anthracyclines: cardiotoxicity andits preventionand irreversible cardiotoxicity, the full extent...

Archives of Disease in Childhood 1994; 71: 457-462

CURRENT TOPIC

Anthracyclines: cardiotoxicity and its prevention

J P Hale, I J Lewis

The anthracycline antibiotics have now been inuse in curative chemotherapy protocols forover two decades. They are highly activeagainst a wide range of haemopoietic and solidtumours in children and adults. Unfortunately,their antineoplastic use is limited by significantand irreversible cardiotoxicity, the full extentof which continues to emerge. As more child-ren become long term survivors of cancer, therelative impact of anthracycline inducedcardiac damage has increased but, despite anunderstanding of the pathogenesis, consensusstrategies for detection and prevention havenot been established.

Regional Centre forPaediatric Oncology,St James's UniversityHospital, BeckettStreet, Leeds LS9 7TFJ P HaleI J Lewis

Correspondence to:Dr Lewis.

Historical perspectiveDaunorubicin, the first anthracycline anti-biotic to be used, was isolated in Italy in 1963from cultures of Streptomyces peucetius.I Inphase II trials it showed activity against acutelymphoblastic leukaemia and acute myeloidleukaemia. Doxorubicin, isolated shortlyafterwards,2 showed demonstrable activityagainst a wider range of tumours in children,including soft tissue and bone sarcomas,Wilms' tumour and lymphomas, as well as

lymphoblastic and myeloid leukaemia. Earlyphase I trials involved patients with wide-spread disease and reports of cardiotoxicity3could only uncertainly attribute toxicity to thedrug rather than the disease. With increasinguse of anthracyclines, however, it becameapparent that daunorubicin and doxorubicinwere directly cardiotoxic in children4 andadults.5

Originally cardiac toxicity was describedas being acute or subacute and was seen

during or immediately after one dose or

course of anthracycline. Additionally, late or

chronic toxicity occurred after prolongedadministration of anthracycline.6 Acutetoxicity was seen as transient changes on

electrocardiograms, typically ST and Twave flattening, and arrhythmias, typicallysinus tachycardia.5 7 These changes usuallyreverted to normal, were independent of dose,and were mainly seen in adults. The sub-acute toxicity was manifest as a pericarditis-myocarditis syndrome or as acute leftventricular failure6 and was particularly seenin early trials of anthracyclines in adults whenscheduling led to doses as high as 160 mg/m2being delivered over times as short as fourdays.

Late cardiotoxicity developing after repeated

administration of anthracyclines was firstrecognised when patients presented clinicallywith symptoms of congestive heart failure5within six months of their last dose of anthra-cycline. The outcome was often fatal andnecropsy showed the cytoplasmic vacuolisationand myocyte necrosis since recognised astypical of anthracycline damage.5 This reviewwill focus on late toxicity.

Risk factors and incidenceThe prime factor determining the risk of devel-oping anthracycline induced cardiomyopathyis the cumulative dose of anthracyclinereceived.5 Estimates of the incidence ofcardiomyopathy vary depending on the endpoint being used. Von Hoff and coworkers'large series from the 1970s7 8 using earlyclinically overt congestive cardiac failure asthe end point estimated the incidence ofdaunorubicin induced cardiomyopathy to be2% at 600 mg/M2 and 17% at 1050 mg/M2.The equivalent figures for doxorubicin were7% at 550 mg/mi2 and 18% at 700 mg/M2. Therapid increase in toxicity above 550 mg/M2 ledto this being used as a limiting dose for the twodrugs.The risk of cardiotoxicity is potentiated by

mediastinal radiation,9 advancing age,8 9 andpre-existing cardiac disease or hypertension,8although the effects of advancing age andcardiac disease may not be independent. Theeffect of young age at the time of treatment isless certain. Von Hoff and coworkers7 8 foundthat children younger than 15 years had ahigher risk of developing cardiac failure thanadults with daunorubicin and doxorubicin.Within a paediatric group Lipshultz et alfound an increased risk for children treatedunder 4 years of age of developing subclinicalcardiomyopathy with long follow up,10 butthis was not confirmed by Steinherz et al.11Although early studies found no effect ofgender, it has been suggested that girls may beat increased risk of cardiotoxicity. 12The increasing numbers of long term

survivors and the availability of routine echo-cardiography in the late 1 970s meant thatreports began to appear of cardiac decompen-sation occurring for the first time 10 years afterinitial treatment13 and of subclinical cardio-myopathy.14 Steinherz et al assessed cardiacfunction by measuring fractional shortening inpatients treated in childhood up to 20 yearsfrom completing treatment and found an

457

on August 17, 2021 by guest. P

rotected by copyright.http://adc.bm

j.com/

Arch D

is Child: first published as 10.1136/adc.71.5.457 on 1 N

ovember 1994. D

ownloaded from

http://adc.bmj.com/

Hale, Lewis

overall 23% incidence of abnormal cardiacfunction."1 This increased to 63% in thesubgroup who had received more than 500mg/m2 and had been followed up for morethan 10 years. Late decompensation as aresult of pregnancy or weight training waswell recognised. Using more sensitive loadindependent echocardiographic measures ofcardiac function, Lipshultz et al found anoverall incidence of 57% abnormalities,10with up to 15 years' follow up. The full extentof the problem and the functional significanceof asymptomatic abnormalities is as yetunclear.

