Tomorrow's targeted therapies in breast cancer patients: What is the risk for increased radiation-induced cardiac toxicity?

2.1. Trastuzumab: the unexpected potential for cardiac toxicity 

Approximately 20–25% of human breast cancers overexpress or amplify the human epidermal growth factor receptor 2 (HER2), which is associated with a poor prognosis [3]. Humanized monoclonal antibody trastuzumab (Herceptin; Genentech Inc., South San Francisco, CA) selectively binds to the extracellular domain of the HER2 protein [4]. The addition of trastuzumab to conventional chemotherapy resulted in a survival advantage in patients with metastatic disease [3]. However, an unexpected adverse event observed during the pivotal trials of trastuzumab in the metastatic setting was cardiac dysfunction, improving in 75% of patients after the initiation of standard medical management. The incidence of severe cardiac dysfunction was highest among patients who had received an anthracycline, cyclophosphamide, and trastuzumab (16%) [3]. Although these observations have encouraged further investigations aimed to explore the mechanisms of toxicity, the pathophysiology of cardiac morbidity associated with trastuzumab is not clearly understood. In contrast to radiation or anthracycline cardiac toxicity, trastuzumab-related cardiac toxicity is not associated with ultrastructural changes. Three hypotheses have been suggested for cardiac toxicity of trastuzumab: inherent toxicity, sequential stresses following anthracycline administration, and observational artefacts [5]. The potential role of the HER2 receptor in myocyte survival or maintenance of cardiomyocyte contractility was also suggested [6]. It was demonstrated that HER2 was essential for the normal development of embryonic cardiac cells. Moreover, the neuregulin-1/ErbB signaling pathway is involved not only in the cardiac development, but also in the regulation of cardiac sympathovagal balances through interactions with the muscarinic cholinergic system and by counterbalancing adrenergic stimulation of the myocardium [7]. Preclinical studies also suggested that HER2 signaling pathway was essential for the prevention of dilated cardiomyopathy. Finally, loss of HER2 function may render cardiac cells more sensitive to anthracycline toxicity. In a retrospective analysis of data collected from 1219 patients included in clinical trials, it was reported that most trastuzumab-treated patients who developed cardiac dysfunction were symptomatic, but the majority of patients (79%) responded to medical therapy [8]. Increasing age and chemotherapy with anthracycline or cyclophosphamide were statistically significant predictive factors. A previous history of radiotherapy was not found to be significantly associated with cardiac function. It was suggested that the cardiac risk associated with the use of trastuzumab in the metastatic setting would be justified, given the poor prognosis associated with HER2-positive breast cancer and the 25% improvement in overall survival. In the adjuvant setting, this risk should be taken into consideration. In 2004, Perez and Rodeheffer reviewed the current understanding on the clinical cardiac tolerability of trastuzumab in the clinical settings. Patients treated with trastuzumab were found to have increased risk for cardiac dysfunction [2]. This risk was greatest in patients receiving concurrent anthracyclines.

Although epirubicin seems to be a less cardiotoxic anthracyclin, further evaluation of this regimen concurrently with trastuzumab is mandatory. Decreases of ejection fraction and a few cases of CHF requiring medical therapy had been detected. Increased vigilance is however recommended for higher risk patients and concurrent administration of trastuzumab with anthracyclines should be avoided [5].

