1 592
`
`Trastuzumab-Associated Cardiotoxicity
`
`Deborah L. Keefe, M.D., M.P.H.
`
`Cardiology Service, Memorial Sloan-Kettering Can-
`cer Center, New York, New York.
`
`Dr. Keefe served as a paid consultant to Genetech
`during the Clinical development 0f trastuzumab
`and as chair “the Cardiactvent “New 9mm?“
`tee. She has no current affiliation With or tmancral
`interest
`in Genetech. Dr. Keete currently is an
`employee of Berlex Laboratories, which manufac-
`turers no relevant products.
`
`Address for rep’ims; Deborah L' Keefe’ MD" MPH"
`PD. Box 500, New Rochelle, NY 10804-0500; Fax:
`914-636-0711; E—mail: dkclinpharm@aol.com
`
`Received March 25, 2002; accepted May 1, 2002.
`
`© 2002 American Cancer Society
`
`Trastuzumab is a monoclonal antibody used for the treatment of metastatic breast
`carcinoma in women whose tumors overexpress the HERZ protein. Cardiotoxicity
`has been reported to occur with trastuzumab when administered alone and in
`combination with antineoplastic agents, particularly anthracyclines. The risk of
`cardiotoxicity with trastuzumab has been reported to be 4% with monotherapy and
`27% when administered in combination with an anthracycline and cyclophos-
`phamide, but to the author’s knowledge severe outcomes, such as death or per-
`manent disability, are uncommon. The majority of reported cardiac effects are
`mild to moderate, nonspecific, and medically manageable. Signs and symptoms
`are similar to those observed in patients who develop anthracycline-induced
`cardiomyopathy and include tachycardia, palpitations, and exertional dyspnea,
`which may progress to congestive heart failure. The pathogenesis and histologic
`changes responsible for trastuzumab-associated cardiotoxicity currently are under
`investigation. Unlike anthracycline-induced toxicity, trastuzumab-associated tox~
`icity usually responds to standard treatment or the discontinuation of trastu-
`zumab, and there is no evidence that the toxicity is dose related. Current methods
`for the early detection of cardiotoxicity in trastuzumab-treated patients are similar
`to those used in anthracycline-treated patients. Cardiac function is established at
`baseline and monitored regularly during treatment by physical examination and
`measurement of left ventricular ejection fraction. The majority of patients improve
`with proper treatment, and some are able to continue to receive trastuzumab.
`Cancer 2002;95:1592—600. © 2002 American Cancer Society.
`DOI 10.1002/cncr.10854
`
`KEYWORDS: trastuzumab, cardiotoxicity, anthracyclines, metastatic breast carci-
`noma, HERZ.
`
`he human epidermal growth factor receptor-2 (HERZ) is an im—
`portant mediator of cell growth, differentiation, and survival.1'2
`Approximately 20—25% of breast tumors overexpress HERZF’“7 Over-
`expression is nearly always the result of amplification of the HERZ
`gene (also known as erb B-2 or HERZ/neu), and amplification or
`overexpression in breast carcinoma is associated with poor clinical
`outcome.
`
`Trastuzumab (Herceptin®, Genertech, Inc, South San Francisco,
`is a monoclonal antibody that selectively binds to HERZ. A1-
`CA)
`though the exact mechanism of action of trastuzumab has not been
`.
`.
`.
`.
`.
`.
