First Published Online January 15, 2009 The Oncologist, Vol. 14, No. 1, 1-11, January 2009; doi:10.1634/theoncologist.2008-0137 © 2009 AlphaMed Press
Minimizing Cardiotoxicity While Optimizing Treatment Efficacy with Trastuzumab: Review and Expert RecommendationsaDepartment of Medical Oncology, San Carlos University Hospital, Madrid, Spain; bMD Anderson Cancer Center, Houston, Texas, USA; cDepartment of Oncology, Vírgen de la Victoria University Hospital, Málaga, Spain; dDepartment of Cardiovascular Diseases, Mayo Clinic, Scottsdale, Arizona, USA; eDepartment of Cardiology, San Carlos University Hospital, Madrid, Spain Key Words. Breast cancer • Cardiotoxicity • Heart failure • Left ventricular ejection fraction • Trastuzumab Correspondence: Miguel Martín, M.D., Ph.D., Department of Medical Oncology, Hospital Clínico San Carlos, Profesor Martín Lago s/n, 28040 Madrid, Spain. Telephone: +34-91-3303546; Fax: +34-91-3303544; e-mail: mmartin{at}geicam.org Received June 18, 2008; accepted for publication December 1, 2008; first published online in THE ONCOLOGIST Express on January 15, 2009.
Disclosures
This article is available for continuing medical education credit at CME.TheOncologist.com.
Numerous clinical studies have demonstrated the therapeutic benefit of trastuzumab in women with breast cancer. However, a small but not insignificant proportion of patients have experienced trastuzumab-associated cardiotoxicity during these trials. This phenomenon is generally characterized by an asymptomatic reduction in left ventricular ejection fraction (LVEF) or, less often, congestive heart failure (CHF). Concomitant anthracycline therapy significantly increases the risk for cardiotoxicity during trastuzumab treatment, and such regimens are therefore not recommended. The cardiac dysfunction associated with trastuzumab is most often reversible upon discontinuation of treatment and initiation of standard medical therapy for CHF. Prior to treatment initiation, a risk–benefit analysis should be performed for each individual patient, including a thorough assessment of potential risk factors and cardiac function. Cardiac monitoring should be continued throughout trastuzumab therapy and the follow-up period, because early recognition of trastuzumab-associated cardiac dysfunction can allow effective medical intervention. Following the occurrence of asymptomatic LVEF reduction or CHF and appropriate medical intervention, reintroduction of trastuzumab may be considered in patients following resolution of normal cardiac function, or in those for whom the benefit of antitumor therapy outweighs the risk for CHF.
Trastuzumab (Herceptin®; F. Hoffmann-La Roche, Basel, Switzerland) is a humanized monoclonal antibody that selectively inhibits the growth and proliferation of breast cancer cells overexpressing the type-1 tyrosine kinase membrane receptor human epidermal growth factor receptor (HER)-2/neu [1–6]. Evidence for the efficacy of trastuzumab was initially provided in phase II clinical trials involving patients with breast cancer overexpressing HER-2/neu who were refractory to chemotherapy [7–9] and a pivotal study involving 479 women with metastatic breast cancer (MBC) overexpressing HER-2/neu [10]. Chemotherapy plus first-line trastuzumab resulted in a significantly higher response rate than with chemotherapy alone, as well as longer durations of response and survival. This demonstration of trastuzumab efficacy in treating MBC led to the initiation of four international clinical studies in the postsurgery adjuvant setting, involving a total of nearly 12,000 women with HER-2/neu overexpression. In those trials, the experimental arm included trastuzumab concomitantly administered with chemotherapy or sequentially upon chemotherapy completion. Results indicated substantially better recurrence rates, disease-free survival times, and overall survival times with trastuzumab treatment [11–15]. Different chemotherapeutic regimens plus trastuzumab have also been evaluated in women with HER-2-overexpressing breast cancer in the neoadjuvant setting, with significantly greater pathological complete response rates observed [16–19]. Trastuzumab is generally well tolerated. However, in a number of these clinical studies, cardiac dysfunction was observed, secondary to trastuzumab therapy. This paper aims to identify the incidence of this phenomenon, together with potential risk factors and strategies for minimizing its occurrence in clinical practice. Management of trastuzumab-associated cardiotoxicity is also discussed.