2.2. Cardiac toxicity of radiation therapy 

In contrast to trastuzumab toxicity, radiation-induced toxicity is a late and irreversible effect that can be manifest 10–15 years after radiation therapy. Early cardiac toxicity from radiotherapy results from direct cellular toxicity through apoptosis, necrosis, and endothelial cell injury, leading to increased vascular permeability and stromal oedema. Delayed vascular toxicity includes epithelial nuclear atypia and development of multinucleate stromal fibroblasts, with subsequent intimal thickening, fibrinoid necrosis, medial hyalinization and parenchymal atrophy. Heart irradiation may favor fibrosis and thickening of the pericardium, occasionally leading to pericarditis [9], [10], myocardial fibrosis and endocardial fibrosis, which may cause valve disease [9], [10], [11], [12], [13], [14]. Accelerated coronary artery fibrosis and atherosclerosis after irradiation have also been reported. Early ECG changes are frequent. The excess risk of radiation-induced cardiac disease has been documented in many trials conducted in postmastectomy chest wall radiotherapy, but also for patients with conservative surgery followed by adjuvant radiotherapy [15], [16], [17], [18], [19], [20], [21], [22], [23]. The Early Breast Cancer Trialists’ Collaborative Group (EBCTCG) meta-analysis of approximately 20,000 women enrolled in 78 randomized trials of postoperative radiotherapy found a 30% increase in cardiac mortality for patients treated in the 1960s and 1970s [24]. A number of randomized trials have reported increased rates of cardiac events in patients with adjuvant radiotherapy. These studies provided important information on the long-term cardiac risks related to radiotherapy. It was advocated that the excess of non-breast cancer deaths among patients who received radiotherapy was mainly due to heart disease and mainly in older trials, which used outdated radiation techniques that exposed more volume of heart to larger doses of irradiation than current practice. The newer megavoltage techniques have significantly reduced the irradiated cardiac volume. Long-term outcome data in the era of tangential radiation have become available, showing more subtle cardiac effects of irradiation. These recent technical changes facilitated sparing the heart and coronaries from unnecessary irradiation, but did not annihilate the risk for subsequent heart disease. High irradiation is still delivered to a small segment of the anterior wall, which partially includes the left anterior descending artery and other major vessels receiving a smaller dose. For this reason, internal-mammary chain (IMC) irradiation is presumed to contribute to a greater dose of radiation to the heart when added to conventional breast radiotherapy. Similarly, it has been shown that patients with left-sided breast tumors treated with adjuvant radiotherapy stand a significantly higher risk for cardiac events, including fatal myocardial infarction 10–15 years later compared with postoperative radiotherapy for patients with right-sided tumors [24].

2.3. What about combination therapy? 

Preclinical results suggested that trastuzumab may enhance radiosensitivity with a dose-modifying factor of 1.11 [25]. However, there is no strong demonstration that trastuzumab may interact with radiation. If there is a potential risk for increased radiation-induced cardiac toxicity, it is believable that the potentiation of effect would be more likely to be additive rather than to be synergistic. Six randomized adjuvant trials of trastuzumab in combination with chemotherapy have been reported [26], [27], [28], [29], [30]. Up to 4% of patients experienced severe CHF during treatment. Reported as a joint analysis, the North Central Cancer Treatment Group (NCCTG) N9831 and the National Surgical Adjuvant Breast and Bowel Project (NSABP) trial B-31 compared adjuvant chemotherapy with or without concurrent trastuzumab [26]. In both trials, radiotherapy was initiated after the completion of chemotherapy, without irradiation of the internal-mammary nodes. A 33% decrease in the risk of death specific cancer was demonstrated, leading to the adoption of this targeted agent into clinical adjuvant practice for HER2-positive breast cancer patients. The 3-year cumulative incidence of class III or IV CHF or death from cardiac causes in the trastuzumab group was 2.9% in trial N9831 and 4.1% in trial B-31. In the N9831 trial, the cumulative incidence of cardiac events at 3 years was higher in patients treated with trastuzumab (2.8–3.3%) versus the control arm (0.3%). The incidence of asymptomatic left ventricular ejection fraction (LVEF) decreases requiring holding trastuzumab was 8–10% [27]. In the NSABP trial B-31, a significant proportion of patients sustained decrements in LVEF to less than 50%. Trastuzumab was discontinued permanently for cardiac reasons in 133 (19%) of 714 patients, including asymptomatic decreases in LVEF in 102 (14.2%) of 714 patients. Twenty-one of these 102 patients subsequently developed symptoms. Risk factors for CHF included age more than 50 years, requirement for hypertension medications, and post-AC LVEF values of 50–54% [28]. Two thirds of patients with a cardiac event continued to require cardiac medications and 71% had a persistent cardiac dysfunction. In the unpublished Breast Cancer International Research Group trial (BCIRG) 006 trial, significant cardiac disease was observed in 1040 patients treated with adjuvant AC, then trastuzumab delivered concurrently with docetaxel and continued for 1 year. In this group, 180 (17.3%) patients sustained a more than 10% LVEF decrease relative to the baseline. Among 145 patients with a LVEF measurement after 42 days, 37 (26%) had persistent decline in LVEF [29]. In the Herceptin Adjuvant (HERA) trial, 94% of patients had received an anthracycline and 1308 (77.2%) patients had received radiotherapy [30]. No case of CHF was reported in patients treated with 9 weeks of trastuzumab. Only 4.3% of patients needed permanent trastuzumab discontinuation. In the Finher trial, trastuzumab was not associated with decreased LVEF or cardiac failure. Trastuzumab was administered before other cardiotoxic treatments and concomitantly with potentially synergistic chemotherapy for only 9 weeks [31]. Safety analysis of the PACS-04 trial found low cardiac toxicity with sequential trastuzumab [32].