PathologyElectron microscopic examination of endo-myocardial biopsy samples from patientstreated with anthracyclines shows myofibrilloss, swelling of the sarcoplasmic reticulum andmitochondria, cytoplasmic vacuolisation, andwidespread damage with necrosis of myocytes.9These changes may be focal or diffuse.The number of myocytes present in the adult

heart is established by the age of 6 months.Myocytes lost by necrosis are not replaced, butthe remaining myocytes increase in size to com-pensate. There is also an increase in the amountof interstitial tissue and fibrosis. The changesseen in children and adults are the same.For children sustaining anthracycline

cardiotoxicity, myocyte loss means the leftventricular wall becomes relatively thin andparticularly fails to keep pace with pubertal orgrowth hormone induced growth spurts,1013occasionally causing subclinical damage tobecome clinically overt at this time.

Mechanism ofdamageAt the current level of understanding themechanisms by which anthracyclines produceantitumour activity and cardiotoxicity appeardifferent. The antitumour activity of anthra-cyclines seems to be exerted by intercalation indouble stranded DNA and stabilisation of thetopoisomerase II-DNA complex, which is anessential stage of DNA replication, resultingin DNA strand breaks.15 In contrast, thecardiotoxicity of anthracyclines is almostcertainly mediated by free radical damagewhich is not involved in the antitumour effect.Anthracyclines, particularly when complexedwith iron, can generate superoxide andhydroxyl radicals either by redox cycling orintramolecular reduction of the chelatediron.16 These free radicals then cause lipidperoxidation'7 and damage a variety of cellularmembranes, causing cell death. The heart isthought to be particularly susceptible to thisdamage because it has a large number ofmitochondria, which are a site of free radicalgeneration, and because it has low levels ofantioxidant enzymes.'8

Detection of cardiotoxicityDetection of cardiac damage at the point ofcardiac failure is too late. There is a need for a

method of cardiac assessment which detectsearly cardiac damage, is easily performed, andhas predictive power.The electrocardiogram was the earliest

method used for monitoring toxicity, but thetransient changes seen with acute toxicity hadlittle predictive value for the development ofcardiomyopathy.7 Schwartz et al suggest thatthe prolongation of the QT interval (QTc) is auseful screening test for early cardiac damagewhen echocardiography remains normal,'9 butits relation with clinical outcome is unproved.

In Stanford, Billingham and coworkerspioneered the development of taking rightventricular endomyocardial biopsy samplesand a biopsy scoring system which, in thehands of an experienced pathologist, correlateswell and linearly with the cumulative anthra-cycline dose.9 Using endomyocardial biopsysamples it is possible to predict the rate ofprogression ofdamage for any individual,20 butthere is an associated morbidity and thereforethere are concerns about the invasiveness ofrepeated treating in children.2'

Echocardiographic measurement of frac-tional shortening and ejection fraction asindicators of left ventricular systolic function isthe most widely used method of assessingcardiotoxicity, but is complicated by the abilityof the myocardium to compensate for earlydamage. The sensitivity of systolic echocardi-ography can be improved by exercise testing22when failure of the fractional shortening toincrease in response to exertion may indicateearly cardiomyopathy. Similar, asympto-matic cardiomyopathy can be unmasked byechocardiography during low dose dobut-amine infusions at a time when other physio-logical studies are normal.23

Fractional shortening is affected by variablessuch as anaemia, fever, and volume infusionsthat influence left ventricular loading condi-tions and interfere with the relation betweenfractional shortening and contractility. Atten-tion has focused on load independent echo-cardiographic measures of systolic functionsuch as end systolic wall stress and the stress-velocity index. These are sensitive indicators ofcardiac dysfunction,'024 but are complicatedto measure and require cooperation in youngchildren. Follow up studies using thesemeasures10 show increasing end systolic wallstress with increasing anthracycline dose. Thisis the result of increased left ventricularafterload, which is the result of reduced leftventricular wall thickness and seems a particu-lar problem in children treated at young age.Direct measurement of the left ventricularposterior wall dimension and thickening hasalso proved a sensitive indicator of asympto-matic damage.23 The value of these measuresin prospective studies remains to be proved.

