Anthracyclines: Cardiotoxicity and Its Prevention

Archives ofDisease in Childhood 1994; 71: 457-462 457

CURRENT TOPIC Arch Dis Child: first published as 10.1136/adc.71.5.457 on 1 November 1994. Downloaded from

Anthracyclines: cardiotoxicity and its prevention

J P Hale, I J Lewis

The anthracycline antibiotics have now been in administration of anthracyclines was first use in curative chemotherapy protocols for recognised when patients presented clinically over two decades. They are highly active with symptoms of congestive heart failure5 against a wide range of haemopoietic and solid within six months of their last dose of anthra- tumours in children and adults. Unfortunately, cycline. The outcome was often fatal and their antineoplastic use is limited by significant necropsy showed the cytoplasmic vacuolisation and irreversible cardiotoxicity, the full extent and myocyte necrosis since recognised as of which continues to emerge. As more child- typical of anthracycline damage.5 This review ren become long term survivors of cancer, the will focus on late toxicity. relative impact of anthracycline induced cardiac damage has increased but, despite an understanding of the pathogenesis, consensus Risk factors and incidence strategies for detection and prevention have The prime factor determining the risk of devel- not been established. oping anthracycline induced cardiomyopathy is the cumulative dose of anthracycline received.5 Estimates of the incidence of Historical perspective cardiomyopathy vary depending on the end Daunorubicin, the first anthracycline anti- point being used. Von Hoff and coworkers' biotic to be used, was isolated in Italy in 1963 large series from the 1970s7 8 using early from cultures of Streptomyces peucetius.I In clinically overt congestive cardiac failure as phase II trials it showed activity against acute the end point estimated the incidence of lymphoblastic leukaemia and acute myeloid daunorubicin induced cardiomyopathy to be leukaemia. Doxorubicin, isolated shortly 2% at 600 mg/M2 and 17% at 1050 mg/M2. afterwards,2 showed demonstrable The for doxorubicin were activity equivalent figures http://adc.bmj.com/ against a wider range of tumours in children, 7% at 550 mg/mi2 and 18% at 700 mg/M2. The including soft tissue and bone sarcomas, rapid increase in toxicity above 550 mg/M2 led Wilms' tumour and lymphomas, as well as to this being used as a limiting dose for the two lymphoblastic and myeloid leukaemia. Early drugs. phase I trials involved patients with wide- The risk of cardiotoxicity is potentiated by spread disease and reports of cardiotoxicity3 mediastinal radiation,9 advancing age,8 9 and could only uncertainly attribute toxicity to the pre-existing cardiac disease or hypertension,8 drug rather than the disease. With increasing although the effects of advancing age and on September 25, 2021 by guest. Protected copyright. use of anthracyclines, however, it became cardiac disease may not be independent. The apparent that daunorubicin and doxorubicin effect of young age at the time of treatment is were directly cardiotoxic in children4 and less certain. Von Hoff and coworkers7 8 found adults.5 that children younger than 15 years had a Originally cardiac toxicity was described higher risk of developing cardiac failure than as being acute or subacute and was seen adults with daunorubicin and doxorubicin. during or immediately after one dose or Within a paediatric group Lipshultz et al course of anthracycline. Additionally, late or found an increased risk for children treated chronic toxicity occurred after prolonged under 4 years of age of developing subclinical administration of anthracycline.6 Acute cardiomyopathy with long follow up,10 but toxicity was seen as transient changes on this was not confirmed by Steinherz et al.11 electrocardiograms, typically ST and T Although early studies found no effect of wave flattening, and arrhythmias, typically gender, it has been suggested that girls may be sinus tachycardia.5 7 These changes usually at increased risk of cardiotoxicity. 12 reverted to normal, were independent of dose, The increasing numbers of long term and were mainly seen in adults. The sub- survivors and the availability of routine echo- Regional Centre for acute toxicity was manifest as a pericarditis- cardiography in the late 1 970s meant that Paediatric Oncology, myocarditis syndrome or as acute left reports began to appear of cardiac decompen- St James's University ventricular failure6 and was particularly seen sation occurring for the first time 10 years after Hospital, Beckett in early trials of anthracyclines in when Street, Leeds LS9 7TF adults initial treatment13 and of subclinical cardio- J P Hale scheduling led to doses as high as 160 mg/m2 myopathy.14 Steinherz et al assessed cardiac I J Lewis being delivered over times as short as four function by measuring fractional shortening in Correspondence to: days. patients treated in childhood up to 20 years Dr Lewis. Late cardiotoxicity developing after repeated from completing treatment and found an 458 Hale, Lewis