`defimtlvely 1dent1fied, several proposed mechanlsms include antago-
`nism 0f the growth-stimulating properties 0f HERZ by down-regulat-
`ing receptor expression, activation of complement-mediated tumor
`cell lysis, and augmentation of chemotherapy-induced cytotoxicity.8
`Approved for use in the US. in 1998, trastuzumab is indicated for the
`treatment of metastatic breast carcmoma 1n women whose tumors
`overexpress the HERZ protein. It is approved for use both as mono-
`therapy in women who have previously received one or more che-
`
`Genentech 21 1 7
`
`Hospira v. Genentech
`|PR2017-OO737
`
`Genentech 2117
`Hospira v. Genentech
`IPR2017-00737
`
`

`

`motherapy regimens for their disease and for admin-
`istration in combination with paclitaxel in women
`who have not previously received chemotherapy in
`the metastatic setting. When administered in combi-
`nation with chemotherapy, there are statistically sig-
`nificant improvements in overall response rates, me-
`dian time to tumor progression, and median overall
`survival compared with chemotherapy alone.9 For ex-
`ample, the median survival was 25.1 months in the
`trastuzumab plus chemotherapy group versus 20.3
`months in the group receiving chemotherapy‘alone (P
`= 0.046). Trastuzumab also produces durable, objec-
`tive responses when administered alone.”11
`Cardiotoxicity as an adverse effect of trastuzumab
`therapy did not become evident until Phase III clinical
`trials.12 Because cardiac problems were not antici-
`pated, cardiac monitoring was not mandated in many
`studies, and even among studies in which cardiac
`monitoring was performed, measurements were not
`consistent from one study to another. This makes it
`difficult to compare results between studies and to
`identify predisposing risk factors. It also is difficult to
`compare trastuzumab-associated cardiotoxicity with
`that related to other chemotherapeutic agents, be-
`cause the nature of cardiac dysfunction associated
`with all these agents has not yet been characterized
`fully. For example, doxorubicin has been documented
`to cause a dose-related cardiomyopathy, whereas tras-
`tuzumab has been shown to be associated with
`
`asymptomatic drops in left ventricular ejection frac-
`tion (LVEF), which do not appear to be dose related, as
`well as clinical congestive heart failure.
`Although the cardiotoxicity with which trastu-
`zumab is associated has become better characterized
`
`as experience with this agent has increased and risk
`factors and appropriate monitoring and treatment
`procedures are beginning to be identified, there still is
`much work to be done to characterize this event ac—
`
`curately. This article reviews the epidemiology, signs
`and symptoms, and proposed pathogenesis of trastu-
`zumab-associated cardiotoxicity and presents guide-
`lines for minimizing, detecting, and managing this
`condition.
`
`Trastuzumab-Associated Cardiotoxicity/Keefe
`
`1593
`
`cardiomyopathy during the maintenance phase of this
`study. However, this patient had numerous risk fac-
`tors for cardiomyopathy, including a previous cumu-
`lative anthracycline dose of 420 mg/mz, prior chest
`wall irradiation, hypertension, and diabetes.
`
`Cardiac review and evaluation committee analysis
`As data from the Phase III trials accumulated, reports
`of trastuzumab-related cardiotoxicity began to sur-
`face. These reports prompted a retrospective analysis
`of cardiac events in seven Phase II and III trastuzumab
`
`clinical trials by an independent, blinded Cardiac Re-
`view and Evaluation Committee (CREC).l4 The CREC
`defined cardiac dysfunction as 1) a decline in LVEF of
`5% to < 55% with accompanying signs or symptoms of
`congestive heart failure or a decline in LVEF of at least
`10% to < 55% without accompanying signs or symp-
`toms, 2) symptoms of congestive heart failure, 3) as—
`sociated signs of congestive heart failure including 83
`gallop and/ or tachycardia, or 4) cardiomyopathy char-
`acterized by a decline in LVEF that either was global or
`more severe in the septum.
`One of the trials analyzed by the CREC was a
`randomized, multicenter study of trastuzumab mono-
`therapy as first-line therapy for metastatic breast car—
`cinoma.15 Approximately 58% of the patients in this
`study had received prior anthracycline therapy. The
`CREC identified 3 of 114 women as having cardiac
`events that possibly were related to trastuzumab. Two
`of the three patients had histories of cardiac disease.
`One patient, who had a history of hypertension and
`coronary and cerebrovascular disease, was found to
`have bilateral atrial and right ventricular enlargement
`after 28 weeks of trastuzumab treatment. Intermittent
`
`pleural effusions subsequently developed. The second
`patient had a baseline LVEF of 35— 40% and a history of
`prior anthracycline therapy and myocardial infarction.
`Her LVEF was 31% after 1 year of trastuzumab therapy
`while cardiac drugs were withheld. The cardiac event
`in the third patient, reduction in LVEF after relief of
`pericardial tamponade, was attributed to breast carci-
`noma. None of these three patients required further
`intervention for cardiac events after the discontinua-
`tion of trastuzumab.