There has been interest in echocardio-graphic measurement of left ventriculardiastolic function, in particular early diastolicfilling, as a more sensitive measure of cardiacdamage.24 25 Again this is awkward to measurein children and comparison with stressedsystolic assessment has shown diastolicmeasures to be less sensitive.23

458



j.com/

Arch D


ovember 1994. D

ownloaded from

http://adc.bmj.com/


Left ventricular contractility can also bemeasured by radionuclide angiography. Twostudies in adults have used radionuclideangiography to measure ejection fractionprospectively26 27 and developed guidelines totailor anthracycline dose to individualtolerance, thereby preventing early onsetcardiac failure. Radionuclide angiography hasnot been used in any large paediatric seriesand accuracy might be expected to be more ofa problem in children where movement andanatomical considerations make the imagesless clear.

Overall, the issue of selecting an idealmonitoring investigation is unresolved.Endomyocardial biopsy probably remains theinvestigation with the best linear correlationwith cardiac toxicity and power of predictingthe onset of early cardiac failure. It has notbeen validated against the risk of developinglate cardiac decompensation or subclinicalcardiomyopathy, however. On the other hand,load independent and stressed measurementof systolic function can detect a large amountof subclinical cardiomyopathy, but the impor-tance of this for future cardiac function has notbeen determined and the predictive power ofthis measure has not been validated prospec-tively. Fractional shortening is easily measuredand testing is widely available. Increasing itssensitivity by exercise stressing may well makeit the best measure to use prospectively tomonitor cardiotoxicity and limit future cardiacdysfunction.

Prevention of cardiotoxicityDOSE LIMITATIONThe earliest strategy of cardioprotectionadopted was limitation of the cumulative doseof anthracycline received. Studies showinga small risk of early congestive heart failurebelow cumulative doses of 550 mg/M2 7 8meant that this was used as a safe dose.

Current paediatric protocols rarely use totalanthracycline doses greater than 450 mg/m2 andthis has been one of the most important factorsto date in limiting the incidence of heart failure.The fallacy of this approach is that there is nosafe dose as the risk of early cardiomyopathyincreases progressively and not in an all or nonefashion. It also does not take into account therisks for subclinical cardiomyopathy, althoughthe long term functional significance of thisremains unknown.

Variation in the individual tolerance ofanthracyclines is unacknowledged, to thedetriment of patients who develop significantcardiac damage at doses below those deemedsafe, and also to the detriment of those who aremore tolerant but are denied further doses of aneffective antitumour drug above the safetythreshold.

SERIAL MONITORINGThe difficulty of choosing a single method formonitoring cardiac function which is sensitiveand predictive of future dysfunction wasaddressed earlier. It would, however, allow

tailoring of cumulative anthracycline dose toindividual tolerance.

Endomyocardial biopsy samples20 andradionuclide angiography25 scans have beenused in prospective monitoring studies and inboth instances management protocols weredevised which limited anthracycline dosebased on changes in the biopsy score orejection fraction. The incidence of early onsetcardiac failure was successfully reduced but, asis now understood, this is only a small part ofthe anthracycline cardiotoxicity problem.

Echocardiography is widely used for moni-toring in paediatric practice. Because of thepatchy nature of early cardiac monitoring thereare few serial data with long follow up to allowpredictions of risk to be developed. Steinherzet alII and Lipshultz,'I however, showed thatfractional shortening at the end of treatmentcorrelated fairly well with fractional shorteningat long term follow up. In Steinherz's seriesonly 1 1/% of children with normal end of treat-ment fractional shortening developed mildlyabnormal fractional shortening at follow up.For children with abnormal end of treatmentfractional shortening there was some con-tinued deterioration. Unfortunately, fractionalshortening does not necessarily change in alinear fashion with increasing dose, but theChildren's Cancer Study Group (CCSG) haveproposed guidelines for dose limitation basedon fractional shortening monitoring.28

ANTHRACYCLINE ANALOGUESThe search for analogues of doxorubicin anddaunorubicin with a better therapeutic indexhas been relatively disappointing. Epirubicin,idarubicin, and aclarubicin A are semisyntheticanthraquinones closely related structurally todoxorubicin and daunorubicin. All threedrugs are effective antitumour agents, withepirubicin being similar to doxorubicin in therange of tumours affected and idarubicin andaclarubicin A being active in acute myeloid andlymphoblastic leukaemia.When epirubicin was first introduced, phase

I studies suggested it should be used in a doserelative to doxorubicin 1 x2: 1. This recommen-dation was based on the doses at which doselimiting haematological toxicity was produced.Unfortunately, epirubicin produces identicalcardiotoxicity to doxorubicin, although athigher doses. The dose ratio at which about 5%congestive heart failure occurs is of the order of1 x8-2x0: 1 (epirubicin:doxorubicin) based ondata gathered by Farmitalia in 9144 patients29and compared with the series of Von Hoffet al 8 for doxorubicin. The lower cardiotoxicityof epirubicin has been confirmed by endo-myocardial biopsy samples.30 Analysis ofseveral patient series using epirubicin ordoxorubicin as a single drug on the sameschedule in breast cancer and soft tissuesarcomas in adults suggests that epirubicin isequipotent with doxorubicin for antitumoureffect in these situations,3' but there are noequivalent data in paediatric tumours.Furthermore, short term response rates do notnecessarily correlate with long term outcome.