overall 23% incidence of abnormal cardiac method of cardiac assessment which detects

function."1 This increased to 63% in the early cardiac damage, is easily performed, and Arch Dis Child: first published as 10.1136/adc.71.5.457 on 1 November 1994. Downloaded from subgroup who had received more than 500 has predictive power. mg/m2 and had been followed up for more The electrocardiogram was the earliest than 10 years. Late decompensation as a method used for monitoring toxicity, but the result of pregnancy or weight training was transient changes seen with acute toxicity had well recognised. Using more sensitive load little predictive value for the development of independent echocardiographic measures of cardiomyopathy.7 Schwartz et al suggest that cardiac function, Lipshultz et al found an the prolongation of the QT interval (QTc) is a overall incidence of 57% abnormalities,10 useful screening test for early cardiac damage with up to 15 years' follow up. The full extent when echocardiography remains normal,'9 but of the problem and the functional significance its relation with clinical outcome is unproved. of asymptomatic abnormalities is as yet In Stanford, Billingham and coworkers unclear. pioneered the development of taking right ventricular endomyocardial biopsy samples and a biopsy scoring system which, in the Pathology hands of an experienced pathologist, correlates Electron microscopic examination of endo- well and linearly with the cumulative anthra- myocardial biopsy samples from patients cycline dose.9 Using endomyocardial biopsy treated with anthracyclines shows myofibril samples it is possible to predict the rate of loss, swelling ofthe sarcoplasmic reticulum and progression ofdamage for any individual,20 but mitochondria, cytoplasmic vacuolisation, and there is an associated morbidity and therefore widespread damage with necrosis ofmyocytes.9 there are concerns about the invasiveness of These changes may be focal or diffuse. repeated treating in children.2' The number of myocytes present in the adult Echocardiographic measurement of frac- heart is established by the age of 6 months. tional shortening and ejection fraction as Myocytes lost by necrosis are not replaced, but indicators of left ventricular systolic function is the remaining myocytes increase in size to com- the most widely used method of assessing pensate. There is also an increase in the amount cardiotoxicity, but is complicated by the ability of interstitial tissue and fibrosis. The changes of the myocardium to compensate for early seen in children and adults are the same. damage. The sensitivity of systolic echocardi- For children sustaining anthracycline ography can be improved by exercise testing22 cardiotoxicity, myocyte loss means the left when failure of the fractional shortening to ventricular wall becomes relatively thin and increase in response to exertion may indicate particularly fails to keep pace with pubertal or early cardiomyopathy. Similar, asympto- growth hormone induced growth spurts,1013 matic cardiomyopathy can be unmasked by occasionally causing subclinical damage to echocardiography during low dose dobut- become clinically overt at this time. amine infusions at a time when other physio- logical studies are normal.23 http://adc.bmj.com/ Fractional shortening is affected by variables Mechanism ofdamage such as anaemia, fever, and volume infusions At the current level of understanding the that influence left ventricular loading condi- mechanisms by which anthracyclines produce tions and interfere with the relation between antitumour activity and cardiotoxicity appear fractional shortening and contractility. Atten- different. The antitumour activity of anthra- tion has focused on load independent echo- cyclines seems to be exerted by intercalation in cardiographic measures of systolic function on September 25, 2021 by guest. Protected copyright. double stranded DNA and stabilisation of the such as end systolic wall stress and the stress- topoisomerase II-DNA complex, which is an velocity index. These are sensitive indicators of essential stage of DNA replication, resulting cardiac dysfunction,'024 but are complicated in DNA strand breaks.15 In contrast, the to measure and require cooperation in young cardiotoxicity of anthracyclines is almost children. Follow up studies using these certainly mediated by free radical damage measures10 show increasing end systolic wall which is not involved in the antitumour effect. stress with increasing anthracycline dose. This Anthracyclines, particularly when complexed is the result of increased left ventricular with iron, can generate superoxide