A summary of the incidence of cardiotoxicity and the definitions used in the key clinical trials evaluating trastuzumab is presented in Table 1
MBC Trastuzumab-associated cardiotoxicity was first described in a small number of patients with MBC who participated in two phase II studies. The reported incidence of cardiac dysfunction was 2% of patients receiving trastuzumab alone and 31% of patients treated with paclitaxel plus trastuzumab [9, 20]. In the pivotal phase III trials, trastuzumab-associated cardiotoxicity was an unexpected finding; prospective cardiac monitoring was therefore not conducted at that time. Cardiotoxicity occurred more frequently in patients who received combined chemotherapy and trastuzumab treatment, particularly in those patients treated with anthracyclines (27%, versus 13% in patients treated with paclitaxel plus trastuzumab) [10]. In the chemotherapy arms, the incidence of cardiotoxicity was 8% in patients receiving anthracyclines, and it was 1% in those treated with paclitaxel. Subsequently, these pivotal phase II and III studies were reviewed by a cardiac review and evaluation committee (CREC) [21]. The CREC defined trastuzumab-associated cardiotoxicity as a complication consisting of either asymptomatic decreased left ventricular ejection fraction (LVEF) or the presence of symptoms consistent with grade III or IV congestive heart failure (CHF) (Table 2). Using the New York Heart Association (NYHA) criteria [22], toxicity was identified in nearly 9% of patients, and, although it was observed that trastuzumab by itself was able to induce cardiac dysfunction (2%), the incidence of this complication was substantially greater when trastuzumab was administered concomitantly with anthracyclines (16%). Most of the patients suffering from trastuzumab-induced cardiotoxicity were asymptomatic, and 79% responded to medical treatment. This finding led to a general consensus that the concomitant use of trastuzumab and anthracyclines should be avoided and that cardiac monitoring should be included in all subsequent clinical trials involving trastuzumab.
In a study including cardiac monitoring that assessed weekly paclitaxel plus trastuzumab, 7% of patients developed asymptomatic decreased LVEF, but only 3% presented with symptoms suggestive of CHF [23]. The M77001 trial assessed the combination of trastuzumab and docetaxel weekly or every 3 weeks and included monitoring of LVEF every three treatment cycles. Only 1% of the patients receiving combination therapy developed CHF symptoms, whereas 17% experienced asymptomatic reductions in LVEF 15%, compared with 8% of patients receiving docetaxel monotherapy [24]. In a single-center study including cardiac monitoring, the combination of vinorelbine and trastuzumab was associated with asymptomatic decreased LVEF in three of 40 patients, with none developing CHF [25]. A separate phase II trial evaluating combination therapy with trastuzumab and vinorelbine reported that two patients (4%) experienced a >20% reduction from baseline in LVEF and two (4%) experienced a >10% reduction from baseline in LVEF [26]. The results from clinical trials evaluating the combination of trastuzumab and gemcitabine or capecitabine in patients who had previously received anthracyclines and taxanes indicate no increased risk for CHF with these regimens [27–29]. The results from an interim safety analysis of preliminary data from an ongoing randomized study of docetaxel plus trastuzumab with or without capecitabine in patients with locally advanced breast cancer or MBC reported a CHF incidence of 2% [30]. Combinations of trastuzumab with other anthracyclines less cardiotoxic than doxorubicin, such as epirubicin or liposomal doxorubicin, have also been investigated. Preliminary analyses during stage I of the Herceptin®, Cyclophosphamide, and Epirubicin study showed that there was no dose-limiting toxicity. Of the 50 patients assigned to trastuzumab, only three experienced asymptomatic decreased LVEF, and one had an episode of uncomplicated tachyarrhythmia [31]. The combination of liposomal doxorubicin with paclitaxel and trastuzumab exhibits potent antitumor activity with no cases of cardiotoxicity observed [32].
Early Breast Cancer: Adjuvant Therapy In the Herceptin® Adjuvant trial, patients were randomly assigned to receive adjuvant trastuzumab for 1 year, adjuvant trastuzumab for 2 years, or observation, upon chemotherapy completion [12]. Patients who had received a cumulative doxorubicin dose >360 mg/m2 or a cumulative epirubicin dose >720 mg/m2 or who exhibited marginal decreased LVEF were excluded. LVEF was assessed at baseline and then regularly until 60 months after randomization. Decreased LVEF was seen in 7.1% of patients treated with trastuzumab, compared with 2.2% of patients in the control arm. Only 1.7% of patients experienced symptomatic CHF. After 2 years of follow-up, 0.6% of patients in the trastuzumab group had developed severe CHF (versus 0% in the control group; p < .0001). In addition, symptomatic CHF was experienced by 2% of patients receiving trastuzumab, compared with 0.1% of control patients (p < .0001). Significant reductions in LVEF were observed in 3% and 0.5% of patients receiving trastuzumab and no treatment, respectively (p < .0001) [13].