459



j.com/

Arch D


ovember 1994. D

ownloaded from

http://adc.bmj.com/

Hale, Lewis

The relative cardiotoxicity of idarubicin andaclarubicin A remains to be establishedalthough damage is produced by the twodrugs.

In an attempt to synthesise an anthracyclineanalogue without cardiotoxicity the struc-turally less similar anthraquinone, mitox-antrone, was developed. It has antitumouractivity in leukaemias, but again causes cardiacdamage with an incidence of 3% of heartfailure at cumulative doses greater than 160mg/M2.32 Antineoplastic doses are also lowerthan doxorubicin, but there has been no directcomparison of cardiotoxicity for equivalentantineoplastic doses.

Several synthetic anthrapyrazole drugs arecurrently undergoing evaluation. Their imino-quinine structure should prevent free radicalgeneration, so assessment of their cardiotoxicpotential may be promising.

LIPOSOMAL ANTHRACYCLINESInvestigation of liposomes as potential drugcarriers began in the 1970s. Animal studiesof liposome encapsulated doxorubicin clearlyshowed reduced cardiotoxicity33 and in experi-mental models antitumour activity was good,particularly for hepatic tumours, as might beexpected from the preferential hepatic andsplenic uptake of liposomes. Liposomaldoxorubicin was well tolerated in phase Iclinical trials,34 although at relatively lowdoses, and has entered phase II trials.35 Theclinical role of this mode of deliveringanthracyclines remains to be established.

SCHEDULINGRather than reducing cardiotoxicity bychanging the form of anthracyclines, otherinvestigators have explored the effect of alter-ing drug scheduling.36The original studies of doxorubicin used

doses of 60-90 mg/iM2 given every three weeks.The dose was either given as a singleintravenous bolus or fractionated over three tosix days. Initially there was no evidence thatcardiotoxicity may be schedule dependentuntil pilot studies37 suggested that dividingthe dose every three weeks into weeklyboluses was associated with a lower incidenceof cardiomyopathy. The retrospective study ofVon Hoff confirmed that, independent of otherrisk factors, there was a significant reduction inthe probability of clinically overt cardio-myopathy occurring at a cumulative dose of550 mg/M2 when doxorubicin was givenweekly as opposed to every three weeks.8There was little difference between fraction-ated and single doses every three weeks.Subsequent prospective studies,38 whichincluded examination for subclinical cardio-toxicity, also confirmed the cardioprotectiveeffect of weekly bolus administration for thesame dose intensity.

In the belief that the reduced cardiotoxicityof weekly doxorubicin was the result ofreduced peak plasma concentrations, othergroups investigated enhancing this effect by

administering doxorubicin as a prolongedintravenous infusion. Unequivocal datasupporting reduced toxicity with prolongedinfusions are limited, with much of the workbeing carried out in adults in small serieswith historical controls. Some of the mostconvincing data came from Legha et al at theM D Anderson Center39 where, in a non-randomised study monitoring changes inendomyocardial biopsy score, reduced cardio-toxicity was seen for the group receiving 24-96hour infusions as opposed to bolus doses. Theinfusion time for the prolonged infusion groupof 21 patients was escalated in steps from 24to 96 hours. Five patients were maintained on48 hour infusions, the remainder at 96 hoursmaking it impossible to ascertain the relativetoxicity of the different times, but it seemslikely that extending the infusion time to 96hours is cardioprotective.

Casper et al, in a prospective randomisedstudy, showed using radionuclide angiographythat doxorubicin given over 72 hours was lesscardiotoxic than bolus injections for a givencumulative dose, but that protection was byno means absolute.40 Two small series withhistorical controls have suggested that 24and 48 hour infusions may also be lesstoxic.41 42The original study of Speyer et al of six hour

infusions was uncontrolled, but again showedthat prolonging doxorubicin infusion time is notabsolutely protective, with 48% of patientsshowing a decrease in ejection fraction of 10%or more.43 This is comparable with the 42%showing a 10% decrease in Casper's study of 72hour infusions. Subsequently, Shapira et alhave confirmed in a controlled trial that sixhour infusions are less cardiotoxic than bolusdoses.44