and afterload, which is the result of reduced left hydroxyl radicals either by redox cycling or ventricular wall thickness and seems a particu- intramolecular reduction of the chelated lar problem in children treated at young age. iron.16 These free radicals then cause lipid Direct measurement of the left ventricular peroxidation'7 and damage a variety of cellular posterior wall dimension and thickening has membranes, causing cell death. The heart is also proved a sensitive indicator of asympto- thought to be particularly susceptible to this matic damage.23 The value of these measures damage because it has a large number of in prospective studies remains to be proved. mitochondria, which are a site of free radical There has been interest in echocardio- generation, and because it has low levels of graphic measurement of left ventricular antioxidant enzymes.'8 diastolic function, in particular early diastolic filling, as a more sensitive measure of cardiac damage.24 25 Again this is awkward to measure Detection of cardiotoxicity in children and comparison with stressed Detection of cardiac damage at the point of systolic assessment has shown diastolic cardiac failure is too late. There is a need for a measures to be less sensitive.23 Anthracyclines: cardiotoxicity and its prevention 459

Left ventricular contractility can also be tailoring of cumulative anthracycline dose to

measured by radionuclide angiography. Two individual tolerance. Arch Dis Child: first published as 10.1136/adc.71.5.457 on 1 November 1994. Downloaded from studies in adults have used radionuclide Endomyocardial biopsy samples20 and angiography to measure ejection fraction radionuclide angiography25 scans have been prospectively26 27 and developed guidelines to used in prospective monitoring studies and in tailor anthracycline dose to individual both instances management protocols were tolerance, thereby preventing early onset devised which limited anthracycline dose cardiac failure. Radionuclide angiography has based on changes in the biopsy score or not been used in any large paediatric series ejection fraction. The incidence of early onset and accuracy might be expected to be more of cardiac failure was successfully reduced but, as a problem in children where movement and is now understood, this is only a small part of anatomical considerations make the images the anthracycline cardiotoxicity problem. less clear. Echocardiography is widely used for moni- Overall, the issue of selecting an ideal toring in paediatric practice. Because of the monitoring investigation is unresolved. patchy nature of early cardiac monitoring there Endomyocardial biopsy probably remains the are few serial data with long follow up to allow investigation with the best linear correlation predictions of risk to be developed. Steinherz with cardiac toxicity and power of predicting et alII and Lipshultz,'I however, showed that the onset of early cardiac failure. It has not fractional shortening at the end of treatment been validated against the risk of developing correlated fairly well with fractional shortening late cardiac decompensation or subclinical at long term follow up. In Steinherz's series cardiomyopathy, however. On the other hand, only 1 1/% of children with normal end of treat- load independent and stressed measurement ment fractional shortening developed mildly of systolic function can detect a large amount abnormal fractional shortening at follow up. of subclinical cardiomyopathy, but the impor- For children with abnormal end of treatment tance ofthis for future cardiac function has not fractional shortening there was some con- been determined and the predictive power of tinued deterioration. Unfortunately, fractional this measure has not been validated prospec- shortening does not necessarily change in a tively. Fractional shortening is easily measured linear fashion with increasing dose, but the and testing is widely available. Increasing its Children's Cancer Study Group (CCSG) have sensitivity by exercise stressing may well make proposed guidelines for dose limitation based it the best measure to use prospectively to on fractional shortening monitoring.28 monitor cardiotoxicity and limit future cardiac dysfunction. ANTHRACYCLINE ANALOGUES The search for analogues of doxorubicin and Prevention of cardiotoxicity daunorubicin with a better therapeutic index DOSE LIMITATION has been relatively disappointing. Epirubicin, The earliest strategy of cardioprotection idarubicin, and aclarubicin A are semisynthetic http://adc.bmj.com/ adopted was limitation of the cumulative dose anthraquinones closely related structurally to of anthracycline received. Studies showing doxorubicin and daunorubicin. All three a small risk of early congestive heart failure drugs are effective antitumour agents, with below cumulative doses of 550 mg/M2 7 8 