`
`Epidemiology
`Early trials
`Early clinical trials provided little or no indication of
`an increased risk of cardiotoxicity in patients treated
`with trastuzumab. For example, in a Phase II study in
`women with refractory metastatic breast carcinoma,
`Pegram et al.13 reported no National Cancer Institute
`Common Toxicity Criteria (CTC) Grade 3 or 4 cardiac
`toxicity among patients treated with trastuzumab plus
`cisplatin for 8 weeks. One of 19 patients (5%) reported
`
`The CREC also analyzed cardiac events in a mul-
`tinational study of 222 women who received trastu-
`zumab monotherapy as second-line or third-line che-
`motherapy in the setting of metastatic disease.10 All
`patients had received extensive prior therapy, includ-
`ing anthracyclines (94% of women), taxanes (67%),
`radiotherapy (71%), and hormonal therapy (57%). Car-
`diac dysfunction was identified in 10 patients (4.7%),
`including 3 whose cardiac events occurred after tras-
`tuzumab therapy was discontinued. Of these 10 pa-
`
`

`

`1594
`
`CANCER October 1, 2002 / Volume 95 / Number 7
`
`TABLE 1
`Incidence of Cardiac Dysfunction in a Randomized Comparative Clinical Study of Trastuzumab plus Chemotherapy versus Chemotherapy Alone
`_——___—.———_—-——-——
`and in Pooled Studies of Trastuzumab Monotherapy
`
`incidence ('70)
`
`
`Trastuzumab
`
`
`Monotherapy
`[n = 338)
`
`Plus anthracycline
`plus cyclophospharnide
`(n = 143)
`
`Plus paclitaxel
`(n = 91)
`
`Anthracycline plus
`cyclophospharnide
`monotherapy.
`(n = 135)
`
`Paclitaxel
`monotherapy
`[n = 95)
`
`1.0
`8.0
`13.0
`27.0
`4.0
`Cardiac dysfunction
`1.0
`3.0
`2.0
`16.0
`3.0
`Grade 3 or 4 (initial)
`0
`0.7
`0
`6.0
`1.5
`Grade 3 or 4 (posttreatrnent)
`
`
`0.9Death from cardiac dysfunction 0 0.7 0 0.7
`
`
`
`
`Data adapted from Slamon DI. Leyland-Iones B. Shak S, et al. Use of chemotherapy plus a monoclonal antibody against HERZ for metastatic breast cancer that overexpresses HER2. N Eng! 1Med. 2001;344:7213-7512
`and Baselga I. Safety profile of Herceptin as a single agent and in combination with chemotherapy [abstract 1299]. Eur] Cancer. 1999;35:3324.fl
`
`tients, 9 had received prior anthracycline therapy and
`had risk factors for anthracycline-induced cardiac tox-
`icity (total doxorubicin dose > 400 mg/mz, radiother-
`apy to the left chest, age > 70 years, and a history of
`hypertension). The patient who had not had prior
`exposure to anthracyclines had significant cardiac dis-
`ease at the time of study entry. The majority of cardiac
`events were clinically significant but responded to
`routine therapy. There was one death caused by pre-
`sumed ventricular arrhythmia in a patient with previ—
`ous exposure to anthracyclines and significant cardiac
`dysfunction at the time of study entry. Of the eight
`patients who continued treatment with trastuzumab
`despite cardiac dysfunction, six remained stable. Two
`patients with a decreased ejection fraction were
`asymptomatic.
`Finally, the CREC analyzed the results of a ran-
`domized, multinational, Phase III trial comparing tras-
`tuzumab plus chemotherapy with chemotherapy
`alone in 469 patients with metastatic breast carcino-
`ma.9 Cardiotoxicity (both symptomatic and asymp-
`tomatic) was reported in 27% of patients who received
`trastuzumab plus anthracycline and cyclophospha-
`mide versus 8% who received anthracycline and cy—
`clophosphamide without trastuzumab (Table 1). Car-
`diotoxicity was reported in 13% of patients treated
`with paclitaxel plus trastuzumab, and the risks for
`patients receiving trastuzumab (3—7%) or paclitaxel
`(1%) alone were substantially lower. The continued
`use of trastuzumab did not cause further cardiac de-
`
`terioration in the majority of patients; in nearly all
`cases, cardiac function improved after treatment with
`standard therapy.