All studies of prolonged anthracycline infu-sions have been carried out in adults, althoughfeasibility in children has been established insome small paediatric series. No study hasexamined late cardiotoxicity, nor the relativecardiotoxicity of different infusion times. Mostimportantly, no study has established howmuch it is necessary to prolong bolus doses toreduce cardiotoxicity.An essential correlate of reducing toxicity by

increasing infusion times is that antineoplasticefficacy is preserved. In vitro studies suggestedthat the efficacy of weekly or infused anthra-cyclines may be better than conventionalschedules. For weekly doses there are clinicaldata that efficacy in breast carcinoma and softtissue sarcomas is maintained.45 For prolongedinfusions there are only three controlled studies,one of which has no data on antitumoureffect,44 one of which concludes that efficacy issimilar,46 and the study of Casper et al,40 whichsuggests that the prolonged infusion group mayhave a worse outcome. In all other uncontrolledstudies the response rate has been comparedwith historical controls. It must be concludedthat for adults it has not been safely establishedthat antineoplastic efficacy is unaltered byanthracycline scheduling changes, and whetherthe existing data can be extrapolated to childrenwith their very different tumours is uncertain.

460



j.com/

Arch D


ovember 1994. D

ownloaded from

http://adc.bmj.com/


CARDIOPROTECTANTSA further strategy to ameliorate anthracyclineinduced cardiotoxicity has been to developcardioprotective agents. ICRF-1 59 (Razoxane),a cytotoxic chelating agent based on ethylene-diaminetetra-acetic acid (EDTA), was originallydeveloped as a potential antineoplastic agent.47Although early studies were promising, phase IItrials showed no significant antitumour activity.Before cardiotoxicity was understood to be theresult of free radical generation by anthracyclineiron complexes, Herman et al discovered thatICRF-1 59 protected isolated heart preparationsfrom acute anthracycline cardiotoxicity.48 Sub-sequent animal experiments from Herman'slaboratory established that ICRF-159 and itsD-isomer ICRF-187 protected against acute andchronic anthracycline induced cardiotoxicity.

It was postulated that ICRF-187 worked byreacting with the doxorubicin-iron complex tochelate and remove the iron, thereby reducingfree radical formation, and there is evidence tosupport this.49 ICRF-187 did not interferewith antitumour effect in animal studies, butdid add some myelotoxicity of its own. Therehas been a suggestion in clonogenic assaythat ICRF-187 might interfere with thetopoisomerase II inhibitory effects of anthra-cyclines,50 however, this contradicts earlierreports of ICRF-1 87 potentiating the effects ofdoxorubicin.51 Free radical scavengers such as

acetylcysteine and ot-tocopherol had an in-significant cardioprotective effect.

Phase I trials had established the doselimiting toxicities of ICRF-187 to be myelosup-pression and, in children, transient hepatic toxi-city.52 Speyer et al initiated a phase III study inadults with advanced breast cancer,53 whichclearly showed significant cardioprotection byICRF-187 using radionuclide angiography andbiopsy monitoring. There was no significantreduction in antitumour activity and a slightincrease in myelosuppression. The degree ofprotection against changes in ejection fractionand biopsy score was striking, allowing cumula-tive doxorubicin doses up to 2150 mg/iM2.54Unfortunately, disease progression in thesepatients prevented comment about long termcardioprotection. There has been no phase IIItrial in children to date, although case reportssuggest similar cardioprotection.55

In response to the urgent need for a

paediatric phase III study the MedicalResearch Council and UK CCSG were hopingto open a trial to examine the cardioprotectiveeffect of ICRF-1 87 in acute myeloid leukaemiaand Ewing's sarcoma; however, this has beendelayed by the company because of concerns

about possible interference with anthracyclineactivity, as outlined here. There is an urgentneed to resolve this issue.

Our practice

Based on our own analysis of published reportsand experience, current practice is as follows:

(1) Doxorubicin and daunorubicin are

always administered as infusions with a

minimum infusion time of one hour. SomeUK/European study treatment protocols

request administration times of greater thanone hour with which we comply. We havenot been involved in studies of prolongedanthracycline infusion.

(2) Standard monitoring is carried out onall patients receiving anthracyclines usingelectrocardiography and echocardiography.Wherever feasible, baseline assessmenttakes place before the initiation of treatment;however, because assessment is at anotherhospital, it is not always possible to do this inseriously ill children. We broadly follow CCSGguidelines28 in that we obtain further echo-cardiograms or electrocardiograms, or both,before alternate courses of anthracyclines atdoses less than 300 mg/im2 and before everycourse above this dose level.