epirubicin being similar to doxorubicin in the meant that this was used as a safe dose. range of tumours affected and idarubicin and Current paediatric protocols rarely use total aclarubicin A being active in acute myeloid and anthracycline doses greater than 450 mg/m2 and lymphoblastic leukaemia. on September 25, 2021 by guest. Protected copyright. this has been one of the most important factors When epirubicin was first introduced, phase to date in limiting the incidence ofheart failure. I studies suggested it should be used in a dose The fallacy of this approach is that there is no relative to doxorubicin 1 x2: 1. This recommen- safe dose as the risk of early cardiomyopathy dation was based on the doses at which dose increases progressively and not in an all or none limiting haematological toxicity was produced. fashion. It also does not take into account the Unfortunately, epirubicin produces identical risks for subclinical cardiomyopathy, although cardiotoxicity to doxorubicin, although at the long term functional significance of this higher doses. The dose ratio at which about 5% remains unknown. congestive heart failure occurs is of the order of Variation in the individual tolerance of 1 x8-2x0: 1 (epirubicin:doxorubicin) based on anthracyclines is unacknowledged, to the data gathered by Farmitalia in 9144 patients29 detriment of patients who develop significant and compared with the series of Von Hoff cardiac damage at doses below those deemed et al 8 for doxorubicin. The lower cardiotoxicity safe, and also to the detriment of those who are of epirubicin has been confirmed by endo- more tolerant but are denied further doses of an myocardial biopsy samples.30 Analysis of effective antitumour drug above the safety several patient series using epirubicin or threshold. doxorubicin as a single drug on the same schedule in breast cancer and soft tissue sarcomas in adults suggests that epirubicin is SERIAL MONITORING equipotent with doxorubicin for antitumour The difficulty of choosing a single method for effect in these situations,3' but there are no monitoring cardiac function which is sensitive equivalent data in paediatric tumours. and predictive of future dysfunction was Furthermore, short term response rates do not addressed earlier. It would, however, allow necessarily correlate with long term outcome. 460 Hale, Lewis

The relative cardiotoxicity of idarubicin and administering doxorubicin as a prolonged

aclarubicin A remains to be established intravenous infusion. Unequivocal data Arch Dis Child: first published as 10.1136/adc.71.5.457 on 1 November 1994. Downloaded from although damage is produced by the two supporting reduced toxicity with prolonged drugs. infusions are limited, with much of the work In an attempt to synthesise an anthracycline being carried out in adults in small series analogue without cardiotoxicity the struc- with historical controls. Some of the most turally less similar anthraquinone, mitox- convincing data came from Legha et al at the antrone, was developed. It has antitumour M D Anderson Center39 where, in a non- activity in leukaemias, but again causes cardiac randomised study monitoring changes in damage with an incidence of 3% of heart endomyocardial biopsy score, reduced cardio- failure at cumulative doses greater than 160 toxicity was seen for the group receiving 24-96 mg/M2.32 Antineoplastic doses are also lower hour infusions as opposed to bolus doses. The than doxorubicin, but there has been no direct infusion time for the prolonged infusion group comparison of cardiotoxicity for equivalent of 21 patients was escalated in steps from 24 antineoplastic doses. to 96 hours. Five patients were maintained on Several synthetic anthrapyrazole drugs are 48 hour infusions, the remainder at 96 hours currently undergoing evaluation. Their imino- making it impossible to ascertain the relative quinine structure should prevent free radical toxicity of the different times, but it seems generation, so assessment of their cardiotoxic likely that extending the infusion time to 96 potential may be promising. hours is cardioprotective. Casper et al, in a prospective randomised study, showed using radionuclide angiography LIPOSOMAL ANTHRACYCLINES that doxorubicin given over 72 hours was less Investigation of liposomes as potential drug cardiotoxic than bolus injections for a given carriers began in the 1970s. Animal studies cumulative dose, but that protection was by of liposome encapsulated doxorubicin clearly no means absolute.40 Two small series with showed reduced cardiotoxicity33 and in experi- historical controls have suggested that 24 mental models antitumour activity was good, and 48 hour infusions may also be less particularly for hepatic tumours, as might be toxic.41 42 expected from the preferential hepatic and The original study of Speyer et al of six hour splenic