Until recently, the cardiac effects of the interaction trastuzumab-irradiation therapy had not been adequately reported. There were major questions regarding the scheme of radiotherapy: what were the dose, fractionation, involved fields, energy and techniques used (2D or 3D)? The consequences of irradiating IMC fields in the right- and left-sided breast cancer patients were not discussed and heart dosimetry data were scarce. However, an important article was recently published in 2009, in which Halyard et al. examined tolerability and adverse event data from the NCCTG Phase III Trial N9831 and directly evaluated the effect of trastuzumab on radiation-induced cardiac toxicity [33]. At a median follow-up of 3.7 years, radiotherapy concurrently with trastuzumab did not increase relative frequency of cardiac events, regardless of treatment side. In 59 patients, Shaffer et al. recently reported on the acute toxicity of IMC irradiation concurrently with trastuzumab [34]. Median absolute decrease in LVEF was 4%, which was not significantly different according to side or inclusion of IMC. However, trastuzumab was stopped in 18.6% patients due to decrease in LVEF. Three patients developed clinical CHF, none of whom received left-sided IMC RT. Belkacémi et al. evaluated the acute toxic effects of such association in 146 patients, and concluded that weekly concurrent trastuzumab and radiotherapy are feasible in daily clinical practice. However, 10% of the patients had a grade ≥2 LVEF decrease [35].

Although the risk of severe CHF was low for trastuzumab in the adjuvant setting, the LVEF declines might be more sustained than previously believed. It is uncertain whether the safety of trastuzumab in the general study population would be reported, and there is a need to focus not only on severe symptomatic cardiac disease. Minor cardiovascular changes may increase mortality with longer follow-up. Several patients experienced asymptomatic decrements in LVEF on follow-up evaluations more than 6 months after the onset of cardiac disease, suggesting a need for continued cardiac follow-up. The lack of thorough, long-term data regarding cardiac toxicity should necessitate caution when assessing concurrent trastuzumab and radiotherapy.

3.1. ErbB targeting 

The data regarding adjuvant trastuzumab, although suggesting that most of its cardiac toxicity can be medically managed, drew particular attention to the potential cardiotoxicity that might be related to new HER2-targeting agents, such as lapatinib. Dual inhibition of ErbB-1 and ErbB-2 tyrosine kinases has been found to exert greater biological effects in the inhibition of proliferative and survival signaling pathways than inhibition of either receptor alone. Lapatinib (Tyverb®; GlaxoSmithKline, Philadelphia), is an oral, small molecule, dual EGFR/ErbB-2 TKI. It was demonstrated that lapatinib was an effective regimen for women with refractory ErbB-2-positive metastatic breast cancer who had developed progressive disease following prior treatment with anthracyclines, taxanes, and trastuzumab [36], [37]. However, most patients with refractory ErbB-2-overexpressing breast cancer have already had exposure to potentially cardiotoxic treatments, including postoperative radiotherapy, anthracyclines and trastuzumab. Unlike the Herceptin metastatic trials, all patients had undergone regular cardiac evaluations, and of course, had normal baseline ejection fractions after prior Herceptin – as some would say, they had all passed the “Herceptin stress test”. Consequently, there is probably a selection bias when assessing the lapatinib-related cardiac toxicity: some high – risk patients were probably excluded from the lapatinib trials, since they developed cardiotoxicity while receiving trastuzumab. In the trial by Geyer et al., asymptomatic cardiac events were identified in four women in the group treated with lapatinib plus capecitabine (2.5%) and in one patient in the group treated with capecitabine alone (<1%) [37], which is less than reported with trastuzumab (up to 4% of severe CHF during treatment in the literature). However, all patients had LVEF values at or above the lower limit of the normal range on subsequent assessment. One patient developed Prinzmetal's angina. No symptomatic cardiac events were related and lapatinib was not discontinued because of a decrease in the LVEF. No significant difference was observed in the mean LVEF values between the two groups at scheduled assessments. Gomez et al. recently assessed the efficacy and tolerability of two lapatinib administration schedules (1500mg once daily, 500mg twice daily) as first-line monotherapy in women with ErbB2-amplified locally advanced or metastatic breast cancer [38]. Cardiac function was assessed at weeks 8 and 12 and every 8 weeks thereafter. Four patients (3%) experienced asymptomatic reductions in LVEF that represented greater than 20% relative decreases from baseline, all of whom previously received anthracyclines. No symptomatic cardiac events were reported. In a phase II study by Lin et al., no patient developed symptomatic CHF. Four patients developed asymptomatic declines in LVEF to less than 50% (range, 44–49%), one of them experiencing a 10% or greater decline in LVEF from baseline [39].