(3) We use fractional shortening as ourmain guide to tailoring treatment. A decreasein fractional shortening of more than 10%from baseline or fractional shortening less than29% will trigger a discussion about furtheranthracycline treatment. We would usuallyrepeat the test one to two weeks later anddiscontinue anthracyclines if the result isconfirmed.

(4) Our follow up policy after completion oftreatment varies depending on the end oftreatment cardiac function. In those with a'normal' or 'acceptable' cardiac function(fractional shortening >29%), echocardi-ography is carried out at one, three, and fiveyears after the end of treatment and our aim isthen to repeat this every five years or sooner ifthere is any other potential problem - forexample, growth hormone treatment orpregnancy. In those with abnormal end oftreatment function, echocardiography iscarried out between three and six months afterthe completion of treatment, at one year, andthen at least annually for the next five years,although a number of children have requiredmore frequent monitoring.

ConclusionsFor paediatric oncologists the use of anthra-cyclines to treat children with malignanciescauses a dilemma. Anthracyclines undoubtedlydamage the myocardium irreversibly and animportant minority of children have survivedtheir tumour to die of cardiomyopathy.Furthermore, because more than 50% ofchildren with cancer become long termsurvivors, the fact that 23-57% of those whoreceived anthracyclines have evidence ofsubclinical myocardial damage has worryingimplications for the future. Anthracyclinesform an important part of the curativechemotherapy given to children with acutemyeloid leukaemia, Wilms' tumour, andsarcomas and may only be discarded at thepotential cost of a decrease in survival rates.Cardioprotective strategies require carefulcontrolled clinical trials before they areaccepted into routine clinical use.

1 Grein A, Spalla C, Di Marco A, et al. Decrizione e classifi-cazione di un attinomicete (Streptomyces peucetius spnova) produttore di un sostanza ad attivita antitumorale:la Daunomicina. G Microbiol 1963; 11: 109-15.

461



j.com/

Arch D


ovember 1994. D

ownloaded from

http://adc.bmj.com/

Hale, Lewis

2 Bonnadonna G, Monfardini S, DeLena M, et al. Clinicalevaluation of Adriamycin, a new antitumour antibiotic.BMY 1969; iii: 503-6.

3 Tan CT, Tanaka H, Yu KP, et al. Daunomycin: an anti-tumour antibiotic in the treatment of neoplastic disease.Cancer 1967; 20: 333-53.

4 Gilladoga AC, Manuel C, Tan CT, et al. The cardiotoxicityof Adriamycin and daunomycin in children. Cancer 1976;37: 1070-8.

5 Lefrak EA, Pitha J, Rosenheim S, et al. A clinicopathologicalanalysis of Adriamycin cardiotoxicity. Cancer 1973; 32:302-14.

6 Bristow MR, Billingham ME, Mason JW, et al. Clinicalspectrum ofanthracycline antibiotic cardiotoxicity. CancerTreatment Reports 1978; 62: 873-9.

7 Von Hoff DD, Rozencweig M, Layard M, et al.Daunomycin-induced cardiotoxicity in children andadults. A review of 1 10 cases. Am JMed 1977; 62: 200-8.

8 Von Hoff DD, Layard MW, Basa P, et al. Risk factors fordoxorubicin-induced congestive heart failure. Ann InternMed 1979; 91: 710-7.

9 Bristow MR, Mason JW, Billingham ME, et al. Doxorubicincardiomyopathy; evaluation by phonocardiography,endomyocardial biopsy and cardiac catheterisation. AnnIntern Med 1978; 88: 168-75.

10 Lipshultz SE, Colan SD, Geiber RD, et al. Late cardiaceffects of doxorubicin therapy for acute lymphoblasticleukaemia in childhood. NEnglJrMed 1991; 324: 808-15.

11 Steinherz LJ, Steinherz PG, Tan CTC, et al. Cardiac toxic-ity 4 to 20 years after completing anthracycline therapy.JAMA 1991; 266: 1672-7.

12 Silber JH, Jakacki RI, Larsen RU, et al. Increased risk of car-diac dysfunction after anthracyclines in girls. Med PediatrOncol 1993; 21: 477-9.

13 Goorin AM, Chauvenet AR, Perez-Atayde AR, et al. Initialcongestive heart failure six to ten years after doxorubicinchemotherapy for childhood cancer. Y Pediatr 1990; 116:144-7.

14 Dresdale A, Bonow R, Wesley R, et al. Prospective evalua-tion of doxorubicin-induced cardiomyopathy resultingfrom post-surgical adjuvant treatment ofpatients with softtissue sarcomas. Cancer 1983; 52: 51-60.

15 Bodley A, Liu LF, Israel M, et al. DNA topoisomeraseII-mediated interaction of doxorubicin and daunorubicincongeners with DNA. Cancer Res 1989; 49: 5969-78.