In the Breast Cancer International Research Group 006 study, three adjuvant treatment arms were compared: (a) doxorubicin plus cyclophosphamide followed by docetaxel (AC Two different approaches have been used with the aim of diminishing, without reducing efficacy, the incidence of cardiotoxicity from adjuvant treatments including trastuzumab: (a) using adjuvant chemotherapy regimens without anthracyclines and (b) reducing the duration of trastuzumab administration. In a small randomized Finnish study, a regimen of adjuvant trastuzumab lasting only 9 weeks combined with chemotherapy (vinorelbine or docetaxel followed by 5-fluorouracil, epirubicin, and cyclophosphamide) was compared with chemotherapy alone [15]. After a median follow-up of 3 years, there were no reports of either severe CHF or asymptomatic decreased LVEF. However, this superior cardiac safety may not be attributable uniquely to the reduced duration of trastuzumab administration, because patients participating in the study received epirubicin instead of doxorubicin, and the cumulative dose (180 mg/m2) was lower than that in other clinical trials in the adjuvant setting.
Data from four large, international trastuzumab adjuvant therapy clinical trials have been pooled and analyzed. Grade III or IV cardiac toxicity was reported for 4.5% of the patients receiving trastuzumab-containing regimens, versus 1.8% of the patients who did not receive trastuzumab [36]. A separate meta-analysis including a total of 11,187 patients revealed that the relative risk for cardiotoxicity with adjuvant trastuzumab versus no trastuzumab was 5.59 (95% confidence interval [CI], 1.99–15.7; p = .011), with an absolute difference of 1.49%. Additionally, a relative risk for a reduction of
Early Breast Cancer: Neoadjuvant Therapy
Multidisciplinary Care Approach: Cardiology–Oncology The selection of breast cancer patients who are potential candidates to receive trastuzumab should be performed by a multidisciplinary team (MDT) comprised of cardiologists and oncologists. At the initial evaluation, a risk–benefit assessment regarding the decision to add trastuzumab to chemotherapy should be performed. First, clinical evaluation is needed, in addition to cardiac imaging, in order to identify additional potential risk factors for cardiotoxicity. Later, both clinical follow-up and cardiac monitoring on a regular basis are required. It is necessary for MDT staff to be familiar with trastuzumab-related cardiotoxicity, because this differs from that induced by chemotherapeutic drugs [40]. Chemotherapy-related cardiac dysfunction (CRCD) directly affects the myocardium and manifests as decreased LVEF, which may progress rapidly to CHF. Type I CRCD is defined as cardiac toxicity caused by anthracyclines and has been widely studied. Trastuzumab-related cardiac toxicity is referred to as type II CRCD. Its mechanism(s) has not yet been completely elucidated. Moreover, the strategy for diagnosis, follow-up, monitoring, treatment, and prevention remains to be established. The most important differences between type I and type II CRCD are that the latter is not dose-dependent, its clinical manifestations vary between patients, it may be reversible, and it apparently does not cause ultrastructural alterations within the myocardium [21, 41–43]. It is crucial to identify those patients for whom trastuzumab treatment should be resumed following initial discontinuation and intervention for decreased LVEF or CHF, because the long-term survival of breast cancer patients with left ventricular dysfunction (LVD) may be worse than that seen in the least advanced stages of the cancer. Thus, the MDT should know and compare the different prognoses of CHF and breast cancer. While the 6-year survival rate is 33% for women with CHF, the 5-year survival rate is 97% for women with very early breast cancer, 77% in the case of breast cancer with locoregional metastases, and just 22% for women with distant metastases [44].