`It also is interesting to note that 64% of patients
`treated with trastuzumab plus the anthracycline com-
`bination presented with significant functional impair-
`
`ment (New York Heart Association classification class
`III or IV toxicity), whereas only 20% of patients treated
`with trastuzumab plus paclitaxel presented with im-
`pairment (all class III toxicities). The majority of pa—
`tients (79%) improved with treatment. Cardiac-related
`deaths in patients treated with trastuzumab mono-
`therapy or combination therapy were uncommon
`(1%).
`
`Further analyses
`A pooled—data analysis of 338 patients treated with
`trastuzumab monotherapy in 3 clinical trials revealed
`a 4% risk of cardiac dysfunction, including a 3% risk of
`CTC Grade 3 or 4 dysfunction, that resolved in 50% of
`patients after treatment (Table 1).16 It also should be
`noted that patients treated with chemotherapy alone
`(an anthracycline plus cyclophosphamide) had an 8%
`risk of cardiac dysfunction, including a 3% risk of
`Grade 3 or 4 dysfunction, and patients treated with
`paclitaxel had a 1% risk of cardiotoxicity. The risk of
`cardiac dysfunction also appears to be greater among
`patients receiving trastuzumab monotherapy as sec-
`ond-line therapy (7% overall; 5% Grade 3 or 4 dysfunc-
`tion) compared with first-line therapy (0.8%; no Grade
`3 or 4 dysfunction), most likely because of the differ-
`ence in prior exposure to anthracyclines in the 2 pop-
`ulations (94% and 51%, respectively).17
`
`Recent trials
`
`trials of trastu-
`Because of these findings, clinical
`zumab adopted more stringent cardiac monitoring,
`and the majority of studies now exclude patients with
`either a significant cardiac history and/ or high doses
`of previous cumulative anthracycline exposure. In a
`recent Phase II trial of 95 women treated with trastu-
`
`zumab plus paclitaxel, Seidman et al.18 reported that
`
`

`

`the median LVEF values remained relatively stable
`throughout the first year of therapy. Although 7 pa-
`tients developed asymptomatic decreases in LVEF of
`> 20% from baseline, therapy with trastuzumab plus
`paclitaxel was continued in all these patients without
`dosage modification. The LVEF improved or returned
`to normal in 5 patients without treatment, such that
`only 2 individuals required medical management of
`their cardiac condition. Three other patients devel-
`oped serious cardiac complications. One patient who
`previously had been treated with a cumulative doxo-
`rubicin dose of 615 mg/m2 developed transient con-
`gestive heart failure. A second patient who previously
`had been treated with a cumulative doxorubicin dose
`of 240 mg/m2 experienced an acute myocardial infarc-
`tion and developed ischemic cardiomyopathy. A third
`patient had a massive myocardial infarction shortly
`after being intubated for respiratory failure secondary
`to pneumonia.
`
`In a study in which trastuzumab plus vinorelbine
`was administered as second-line or third-line therapy
`to 40 women with metastatic breast carcinoma, no
`CTC Grade 3 or 4 cardiac toxicity or symptomatic
`congestive heart failure was reported.19 Three patients
`developed Grade 2 cardiac toxicity (defined as a de-
`cline in LVEF of > 20% or to < 50%) and a fourth
`patient experienced a Grade 1 cardiac toxicity (defined
`as a decline in LVEF of 10—20%). All patients with
`Grade 2 cardiotoxicity had prior doxorubicin exposure
`of > 240 mg/m2 and a baseline LVEF of 50% and 59%
`(P = 0.002). Approximately 58% of the patients had
`received prior anthracycline therapy.
`Preliminary data also are available from two Phase
`II studies of docetaxel plus trastuzumab therapy for
`metastatic breast carcinoma.20 In one study in which
`14 patients had been enrolled at the time of publica-
`tion, none of the patients had experienced symptom-
`atic declines in LVEF, and no CTC Grade 3 or 4 cardiac
`
`toxicity was observed. Multigated radionuclide an-
`giography (MUGA) scans, performed at baseline and
`every 8 weeks, revealed 1 case of an asymptomatic
`reduction in LVEF (from 68% to 52%). However, the
`ejection fraction returned to baseline without discon-
`tinuation of therapy or medical intervention. In a sec-
`ond study of docetaxel plus trastuzumab in which 21
`patients had been treated by the time of publication,
`there were no reports of cardiotoxicity. No patients
`experienced symptoms of congestive heart failure or a
`decline in LVEF of 2 20% to a value of < 50%. It also
`
`should be noted that 9 of these 21 patients had re-
`ceived prior anthracycline therapy, and eligibility cri-
`teria restricted the maximum prior lifetime dose of
`doxorubicin to s 250 mg/mz.