16 Keizer HG, Pinedo HM, Schuurhuis GJ, Joenje H.Doxorubicin (Adriamycin): a critical review of freeradical-dependent mechanisms of cytotoxicity. PharmacolTher 1990; 47: 219-31.

17 Myers CE, McGuire WP, aiss RH, et al. Adriamycin: therole of lipid peroxidation in cardiac toxicity and tumourresponse. Science 1977; 197: 165-7.

18 Doroshow JH, Locker GY, Myers CE. Enzymatic defencesof the mouse heart against reactive oxygen metabolites;alterations produced by doxorubicin. Clin Invest 1980;65: 128-35.

19 Schwartz CL, Hobbie WL, Truesdell A, et al. Corrected QTinterval prolongation in anthracycline-treated survivors ofchildhood cancer. Y Clin Oncol 1993; 11: 1906-10.

20 Billingham ME, Bristow MR. Evaluation of anthracyclinecardiotoxicity: predictive ability and functional correlationof endomyocardial biopsy. Cancer Treatment Symposiums1984; 3: 71-6.

21 Pegelow CH, Popper RW, de Wit SA, et al. Endo-myocardial biopsy to monitor anthracycline therapy inchildren. Y Clin Oncol 1984; 2: 443-6.

22 Weesner KM. Bledsoe M, Chauvenet A, Wofford M.Exercise echocardiography in the detection of anthracy-cline cardiotoxicity. Cancer 1991; 68: 435-8.

23 Klewer SE, Goldberg SJ, Donnerstein RL, et al.Dobutamine stress echocardiography: a sensitive indica-tor of diminished myocardial function in asymptomaticdoxorubicin-treated long-term survivors of childhoodcancer. Am Coll Cardiol 1992; 19: 394-401.

24 Hausdorf G, Morf G, Beron G, et al. Long term doxoru-bicin cardiotoxicity in childhood: non-invasive evaluationof the contractile state and diastolic filling. Br Heart1988; 60: 309-15.

25 Marchandaise B, Schroeder E, Bosly A, et al. Early detec-tion of doxorubicin cardiotoxicity: interest of Dopplerechocardiographic analysis of left ventricular fillingdynamics. Am HeartJ 1989; 118: 92-8.

26 Schwartz RG, McKenzie WB, Alexander J, et al. Congestiveheart failure and left ventricular dysfunction complicatingdoxorubicin therapy: a seven year experience using serialradionuclide angiography. Am Med 1987; 82: 1109-18.

27 Palmeri ST, Bonow RO, Myers CE, et al. Prospective eval-uation of doxorubicin cardiotoxicity by rest and exerciseradionuclide angiography. Am Y Cardiol 1986; 58:607-13.

28 Steinherz U, Graham T, Hurwitz R, et al. Guidelines forcardiac monitoring of children during and after anthracy-cline therapy: report of the Cardiology Committee of theChildren's Cancer Study Group. Pediatrics 1992; 89:942-9.

29 Praga C, Trave F, Petroccione A. Anthracycline-inducedcardiotoxicity and its relevance in cancer treatment. In:Nimmo WS, Tucker GT, eds. Clinical measurement in drugevaluation. London: Wolfe, 1991: 131-42.

30 Bonnadonna G, Gianni L, Santoro A, et al. Drugs ten yearslater: epirubicin. Ann Oncol 1993; 4: 359-69.

31 Lanchbury AP, Habboubi N. Epirubicin and doxorubicin: acomparison of their characteristics, therapeutic activityand toxicity. Cancer Treat Rev 1993; 19: 197-228.

32 Shenkenberg TD, Von HoffDD. Mitoxantrone: a new anti-cancer drug with significant clinical activity. Ann InternMed 1986; 105: 67-81.

33 Herman EH, Rahman A, Ferrans VJ, et al. Prevention ofchronic doxorubicin cardiotoxicity in beagles by liposomalencapsulation. Cancer Res 1983; 43: 5427-32.

34 Gabizon A, Peretz T, Sulkes A, et al. Systemic administra-tion of doxorubicin-containing liposomes in cancerpatients: a phase I study. Eur Cancer Clin Oncol 1989;25: 1795-803.

35 Treat J, Greenspan A, Forst D, et al. Antitumour activity ofliposome-encapsulated doxorubicin in advanced breastcancer: a phase II study. Y Natl Cancer Inst 1990; 82:1706-10.

36 Bielack SS, Erttmann R, Winkler K, Landbeck G.Doxorubicin: effect of different schedules on toxicity andantitumour efficacy. Eur Y Cancer Clin Oncol 1989; 25:873-82.

37 Weiss AJ, Manthel RW. Experience with the use of adria-mycin in combination with other anticancer agents usinga weekly schedule with particular reference to lack of car-diotoxicity. Cancer 1977; 40: 2046-52.