uptake of liposomes. Liposomal infusions was uncontrolled, but again showed doxorubicin was well tolerated in phase I that prolonging doxorubicin infusion time is not clinical trials,34 although at relatively low absolutely protective, with 48% of patients doses, and has entered phase II trials.35 The showing a decrease in ejection fraction of 10% clinical role of this mode of delivering or more.43 This is comparable with the 42% anthracyclines remains to be established. showing a 10% decrease in Casper's study of 72 hour infusions. Subsequently, Shapira et al have confirmed in a controlled trial that six SCHEDULING hour infusions are less cardiotoxic than bolus http://adc.bmj.com/ Rather than reducing cardiotoxicity by doses.44 changing the form of anthracyclines, other All studies of prolonged anthracycline infu- investigators have explored the effect of alter- sions have been carried out in adults, although ing drug scheduling.36 feasibility in children has been established in The original studies of doxorubicin used some small paediatric series. No study has doses of 60-90 mg/iM2 given every three weeks. examined late cardiotoxicity, nor the relative The dose was either given as a single cardiotoxicity of different infusion times. Most on September 25, 2021 by guest. Protected copyright. intravenous bolus or fractionated over three to importantly, no study has established how six days. Initially there was no evidence that much it is necessary to prolong bolus doses to cardiotoxicity may be schedule dependent reduce cardiotoxicity. until pilot studies37 suggested that dividing An essential correlate of reducing toxicity by the dose every three weeks into weekly increasing infusion times is that antineoplastic boluses was associated with a lower incidence efficacy is preserved. In vitro studies suggested of cardiomyopathy. The retrospective study of that the efficacy of weekly or infused anthra- Von Hoffconfirmed that, independent ofother cyclines may be better than conventional risk factors, there was a significant reduction in schedules. For weekly doses there are clinical the probability of clinically overt cardio- data that efficacy in breast carcinoma and soft myopathy occurring at a cumulative dose of tissue sarcomas is maintained.45 For prolonged 550 mg/M2 when doxorubicin was given infusions there are only three controlled studies, weekly as opposed to every three weeks.8 one of which has no data on antitumour There was little difference between fraction- effect,44 one of which concludes that efficacy is ated and single doses every three weeks. similar,46 and the study of Casper et al,40 which Subsequent prospective studies,38 which suggests that the prolonged infusion group may included examination for subclinical cardio- have a worse outcome. In all other uncontrolled toxicity, also confirmed the cardioprotective studies the response rate has been compared effect of weekly bolus administration for the with historical controls. It must be concluded same dose intensity. that for adults it has not been safely established In the belief that the reduced cardiotoxicity that antineoplastic efficacy is unaltered by of weekly doxorubicin was the result of anthracycline scheduling changes, and whether reduced peak plasma concentrations, other the existing data can be extrapolated to children groups investigated enhancing this effect by with their very different tumours is uncertain. Anthracyclines: cardiotoxicity and its prevention 461

CARDIOPROTECTANTS request administration times of greater than

A further strategy to ameliorate anthracycline one hour with which we comply. We have Arch Dis Child: first published as 10.1136/adc.71.5.457 on 1 November 1994. Downloaded from induced cardiotoxicity has been to develop not been involved in studies of prolonged cardioprotective agents. ICRF-1 59 (Razoxane), anthracycline infusion. a cytotoxic chelating agent based on ethylene- (2) Standard monitoring is carried out on diaminetetra-acetic acid (EDTA), was originally all patients receiving anthracyclines using developed as a potential antineoplastic agent.47 electrocardiography and echocardiography. Although early studies were promising, phase II Wherever feasible, baseline assessment trials showed no significant antitumour activity. takes place before the initiation of treatment; Before cardiotoxicity was understood to be the however, because assessment is at another result offree radical generation by anthracycline hospital, it is not always possible to do this in iron complexes, Herman et al discovered that seriously ill children. We broadly follow CCSG ICRF-1 59 protected isolated heart preparations guidelines28 in that we obtain further echo- from acute anthracycline cardiotoxicity.48 Sub- cardiograms or electrocardiograms, or both, sequent animal experiments from Herman's before alternate courses