`
`Trastuzumab-Associated Cardiotoxicity/Keefe
`
`1595
`
`TABLE 2
`Risk Factors for Trastuzumab Cardiotoxicity
`
`Documenteda
`
`0 Treatment with trastuzumab plus chemotherapy (highest risk
`combination is trastuzumab plus concurrent anthracyclines)
`0 Age > 60 yrs
`0 Previous cumulative anthracycline dose a 400 mg/m2
`0 Prior chest wall irradiation
`0 Preexisting cardiac dysfunction
`
`Suspected
`
`“ Data adapted from Sparano IA Cardiac toxicity of trastuzumab (Herceptin): implications for the
`design of adjuvant trials. Semin Oncol. 2001;28:20—27.
`
`Risk Factors
`
`Risk factors specific for trastuzumab-associated car-
`diotoxicity have not been clearly established, although
`the majority of patients had other risk factors for car-
`diac dysfunction. Analyses of the potential risk factors,
`including age, weight, history of hypertension, cumu-
`lative doxorubicin dose, HER2 expression level, and
`treatment, have revealed that only age and concurrent
`doxorubicin therapy are significantly associated with
`an increased risk of cardiac dysfunctionf“l7 Data are
`inadequate to confirm other risk factors, such as the
`cumulative dose of prior anthracyclines, preexisting
`heart disease, or radiation to the chest, although in-
`vestigators suspect such associations may exist (Table
`2). Because the majority of patients receiving trastu-
`zumab monotherapy previously were exposed to an
`anthracycline, it is uncertain whether trastuzumab is
`associated with cardiotoxicity in patients who were
`never exposed to anthracyclines.
`
`Characterization
`
`Few detailed, prospective assessments of trastu-
`zumab-associated cardiotoxicity have been con-
`ducted, which makes the characterization of this event
`
`difficult. Most reported trastuzumab—related cardiac
`. events appear to be asymptomatic decreases in LVEF,
`the significance of which is unclear. When signs and
`symptoms are reported, they are mild to moderate in
`severity and include tachycardia, palpitations, non-
`specific dyspnea, and occasional chest pain, which
`may progress to congestive heart failure. These obser—
`vations should be interpreted with caution. Analysis of
`trastuzumab-related events is limited by the retro-
`spective nature of their collection. Furthermore, to be
`able to understand the true significance of trastu-
`zumab-related events, it is important to determine the
`true incidence of LVEF drops in patients receiving
`chemotherapy alone.
`Tachycardia is an early sign of anthracycline-in-
`duced cardiotoxicity, which may be present in asymp-
`tomatic patients without altered systolic left ventricu-
`
`

`

`1596
`
`CANCER October 1, 2002 I Volume 95 I Number 7
`
`lar function.21 Although unproved, tachycardia also
`may be a sensitive tool with which to identify the first
`signs of trastuzumab-associated cardiotoxicity. How-
`ever, although the signs and symptoms of anthracy-
`cline-induced and trastuzumab-associated cardiotox-
`
`icity may be similar,22 closer examination reveals
`important differences, including dose-relatedness and
`the degree of reversibility. Whereas the risk of anthra-
`cycline-induced cardiotoxicity increases as the cumu-
`lative anthracycline dose increases,22 there is no evi-
`dence that the risk of cardiotoxicity is dose related in
`patients treated with trastuzumab. Although many pa-
`tients with anthracycline-induced cardiotoxicity re-
`cover from acute congestive heart failure, their cardiac
`reserve often is limited, and they require careful long-
`term management.23 In contrast,
`trastuzumab ap-
`pears to be associated less frequently with severe tox-
`icity, and it is at least partially reversible in many
`cases. There usually is improvement in functional
`class and ejection fractions after standard treatment
`or trastuzumab discontinuation,9’24 although a small
`number of patients may be left with severe heart fail-
`ure. The irreversibility of doxorubicin-induced dam-
`age to the heart is believed to result from the genera-
`tion of free radicals that cause cellular injury, a
`concurrent decrease in endogenous antioxidants that
`results in oxidative stress, and subcellular changes
`that include loss of myofibrils and vacuolization of
`myocardial cells.22
`
`Pathogenesis
`The pathogenesis of trastuzumab-associated cardio-
`toxicity currently is under investigation. HER2 is
`known to play an essential role in embryonic cardio-
`genesis and cardiac hypertrophy,25 and a recent case
`report has demonstrated the presence of HER2 in
`human cardiac tissue.24 Thus it is possible that tras-
`tuzumab has a direct cardiotoxic effect that is medi-
`
`ated by the cardiac HER2 receptor. In fact, Fuchs et
`al.26 theorized that trastuzumab-associated cardiotox-
`
`icity is a consequence of interaction with myocardial
`cells that pathologically overexpress HER2. However,
`analysis of HER2 status in heart biopsy tissue from
`patients with decreased left ventricular function but
`without previous exposure to trastuzumab did not
`reveal overexpression or gene amplification.