Definitions and Classification of Cardiac Dysfunction
Evaluation of Potential Risk Factors
Currently, it is unclear whether or not classical cardiac risk factors, for example, hypertension and diabetes, are predisposing factors for trastuzumab-related cardiotoxicity. In the review conducted by Seidman et al. [21], a multivariate analysis for potential risk factors (i.e., age, hypertension, previous radiation therapy to the chest wall, cumulative anthracycline dose, baseline LVEF) was performed. Of these, only age (when trastuzumab was administered concomitantly with doxorubicin) was significantly positively associated with the risk of developing trastuzumab-related cardiotoxicity (risk ratio, 1.56 per 10-year increase in age). All trials with trastuzumab have excluded those patients with pre-existing CHF, and therefore the presence of LVD or CHF symptoms should exclude the use of trastuzumab even in those patients with borderline postchemotherapy LVEF <50%. In a retrospective clinical experience study, the authors attempted to identify those pre-existing factors that may be used as predictors for the development of trastuzumab-related cardiotoxicity [48]. Results suggested that advanced age, hypertension, radiation therapy to the left chest wall, or previous exposure to anthracyclines did not result in a higher risk for cardiotoxicity, but a clear correlation between baseline LVEF and the risk for cardiac events was established. However, the number of cardiac events was too small, which notably diminished the statistical power of the study. Diabetes, history of coronary artery disease, and valvular disease were all factors clearly associated with a higher incidence of cardiac events, though these associations did not reach statistical significance, probably as a result of this lack of sufficient statistical power. These data must be considered with caution, and not as definitive results, because in another study smoking, family history, taking lipid-lowering or hypoglycemic medications, and radiation to the left side of the chest were not identified as risk factors [49]. Other conditions usually considered to exclude patients from enrollment in large trials of trastuzumab are a history of myocardial infarction or angina, uncontrolled hypertension, valvular disease, or arrhythmia, so limited knowledge exists regarding the effects of trastuzumab in patients with these conditions. Every patient should be interrogated regarding these problems and, if any are present, treatment with trastuzumab should be reconsidered and a risk–benefit assessment should be carefully performed.
Evaluation of Cardiac Function Before and During Trastuzumab Therapy Electrocardiography (ECG) should be routinely used in every patient prior to trastuzumab therapy, although rhythm disturbances or intraventricular conduction abnormalities usually appear only in advanced stages of cardiac disease. Currently, a diagnosis of CHF relies on clinical evaluation based on history, physical examination, and complementary tests [47]. The main components are the presence of symptoms (fatigue, breathlessness, ankle swelling) and objective evidence of cardiac dysfunction at rest. But such symptoms are often nonspecific and may mimic those of pulmonary disease. Furthermore, although a large proportion of patients with cardiac dysfunction induced by trastuzumab present with clinical impairment (64%) [21], a considerable number of them remain asymptomatic. The use of adequate imaging methods to monitor cardiac function during treatment is therefore of critical importance in these patients. Alongside ECG, measuring plasma concentrations of natriuretic peptides can be helpful in ruling out significant cardiac disease, whereas elevated concentrations may indicate the presence of diastolic dysfunction [47]. However, the HFSA recommends using this test only in patients presenting with symptoms of suspected HF [46]. Echocardiography is the most widely used imaging method for the evaluation of left ventricular function. It is also valuable for the detection of pulmonary hypertension, myocardial damage, valvular disease, and hemodynamic abnormalities [45]. However, measurement is somewhat dependent upon the individual who performs the procedure and on the acoustic window of the patient, which is one of the main determinants of image quality. The rates of inter- and intraobserver variability are relatively high, and variations in LVEF >10% are often found in the absence of any true modification [12]. Three-dimensional (3D) echocardiography and cardiac magnetic resonance imaging (MRI) have greater reproducibility in evaluating LVEF [51–53]. This technique provides morphological, functional, perfusion, and viability information in one assessment. It is expensive and time-consuming but is the diagnostic method of choice for patients with technically limited images from ECG and in patients with discordant information that is clinically significant from prior tests [54].