38 Torti FM, Bristow MR, Howes AE, et al. Reduced car-diotoxicity of doxorubicin delivered on a weekly schedule.Ann Intern Med 1983; 99: 745-9.

39 Legha SS, Benjamin RS, Mackay B, et al. Reduction ofdoxorubicin cardiotoxicity by prolonged continuousintravenous infusion. Ann Intern Med 1982; 96:133-9.

40 Casper ES, Gaynor JJ, Hajdu SI, et al. A prospectiverandomized trial of adjuvant chemotherapy with bolusversus continuous infusion of doxorubicin in patients withhigh-grade extremity soft tissue sarcoma and an analysisof prognostic factors. Cancer 1991; 68: 1221-9.

41 Speyer JL, Green MD, Bottino J, et al. A phase I/II study of6 hour and 24 hour intravenous infusion of doxorubicin.In: Muggia F, Young C, Carter S, eds. Anthracycline anti-biotics in chemotherapy. The Hague: Martinus Nijhoff,1982: 398-411.

42 Tannir N, Yap HY, Hortobagyi GH, et al. Sequential con-tinuous infusion with doxorubicin and vinblastine: aneffective chemotherapy combination for patients withadvanced breast cancer previously treated with cyclophos-phamide, methotrexate, 5-FU, vincristine. CancerTreatment Reports 1984; 68: 1039-41.

43 Speyer JL, Green MD, Dubin N, et al. Prospective evalua-tion of cardiotoxicity during a 6-hour doxorubicininfusion regimen in women with adenocarcinoma of thebreast. Am JMed 1985; 78: 555-63.

44 Shapira J, Gotfried M, Lishner M, Ravid M. Reduced car-diotoxicity of doxorubicin by a 6-hour infusion regimen: aprospective randomized evaluation. Cancer 1990; 65:870-3.

45 Borden EC, Amato DA, Rosenbaum DH, et al.Randomized comparison of three Adriamycin regimensfor metastatic soft tissue sarcomas. 7 Clin Oncol 1987; 5:840-50.

46 Zalupski M, Metch B, Balcerzak S. Phase III comparisonof doxorubicin and dacarbazine given by bolus versusinfusion in patients with soft tissue sarcomas: a SouthwestOncology Group study. Y Nail Cancer Inst 1991; 83:926-32.

47 Hellman K, Newton KA, Whitmore DN, et al. Preliminaryclinical assessment of ICRF-159 in acute leukaemia andlymphosarcoma. BMJ 1969; i: 822-4.

48 Herman EH, Mhatre RM, Lee IP, Waravdekar VS.Prevention of the cardiotoxic effects of Adriamycin anddaunomycin in the isolated dog heart. Proc Soc Exp BiolMed 1972; 140: 234-49.

49 Rajagopalan S, Politi PM, Sinha BK, Myers CE.Adriamycin-induced free radical formation in the per-fused rat heart: implications for cardiotoxicity. Cancer Res1988; 48: 4766-9.

50 Sehested M, Jensen PB, Sorensen BS, et al. Antagonisticeffect of the cardioprotector (+)-1,2-bis (3,5-dioxopiper-azinyl-l-YL) propane (ICRF-187) on DNA breaks andcytotoxicity induced by the topoisomerase II directeddrugs daunorubicin and etoposide (VP- 16). BiochemPharmacol 1993; 46: 389-93.

51 Monti E, Sinha BK. Potentiation of doxorubicincytotoxicity by (+)-1,2-bis (3,5-dioxopiperazinyl-1-yl)propane (ICRF-187) in human leukamic HL-60 cells.Cancer Communications 1990; 2: 145-9.

52 Holcenberg JS, Tutsch KD, Earhart RH, et al. Phase I studyof ICRF-187 in pediatric cancer patients and comparisonof its pharmacokinetics in children and adults. CancerTreatment Reports 1986; 70: 703-9.

53 Speyer JL, Green MD, Kramer E, et al. Protective effect ofthe bispiperazinedione ICRF-187 against doxorubicin-induced cardiac toxicity in women with advanced breastcancer. N Engl_'Med 1988; 319: 745-52.

54 Speyer JL, Green MD, Zeleniuch-Jacquotte A, et al. ICRF-187 permits longer treatment with doxorubicin in womenwith breast cancer. Clin Oncol 1992; 10: 1 7-27.

55 Bu'lock FA, Gabriel HM, Oakhill A, et al. Cardioprotectionby ICRF-187 against high dose anthracycline toxicity inchildren with malignant disease. Br Heart 1993; 70:185-8.

462



j.com/

Arch D


ovember 1994. D

ownloaded from

http://adc.bmj.com/

anthracyclines: cardiotoxicity andits preventionand irreversible cardiotoxicity, the full extent...

Documents