`Other hypotheses have been proposed for the
`mechanism of trastuzumab-associated cardiotoxic-
`
`ity.27 The first is that, similar to the anthracyclines,
`trastuzumab may have inherent cardiotoxicity. The
`second is that trastuzumab amplifies or has an addi-
`tive cardiotoxic effect when administered after or con-
`
`currently with anthracyclines or other cardiotoxic
`agents. To date, nearly all patients treated with tras-
`
`tuzumab have been exposed to anthracyclines, either
`before or concurrently with trastuzumab administra-
`tion. This makes the assessment of inherent trastu-
`
`zumab toxicity impossible using available data but
`supports the hypothesis of an amplified or additive
`cardiotoxicity.27 An extension of the amplified or ad-
`ditive cardiotoxicity hypothesis would include in-
`creased vulnerability to any stress, which would cause
`activation of cellular injury pathways. Basic scientific
`evidence supporting this theory includes the work of
`Chien.28'29 Ventricle-restricted erb B-2 knockout mice
`
`develop dilated cardiomyopathy with impaired con-
`tractility, relaxation, and response to B-agonists. Im-
`pairment of this pathway also impairs the ability of
`mice to withstand a stress such as aortic banding.
`However, these mice can be rescued by overexpres-
`sion of BcL-xL, which is a salvage mechanism in apo-
`ptosis. These and other data suggest that erb B-2 re-
`ceptor—dependent myocyte survival pathways may
`create a susceptibility for the onset of heart failure in
`response to the cardiotoxicity of anthracycline treat-
`ment.
`Preclinical animal and small clinical studies cur-
`
`rently are underway to determine the mechanism of
`action of trastuzumab-associated cardiotoxicity. By
`conducting these studies, more detailed information
`regarding the nature of this adverse event will be
`gained.
`
`Detection and Management
`Methods for the early detection of cardiotoxicity in
`trastuzumab-treated patients should be similar to
`those used for anthracyclines. The standard clinical
`approach should include the assessment of baseline
`cardiac performance before therapy, regular monitor-
`ing during treatment, and follow-up after the comple-
`tion of therapy. When physicians first began to diag-
`nose anthracycline-induced cardiotoxicity, they relied
`on physical examination and electrocardiography.22
`Although these surveillance methods can identify gal-
`lop rhythms, reversible arrhythmias (most commonly
`sinus tachycardia),
`jugular vein distention, ankle
`edema, and hepatomegaly, these symptoms are not
`specific to doxorubicin-induced heart failure and are
`common to many pathologic conditions;
`therefore
`physicians should not depend on them to establish a
`diagnosis. However, new-onset gallop rhythms and
`jugular vein distention may be a sign of impending
`congestive heart failure. In addition, electrocardiogra-
`physhould be used in conjunction with echocardiog-
`raphy and MUGA scans.
`MUGA scans, which can be used to measure ejec-
`tion fraction, are noninvasive, which makes them a
`favorable choice for routine clinical monitoring. How-
`
`

`

`TABLE 3
`
`Indications for a Formal Assessment of Cardiac Function by LVEF in
`Patients Receiving Trastuzumab
`
`I Heart rate increase in bpm from baseline
`Baseline heart rate < 80 bpm, increasing to > 90 bpm
`Baseline heart rate 2 80 bpm ’and s 100 bpm, increasing to > 100 bpm
`Baseline heart rate > 100 bpm, increasing to > 120 bpm
`0 Weight increase of 2 2 kg in 1 week
`0 Spontaneous report of new symptoms of dyspnea on exertion
`
`LVEF: left ventricular ejection fraction; bpm: beats per minute.
