuzumab therapy.


The diagnosis of trastuzumab-induced CM is often made by the detection of an asymptomatic decrease in LVEF and now considered to be an “on-target” adverse event. Similarly to anthracycline-induced CM, tachycardia may be an early clinical indicator, and the late constellation of a dilated and hypokinetic myocardium is a late manifestation. Dissimilar to anthracyclines, trastuzumab-induced CM does not depend on the CD of the antibody and is often self-limited with supportive measure. For example, with echocardiography follow-up, the use of adjuvant trastuzumab had no additive adverse effect when used after anthracycline therapy.48 To date, there remain no definitive predictive tests other than documenting a low LVEF prior to therapy and no preventive measures to ameliorate the development of trastuzumab-induced CM. Lastly, concurrent trastuzumab-anthracycline exposure is contraindicated. Guidelines have restricted eligibility to those with LVEF of ≥55% and delay of initiation to 1 to 2 months if following anthracyclines.48 In such criteria, trastuzumab was discontinued in only 4.3% of such patients for CM. In practice, active monitoring of LVEF is recommended every 4 months during therapy and every 6 months thereafter for up to 2 years.


Because of the previously described cardiac toxicity associated with trastuzumab, T-DM1, a recently approved antibody drug conjugate against HER-2 and used in HER-2–positive breast cancer while in development, was carefully assessed for CM by surveillance imaging. In 882 patients treated with T-DM1, defined cardiac-specific events were seen in only 1.5% (13 patients) with nearly all of the events being low grade.49 Further clinical testing in combination strategies are ongoing, but lessons learned from the trastuzumab era will hopefully reduce the incidence of CM as the drug moves forward into the up-front setting.


Rituximab


Rituximab is a CD20 monoclonal antibody widely used in the management of non-Hodgkin’s lymphoma (HL). In early development of the antibody, an initial concern for CM was seen that was felt to be related to the rate of infusion; however, those findings went unfounded when used in combination regimens.50 There have been no long-term effects from rituximab; however, arrhythmias with cardiac death have been reported related to infusion reaction most commonly seen with the initial infusion but at a rate of <0.07%.51,52


Bevacizumab


Bevacizumab is a humanized monoclonal antibody that inhibits the vascular endothelial growth factor. Currently, bevacizumab is approved for use in combination with 5-FU–based chemotherapy in metastatic colon cancer. A recent US Food and Drug Administration black box warning was placed on bevacizumab after an increase in risk of MI, angina, and heart disease. The risk was greatest in those who are older than 65 with prior arterial events. The risk of grade 2 to 4 CM was 1.2% when used as a single agent, but increased to 3.8% in those with prior anthracycline exposure.


RADIOTHERAPY-ASSOCIATED CARDIAC SEQUELAE


In the past, cardiac complications resulting from mediastinal irradiation were considered rare and insignificant.53 Since the mid-1960s, when follow-up information on a large number of patients who had been cured of HL with higher doses of radiation became available, the heart has no longer been considered radioresistant.54 Radiation-associated heart disease has now been clinically described and the pathologic features of the damage have been described with regard to the coronary arteries and all three layers of the heart.


Pericarditis and pericardial effusion have been regarded as the most common side effects of cardiac irradiation.55 However, modern techniques of irradiation, dose decrease, and reduction of the heart volume irradiated in most malignancies have substantially reduced the frequency of this complication. At the same time, evidence has accumulated to suggest that ischemic heart disease resulting from radiation-induced coronary heart disease (CHD) is the most concerning long-term risk of cardiac irradiation, particularly in patients at high-risk for ischemic disease.5660


The clinical spectrum of radiation-induced heart disease involves most structures of the heart and is summarized as follows:


Pericardial disease


Acute pericarditis during irradiation


Delayed acute pericarditis


Pericardial effusion


Constrictive pericarditis


Myocardial dysfunction


Valvular heart disease


Electrical conduction abnormalities


CHD


Although the pathologic and clinical manifestations of radiation-induced heart disease may overlap in many patients, they are discussed separately in the following paragraphs.


Pericardial Disease


Incidence


The risk of radiation-induced pericardial disease depends on the dose given and on the volume of the heart irradiated.61,62 Even when a large volume of the heart (≥60%) is irradiated ≤40 Gy, the risk for mild pericarditis is <5%, and severe pericarditis is rare.63 Smaller heart volumes (20% to 30%) may tolerate up to 60 Gy, with an expected 2% risk of mild pericarditis. In the past, when the whole pericardium was irradiated, the pericarditis incidence was 20%, but when most of the left ventricle was excluded, it was reduced to 7%.64 An updated analysis of these patients showed a sharp decrease in the risk of death from cardiac complications other than acute MI for patients who received mediastinal irradiation after 1972, reflecting the change in cardiac shielding.59 Indeed, all series that showed a high risk of pericarditis are of patients treated with a radiation technique, energy, and total dose and fractionation that are no longer considered to be an acceptable standard of care in most centers.58 With current RT techniques for HL and breast cancer, pericarditis is an infrequent event.


Acute Pericarditis During Radiation

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Jun 18, 2016 | Posted by in ONCOLOGY | Comments Off on Cardiac Toxicity

Full access? Get Clinical Tree

Get Clinical Tree app for offline access