of anthracyclines at laboratory established that ICRF-159 and its doses less than 300 mg/im2 and before every D-isomer ICRF-187 protected against acute and course above this dose level. chronic anthracycline induced cardiotoxicity. (3) We use fractional shortening as our It was postulated that ICRF-187 worked by main guide to tailoring treatment. A decrease reacting with the doxorubicin-iron complex to in fractional shortening of more than 10% chelate and remove the iron, thereby reducing from baseline or fractional shortening less than free radical formation, and there is evidence to 29% will trigger a discussion about further support this.49 ICRF-187 did not interfere anthracycline treatment. We would usually with antitumour effect in animal studies, but repeat the test one to two weeks later and did add some myelotoxicity of its own. There discontinue anthracyclines if the result is has been a suggestion in clonogenic assay confirmed. that ICRF-187 might interfere with the (4) Our follow up policy after completion of topoisomerase II inhibitory effects of anthra- treatment varies depending on the end of cyclines,50 however, this contradicts earlier treatment cardiac function. In those with a reports of ICRF-1 87 potentiating the effects of 'normal' or 'acceptable' cardiac function doxorubicin.51 Free radical scavengers such as (fractional shortening >29%), echocardi- acetylcysteine and ot-tocopherol had an in- ography is carried out at one, three, and five significant cardioprotective effect. years after the end of treatment and our aim is Phase I trials had established the dose then to repeat this every five years or sooner if limiting toxicities of ICRF-187 to be myelosup- there is any other potential problem - for pression and, in children, transient hepatic toxi- example, growth hormone treatment or city.52 Speyer et al initiated a phase III study in pregnancy. In those with abnormal end of adults with advanced breast cancer,53 which treatment function, echocardiography is clearly showed significant cardioprotection by carried out between three and six months after http://adc.bmj.com/ ICRF-187 using radionuclide angiography and the completion of treatment, at one year, and biopsy monitoring. There was no significant then at least annually for the next five years, reduction in antitumour activity and a slight although a number of children have required increase in myelosuppression. The degree of more frequent monitoring. protection against changes in ejection fraction and biopsy score was striking, allowing cumulative doxorubicin doses up to 2150 mg/iM2.54 Conclusions on September 25, 2021 by guest. Protected copyright. Unfortunately, disease progression in these For paediatric oncologists the use of anthra- patients prevented comment about long term cyclines to treat children with malignancies cardioprotection. There has been no phase III causes a dilemma. Anthracyclines undoubtedly trial in children to date, although case reports damage the myocardium irreversibly and an suggest similar cardioprotection.55 important minority of children have survived In response to the urgent need for a their tumour to die of cardiomyopathy. paediatric phase III study the Medical Furthermore, because more than 50% of Research Council and UK CCSG were hoping children with cancer become long term to open a trial to examine the cardioprotective survivors, the fact that 23-57% of those who effect of ICRF-1 87 in acute myeloid leukaemia received anthracyclines have evidence of and Ewing's sarcoma; however, this has been subclinical myocardial damage has worrying delayed by the company because of concerns implications for the future. Anthracyclines about possible interference with anthracycline form an important part of the curative activity, as outlined here. There is an urgent chemotherapy given to children with acute need to resolve this issue. myeloid leukaemia, Wilms' tumour, and sarcomas and may only be discarded at the potential cost of a decrease in survival rates. Our practice Cardioprotective strategies require careful Based on our own analysis of published reports controlled clinical trials before they are and experience, current practice is as follows: accepted into routine clinical use. (1) Doxorubicin and daunorubicin are always administered as infusions with a 1 Grein A, Spalla C, Di Marco A, et al. Decrizione e classifi- minimum infusion time of one hour. cazione di un attinomicete (Streptomyces peucetius sp Some nova) produttore di un sostanza ad attivita antitumorale: UK/European study treatment protocols la Daunomicina. G Microbiol 1963; 11: 109-15. 462 Hale, Lewis