`
`ever, MUGA scans primarily identify drops in LVEF,
`the significance of which are unknown if the patient is
`asymptomatic. Echocardiography, which also is non-
`invasive, also may be of value in identifying treatment-
`induced and nontreatment-induced cardiotoxicity.
`This technique can identify both systolic and diastolic
`dysfunction and valvular and pericardial disease, ei-
`ther underlying or in relation to previous radiation or
`tumor. If alternate causes for symptoms other than
`systolic dysfunction are identified, unnecessary inter-
`ruptions in trastuzumab therapy may be avoided. En-
`domyocardial biopsies offer the greatest specificity
`and sensitivity for doxorubicin-induced cardiotoxic-
`ityzz; however, histologic changes have not been re-
`ported to be associated with trastuzumab, and the
`invasiveness and expense of the procedure limits its
`use to the research setting.
`
`Clinical Guidelines
`
`At Memorial Sloan-Kettering Cancer Center, guide-
`lines were established for monitoring cardiac function
`before and during treatment with trastuzumab. All
`patients in whom treatment with trastuzumab is an-
`ticipated have a baseline measurement of LVEF with
`either a MUGA scan or echocardiograph and a phys-
`ical examination. Only patients with a normal baseline
`ejection fraction and no signs or symptoms of conges-
`tive heart failure on physical examination are consid-
`ered for trastuzumab treatment. In addition, caution
`
`is exercised when treating patients who have a history
`of hypertension, documented coronary artery disease,
`or significant valvular disease.
`Once treatment is initiated, heart rate and body
`weight are monitored weekly; heart rate increases of
`10—20 beats per minute (bpm), body weight increase
`of 2 2 kg in 1 week, or a spontaneous (i.e., unsolicited)
`report of symptoms of congestive heart failure, partic-
`ularly new fatigue or impaired exercise tolerance, sug-
`gest the onset of congestive heart failure (Table 3). The
`occurrence of any of these events triggers a formal
`assessment of LVEF with the same method used at
`
`baseline (Table 4). Because of the variability inherent
`
`Trastuzumab-Associated Cardiotoxicity/Keefe
`
`1597
`
`in the assessment of ejection fraction, repeat studies
`by an alternate method may be conducted to confirm
`abnormalities. Patients who are found to have asymp-
`tomatic drops in LVEF are treated with caution.
`Patients who develop congestive heart failure are
`given standard treatment according to the guidelines
`of the American College of Cardiology/American
`Heart Association and the Heart Failure Society of
`America (Table 5).”31 In addition, Memorial Sloan—
`Kettering Cancer Center guidelines recommend the
`use of angiotensin-converting enzyme inhibitors and
`B-blockers (particularly in clinically compensated pa-
`tients), with or without digoxin and diuretics, to treat
`trastuzumab-associated congestive heart failure. Al-
`though congestive heart failure occurring in patients
`treated with trastuzumab may be moderate or severe,
`in my experience, the majority of patients who de—
`velop congestive heart failure improve with standard
`treatment, and some are able to continue to receive
`
`trastuzumab therapy. The decision to continue or dis—
`continue trastuzumab therapy depends on the pa-
`tient’s clinical examination and their LVEF, as detailed
`in Table 4. For example, patients with asymptomatic
`heart failure and an LVEF of approximately 35—55%
`would likely be treated for heart failure but may be
`allowed to continue treatment with trastuzumab for
`
`advanced metastatic disease. Patients with symptom-
`atic congestive heart failure whose ejection fraction
`falls by no > 10% or whose ejection fraction is > 50%
`also may continue to receive trastuzumab. If trastu-
`zumab has been discontinued because of cardiac dys-
`function and the symptoms subsequently resolve,
`therapy may be reinitiated after the risks and benefits
`of continued use of trastuzumab are discussed with
`
`the patient. If the patient had demonstrated clinical
`benefit from trastuzumab, the continuation of therapy
`may be warranted.