The British Society for Echocardiography has published recommendations for LVEF monitoring in breast cancer patients receiving trastuzumab [55]. These state that an accurate value for LVEF can be determined by 3D echocardiography, provided that the technique is performed by an experienced technician. The technique should be reliable enough to identify 10% changes from baseline in LVEF. In a small percentage of patients, notably in those undergoing surgery or local radiation therapy, the echocardiographic image quality is not good enough; therefore, these patients require alternative imaging techniques. The guidelines suggest LVEF determination prior to trastuzumab treatment initiation and every 3 months thereafter. It is not recommended to start trastuzumab therapy if the LVEF is
Management of Trastuzumab-Associated Cardiotoxicity Following such treatment, repeat echocardiographic evaluations should be performed until symptom and sign stabilization and recovery of ventricular function. In many cases, following such an approach, trastuzumab therapy can be resumed following recovery of LVEF, without the development of further LVEF impairment [28, 38, 43]. However, one in five patients does not regain normal cardiac function following appropriate cardiac therapy [21]. If an individual patient experiences new episodes of HF upon trastuzumab resumption, treatment should be definitively discontinued. As stated previously, it is necessary to perform a risk–benefit analysis on an individual basis when making a decision to continue with trastuzumab in patients presenting either with risk factors or with systolic/diastolic dysfunction. In a percentage of patients who experience cardiac damage, trastuzumab resumption should be considered because of the advanced stage of their cancer, that is, patients with MBC [56], because the risk for cardiac dysfunction is offset by the benefit of antitumor therapy. Although management guidelines published by international cardiology societies have not, as yet, described specific medical therapy for the treatment of trastuzumab-related cardiotoxicity, there is a wealth of evidence to support the use of certain drug classes in any patients with asymptomatic decreased LVEF or HF. The American College of Cardiology/American Heart Association recommend the administration of ACEIs or angiotensin II receptor blockers (ARBs) and beta-blockers in patients with asymptomatic decreases in LVEF, whereas these drugs plus diuretics for the specific treatment of fluid retention are recommended in patients with symptomatic HF [45]. The European Society of Cardiology (ESC) and the HFSA recommend ACEIs as first-line therapy in patients with LVEF <40%, with or without symptoms, administered at doses demonstrated to be effective in large, controlled trials [46, 47]. In patients intolerant to ACEIs, ARBs may be substituted. Diuretics are administered for the symptomatic treatment of fluid retention. According to the ESC guidelines, beta-blockers are added to initial therapy for all patients with NYHA class II–IV HF, starting with a low dose and titrating upward to those demonstrated to be effective in clinical studies of HF [47]. Cardiac glycosides such as digoxin may be beneficial in combination with a beta-blocker in patients with symptomatic HF. The HFSA also recommends beta-blocker therapy for asymptomatic decreased LVEF [46]. A proposed algorithm for the management of patients receiving trastuzumab is presented in Figure 1.
Patient Education In clinical practice, the decision-making process is usually shared between the physician and the patient. Patients make their decisions based not only on the information and guidance provided by the physician, but also on their preferences, moral values, and personal life goals. From an ethical and legal point of view, the adult competent patient has the right to accept or reject treatments proposed by his/her physician. In the case of breast cancer treatment, the physician should first clearly explain the different therapeutic options that are available in the individual case, including expected efficacy and adverse effects, especially irreversible toxic effects. Secondly, the physician must obtain the patient's informed consent, but only when they are certain that the patient has understood the risks, advantages, and disadvantages of the proposed treatment, as well as alternatives. In oncology, the risk–benefit analysis is a routine procedure, based on two major ethical principles: to do what is considered the best for the patient and to not cause him/her excessive damage. In addition, there is another principle that tips the balance: to preserve life, even at the expense of its quality. For the information and education of patients with breast cancer who are candidates for trastuzumab therapy, the following recommendations may be useful:
Trastuzumab may be associated with cardiac dysfunction (mostly an asymptomatic decline in LVEF or overt HF) in a small but significant proportion of patients. Prior to initiating treatment with trastuzumab, a risk–benefit analysis for individual patients must be performed, taking into account their prognosis, pre-existing risk factors, and personal preference. Monitoring of cardiac function must be performed throughout the treatment period and also during follow-up. In many cases, early recognition of trastuzumab-associated cardiac dysfunction can allow effective medical intervention, symptom resolution, and continuation with trastuzumab therapy.
Conception/design: Miguel Martín, Francisco J. Esteva, Bijoy Khandheria, Emilio Alba, Leopoldo Pérez-Isla, José Ángel García-Sáenz, Antonia Márquez, Partho Sengupta, José Zamorano Administrative support: Miguel Martín, José Zamorano Collection/assembly of data: Miguel Martín, José Zamorano Data analysis: Miguel Martín, Francisco J. Esteva, Bijoy Khandheria, Emilio Alba, Leopoldo Pérez-Isla, José Ángel García-Sáenz, Antonia Márquez, Partho Sengupta, José Zamorano Manuscript writing: Miguel Martín, Francisco J. Esteva, Bijoy Khandheria, Emilio Alba, Leopoldo Pérez-Isla, José Ángel García-Sáenz, Antonia Márquez, Partho Sengupta, José Zamorano Final approval of manuscript: Miguel Martín, José Zamorano
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||