2 Bonnadonna G, Monfardini S, DeLena M, et al. Clinical 31 Lanchbury AP, Habboubi N. Epirubicin and doxorubicin: a evaluation of Adriamycin, a new antitumour antibiotic. comparison of their characteristics, therapeutic activity BMY 1969; iii: 503-6. and toxicity. Cancer Treat Rev 1993; 19: 197-228. Arch Dis Child: first published as 10.1136/adc.71.5.457 on 1 November 1994. Downloaded from 3 Tan CT, Tanaka H, Yu KP, et al. Daunomycin: an anti- 32 Shenkenberg TD, Von HoffDD. Mitoxantrone: a new antitumour antibiotic in the treatment of neoplastic disease. cancer drug with significant clinical activity. Ann Intern Cancer 1967; 20: 333-53. Med 1986; 105: 67-81. 4 Gilladoga AC, Manuel C, Tan CT, et al. The cardiotoxicity 33 Herman EH, Rahman A, Ferrans VJ, et al. Prevention of of Adriamycin and daunomycin in children. Cancer 1976; chronic doxorubicin cardiotoxicity in beagles by liposomal 37: 1070-8. encapsulation. Cancer Res 1983; 43: 5427-32. 5 Lefrak EA, Pitha J, Rosenheim S, et al. A clinicopathological 34 Gabizon A, Peretz T, Sulkes A, et al. Systemic administra- analysis of Adriamycin cardiotoxicity. Cancer 1973; 32: tion of doxorubicin-containing liposomes in cancer 302-14. patients: a phase I study. Eur Cancer Clin Oncol 1989; 6 Bristow MR, Billingham ME, Mason JW, et al. Clinical 25: 1795-803. spectrum ofanthracycline antibiotic cardiotoxicity. Cancer 35 Treat J, Greenspan A, Forst D, et al. Antitumour activity of Treatment Reports 1978; 62: 873-9. liposome-encapsulated doxorubicin in advanced breast 7 Von Hoff DD, Rozencweig M, Layard M, et al. cancer: a phase II study. Y Natl Cancer Inst 1990; 82: Daunomycin-induced cardiotoxicity in children and 1706-10. adults. A review of 1 10 cases. Am JMed 1977; 62: 200-8. 36 Bielack SS, Erttmann R, Winkler K, Landbeck G. 8 Von Hoff DD, Layard MW, Basa P, et al. Risk factors for Doxorubicin: effect of different schedules on toxicity and doxorubicin-induced congestive heart failure. Ann Intern antitumour efficacy. Eur Y Cancer Clin Oncol 1989; 25: Med 1979; 91: 710-7. 873-82. 9 Bristow MR, Mason JW, Billingham ME, et al. Doxorubicin 37 Weiss AJ, Manthel RW. Experience with the use of adria- cardiomyopathy; evaluation by phonocardiography, mycin in combination with other anticancer agents using endomyocardial biopsy and cardiac catheterisation. Ann a weekly schedule with particular reference to lack of car- Intern Med 1978; 88: 168-75. diotoxicity. Cancer 1977; 40: 2046-52. 10 Lipshultz SE, Colan SD, Geiber RD, et al. Late cardiac 38 Torti FM, Bristow MR, Howes AE, et al. Reduced car- effects of doxorubicin therapy for acute lymphoblastic diotoxicity of doxorubicin delivered on a weekly schedule. leukaemia in childhood. NEnglJrMed 1991; 324: 808-15. Ann Intern Med 1983; 99: 745-9. 11 Steinherz LJ, Steinherz PG, Tan CTC, et al. Cardiac toxic- 39 Legha SS, Benjamin RS, Mackay B, et al. Reduction of ity 4 to 20 years after completing anthracycline therapy. doxorubicin cardiotoxicity by prolonged continuous JAMA 1991; 266: 1672-7. intravenous infusion. Ann Intern Med 1982; 96: 12 Silber JH, Jakacki RI, Larsen RU, et al. Increased risk of car- 133-9. diac dysfunction after anthracyclines in girls. Med Pediatr 40 Casper ES, Gaynor JJ, Hajdu SI, et al. A prospective Oncol 1993; 21: 477-9. randomized trial of adjuvant chemotherapy with bolus 13 Goorin AM, Chauvenet AR, Perez-Atayde AR, et al. Initial versus continuous infusion of doxorubicin in patients with congestive heart failure six to ten years after doxorubicin high-grade extremity soft tissue sarcoma and an analysis chemotherapy for childhood cancer. Y Pediatr 1990; 116: of prognostic factors. Cancer 1991; 68: 1221-9. 144-7. 41 Speyer JL, Green MD, Bottino J, et al. A phase I/II study of 14 Dresdale A, Bonow R, Wesley R, et al. Prospective evalua- 6 hour and 24 hour intravenous infusion of doxorubicin. tion of doxorubicin-induced cardiomyopathy resulting In: Muggia F, Young C, Carter S, eds. Anthracycline anti- from post-surgical adjuvant treatment ofpatients with soft biotics in chemotherapy. The Hague: Martinus Nijhoff, tissue sarcomas. Cancer 1983; 52: 51-60. 1982: 398-411. 15 Bodley A, Liu LF, Israel M, et al. DNA topoisomerase 42 Tannir N, Yap HY, Hortobagyi GH, et al. Sequential con- II-mediated interaction of doxorubicin and daunorubicin tinuous infusion with doxorubicin and vinblastine: an congeners with DNA. Cancer Res 1989; 49: 5969-78. effective chemotherapy combination for patients with 16 Keizer HG, Pinedo HM, Schuurhuis GJ, Joenje H. advanced breast cancer previously treated with cyclophos- Doxorubicin (Adriamycin): a critical review of free phamide, methotrexate, 5-FU, vincristine. Cancer radical-dependent mechanisms of cytotoxicity. Pharmacol Treatment Reports 1984; 68: 1039-41. Ther 1990; 47: 219-31. 43 Speyer JL, Green MD, Dubin N, et al. Prospective evalua- 17 Myers CE, McGuire WP, aiss RH, et al. Adriamycin: the tion of cardiotoxicity during a 6-hour doxorubicin role of lipid peroxidation in cardiac toxicity and tumour infusion regimen in women with adenocarcinoma of the response. Science 1977; 197: 165-7. breast. Am JMed 1985; 78: 555-63. 