`
`Minimizing Trastuzumab-Associated Cardiotoxicity
`Establishing strategies to minimize trastuzumab-asso-
`ciated cardiotoxicity is difficult because the pathogen-
`esis of the toxicity is unknown. Because the greatest
`risk appears to be in patients who are exposed to both
`trastuzumab and an anthracycline, many physicians
`are trying to avoid exposing patients to these two
`agents concurrently. Although the benefits are unclear
`and need to be studied formally, two strategies have
`been proposed to minimize the potential for trastu-
`zumab-associated cardiotoxicity. The first strategy is
`to decrease the potential for anthracycline-induced
`toxicity. This may be accomplished through close
`monitoring and maintaining the cumulative dose
`within the “safe” range (i.e., < 450 mg/m2)32 or
`through the use of liposome-encapsulated doxorubi-
`
`

`

`1598
`
`CANCER October 1, 2002 / Volume 95 / Number 7
`
`TABLE 4
`
`Proposed Guidelines for the Management of Patients Treated with Trastuzumab, Based on Physical Status and LVEF
`
`Physical
`
`statusa
`LVEF
`Trastuzumab
`LVEF monitoring
`Management
`
`Action
`
`
`Asymptomatic
`
`l but normal
`, > 10 points but normal
`a 10—20 points and LVEF > 40%
`
`Continue
`Continue
`Continue
`
`l > 20 points to < 40% or LVEF < 30%
`
`Hold
`
`Repeat in 4 weeks
`Repeat in 4 weeks
`Repeat in 2 to 4 weeks
`e Improved: monitor
`I Not improved: stop trastuzumab
`Repeat in 2 weeks
`I Improved to > 45%: restart
`trastuzumab
`0 Not improved: stop trastuzumab
`
`Consider fi-blockers
`Treat for CHF
`
`Treat for CHF
`
`Symptomatic
`
`,l < 10 points
`l > 10 points and LVEF > 50%
`
`Continue
`Continue
`
`Search for noncardiac pathology (e.g., anemia)
`Treat for CHF
`
`Repeat in 2—4 weeks
`0 Stable or improved: continue
`trastuzumab
`0 Worsened: stop trastuzumab
`t < 30 points Treat for CHF Stop
`
`
`
`
`
`LVEF: left ventricular ejection fraction; CHF: congestive bean failure.
`a Heart rate and body weight should be monitored weekly. Asymptomatic is defined as changes in heart rate and or weight (as per Table 3) but without symptoms of dyspnea on exertion. Symptomatic is defined
`as a new. spontaneous (i.e.. unsolicited) report of symptoms of dyspnea on exertion, pulmonary vascular congestion, or edema.
`
`
`TABLE 5
`
`American College of Cardiology/American Heart Association Guidelines for the Treatment of CHF, Updated with the Guidelines From the HFSA
`
`
`Diagnosis
`Medication
`Details
`
`Systolic dysfunction
`
`Angiotensin-converting enzyme inhibitors
`
`In symptomatic patients and in those in functional class I who have significantly
`reduced left ventricular function, use indefinitely unless contraindicated or not
`tolerated
`
`Diuretics
`
`Digoxin
`
`fi-blockers
`
`Aldosterone antagonists
`
`In symptomatic patients, use in conjunction with an angiotensin-converting
`enzyme inhibitor. and generally indefinitely, even when rendered free of
`edema
`Effective in many patients with moderate-to-severe heart failure, but its influence
`on mortality is uncertain; recommended for symptomatic patients
`Should be administered routinely to clinically stable patients with left ventricular
`systolic dysfunction, initiated at low dose and up-titrated slowly
`Low-dose spironolactone (12.5—25 mg/day) should be considered for patients
`receiving standard therapy who have severe or recurrent heart failure
`For symptomatic patients, drug of choice
`Diuretics
`For symptomatic patients, drug of choice
`Nitrates
`May be of benefit
`Calcium channel blockers, B-blockers, and angiotensin-
`converting enzyme inhibitors
`
`Diastolic dysfunction“
`
`CHF: congestive heart failure; HFSA: Heart Failure Society of America.
`a The goal of therapy is to control symptoms by reducing ventricular filling pressure without reducing cardiac output. Agents with positive inotropic actions are not indicated if systolic fun