18 Doroshow JH, Locker GY, Myers CE. Enzymatic defences 44 Shapira J, Gotfried M, Lishner M, Ravid M. Reduced car- of the mouse heart against reactive oxygen metabolites; diotoxicity of doxorubicin by a 6-hour infusion regimen: a alterations produced by doxorubicin. Clin Invest 1980; prospective randomized evaluation. Cancer 1990; 65: 65: 128-35. 870-3. 19 Schwartz CL, Hobbie WL, Truesdell A, et al. Corrected QT 45 Borden Amato et EC, DA, Rosenbaum DH, al. http://adc.bmj.com/ interval prolongation in anthracycline-treated survivors of Randomized comparison of three Adriamycin regimens childhood cancer. YClin Oncol 1993; 11: 1906-10. for metastatic soft tissue sarcomas. 7 Clin Oncol 1987; 5: 20 Billingham ME, Bristow MR. Evaluation of anthracycline 840-50. cardiotoxicity: predictive ability and functional correlation 46 Zalupski M, Metch B, Balcerzak S. Phase III comparison of endomyocardial biopsy. Cancer Treatment Symposiums of doxorubicin and dacarbazine given by bolus versus 1984; 3: 71-6. infusion in patients with soft tissue sarcomas: a Southwest 21 Pegelow CH, Popper RW, de Wit SA, et al. Endo- Oncology Group study. Y Nail Cancer Inst 1991; 83: myocardial biopsy to monitor anthracycline therapy in 926-32. children. Y Clin Oncol 1984; 2: 443-6. 47 Hellman K, Newton KA, Whitmore DN, et al. Preliminary 22 Weesner KM. Bledsoe M, Chauvenet A, Wofford M. clinical assessment of ICRF-159 in acute leukaemia and Exercise echocardiography in the detection of anthracy- lymphosarcoma. BMJ 1969; i: 822-4. on September 25, 2021 by guest. Protected copyright. cline cardiotoxicity. Cancer 1991; 68: 435-8. 48 Herman EH, Mhatre RM, Lee IP, Waravdekar VS. 23 Klewer SE, Goldberg SJ, Donnerstein RL, et al. Prevention of the cardiotoxic effects of Adriamycin and Dobutamine stress echocardiography: a sensitive indica- daunomycin in the isolated dog heart. Proc Soc Exp Biol tor of diminished myocardial function in asymptomatic Med 1972; 140: 234-49. doxorubicin-treated long-term survivors of childhood 49 Rajagopalan S, Politi PM, Sinha BK, Myers CE. cancer. Am Coll Cardiol 1992; 19: 394-401. Adriamycin-induced free radical formation in the per- 24 Hausdorf G, Morf G, Beron G, et al. Long term doxoru- fused rat heart: implications for cardiotoxicity. Cancer Res bicin cardiotoxicity in childhood: non-invasive evaluation 1988; 48: 4766-9. of the contractile state and diastolic filling. Br Heart 50 Sehested M, Jensen PB, Sorensen BS, et al. Antagonistic 1988; 60: 309-15. effect of the cardioprotector (+)-1,2-bis (3,5-dioxopiper- 25 Marchandaise B, Schroeder E, Bosly A, et al. Early detec- azinyl-l-YL) propane (ICRF-187) on DNA breaks and tion of doxorubicin cardiotoxicity: interest of Doppler cytotoxicity induced by the topoisomerase II directed echocardiographic analysis of left ventricular filling drugs daunorubicin and etoposide (VP- 16). Biochem dynamics. Am HeartJ 1989; 118: 92-8. Pharmacol 1993; 46: 389-93. 26 Schwartz RG, McKenzie WB, Alexander J, et al. Congestive 51 Monti E, Sinha BK. Potentiation of doxorubicin heart failure and left ventricular dysfunction complicating cytotoxicity by (+)-1,2-bis (3,5-dioxopiperazinyl-1-yl) doxorubicin therapy: a seven year experience using serial propane (ICRF-187) in human leukamic HL-60 cells. radionuclide angiography. Am Med 1987; 82: 1109-18. Cancer Communications 1990; 2: 145-9. 27 Palmeri ST, Bonow RO, Myers CE, et al. Prospective eval- 52 HolcenbergJS, Tutsch KD, Earhart RH, et al. Phase I study uation of doxorubicin cardiotoxicity by rest and exercise of ICRF-187 in pediatric cancer patients and comparison radionuclide angiography. Am Y Cardiol 1986; 58: of its pharmacokinetics in children and adults. Cancer 607-13. Treatment Reports 1986; 70: 703-9. 28 Steinherz U, Graham T, Hurwitz R, et al. Guidelines for 53 Speyer JL, Green MD, Kramer E, et al. Protective effect of cardiac monitoring of children during and after anthracy- the bispiperazinedione ICRF-187 against doxorubicin- cline therapy: report of the Cardiology Committee of the induced cardiac toxicity in women with advanced breast Children's Cancer Study Group. Pediatrics 1992; 89: cancer. N Engl_'Med 1988; 319: 745-52. 942-9. 54 Speyer JL, Green MD, Zeleniuch-Jacquotte A, et al. ICRF- 29 Praga C, Trave F, Petroccione A. Anthracycline-induced 187 permits longer treatment with doxorubicin in women cardiotoxicity and its relevance in cancer treatment. In: with breast cancer. Clin Oncol 1992; 10: 1 7-27. Nimmo WS, Tucker GT, eds. Clinical measurement in drug 55 Bu'lock FA, Gabriel HM, Oakhill A, et al. Cardioprotection evaluation. London: Wolfe, 1991: 131-42. by ICRF-187 against high dose anthracycline toxicity in 30 Bonnadonna G, Gianni L, Santoro A, et al. Drugs ten years children with malignant disease. Br Heart 1993; 70: later: epirubicin. Ann Oncol 1993; 4: 359-69. 185-8.