Perez et al. analyzed cardiac function in patients treated with lapatinib in 18 clinical trials and reported low levels for cardiac toxicity [40]. A total of 3689 patients who had received lapatinib were analyzed, including 2275 patients with breast cancer. Lapatinib infrequently affected the LVEF, with only 62 cardiac events reported in 60 patients 1.6% of patients (60 of 3688). At the time of the event, the mean nadir LVEF was 43.0±6.7% (95% CI: 20.0–54.5%). Asymptomatic cardiac events were reported in 53 patients and symptomatic cardiac events in 7 (0.2%), which is less than that of trastuzumab-treated breast cancer patients. The mean LVEF decrease was 18. 8% (95% CI: 11–32%). Time to onset of an LVEF decrease occurred within 9 weeks of treatment in most patients. Although promptly responding to standard cardiac management, symptomatic decrease in LVEF was observed in 0.2% of patients, with dyspnea, palpitations, and signs of CHF. Interestingly, most patients had received prior anthracycline-based chemotherapy, trastuzumab, and/or prior IMC or left-sided chest radiotherapy. Most decreases in LVEF were moderate. Thirty-five patients (58%) had a full (n=19) or partial (n=16) recovery, regardless of continuation or discontinuation of lapatinib. Five patients (8%) remained asymptomatic and did not recover within the observation period. Data were insufficient to assess recovery in 20 patients (33%). Unlike anthracycline-induced cardiac failure, most of cardiac effects of lapatinib are reversible and non-progressive [41]. The toxicity of ErbB2-targeting with trastuzumab suggests that lapatinib might also be a potential cardiac hazard. Further evaluation is warranted in current and future lapatinib clinical investigations, but the analysis of literature suggests that lapatinib-induced CHF would be a rare complication.

3.2. VEGF targeting 

Recently, antiangiogenic therapy with the monoclonal antibody bevacizumab has demonstrated significant activity in patients with metastatic breast cancer. Bevacizumab (Avastin, Genentech) is a humanized monoclonal antibody directed against all isoforms of VEGF-A, which is potent in inducing vasodilatation and pathologic angiogenesis. Bevacizumab has been shown to have biological activity in breast cancer, alone or in combination with systemic chemotherapy agents such as docetaxel [42], capecitabine [43], and vinorelbine [44].

Torrisi et al. investigated the activity of bevacizumab in combination with chemotherapy, including capecitabine and vinorelbine, and endocrine therapy [45]. In 36 patients assessable, a clinical response rate of 86% (95% CI: 70–95) was observed with manageable toxicity. Two cases of grade 3 hypertension and four grade 3 deep venous thromboses were observed. Ramaswamy and Shapiro evaluated bevacizumab and weekly docetaxel as first- or second-line therapy in patients with metastatic breast cancer [42]. Although active with acceptable overall toxicity, the combination was associated with cardiovascular events. Two pulmonary emboli (7%) and 1 case of hypertension (4%) were observed. Miller et al. demonstrated that bevacizumab plus paclitaxel significantly prolonged progression-free survival, when compared to paclitaxel alone as initial treatment for metastatic breast cancer [46]. Hypertension was more common in patients receiving bevacizumab (14.5% versus 0%, p<0.001). Grade 4 hypertension requiring discontinuation of bevacizumab developed in only one patient. Although thromboembolic events were infrequent, there was a significant increase in cerebrovascular ischemia (1.9% versus 0.0%, p=0.02). No significant difference was observed in terms of left ventricular dysfunction. While the median follow-up was too short for assessment of long-term cardiac toxicity, careful assessment remains necessary, as most breast cancer patients have received or will receive anthracycline-based chemotherapy.

In a prospective phase III trial including 462 heavily pre-treated breast cancer patients, bevacizumab was demonstrated to be well tolerated in combination with capecitabine (2500mg/m2/day, twice daily on days 1–14 every 3 weeks) and to produce a significant increase in response rates when compared to capecitabine alone [43]. Although increased in the combination arm, hypertension was generally managed with medical therapy, no patient experiencing grade 4 hypertension. Four patients discontinued bevacizumab because of hypertension. Seven patients developed grade 3 or 4 CHF or cardiomyopathy in the combination group, versus two in the capecitabine-alone group. However, baseline LVEF was less than 50% in three of bevacizumab-treated patients who developed CHF or cardiomyopathy, suggesting that previous history of systemic treatments may have partially favored the occurrence of subsequent heart failure. Cardiac symptoms improved with medical management in all but one patient. Overall, CHF occurred in 3.1% of the patients treated with bevacizumab and capecitabine and in 0.9% of those with capecitabine alone.

D’Adamo et al. assessed bevacizumab in combination with doxorubicin in 17 patients with metastatic soft-tissue sarcoma and reported high cardiac toxicity rates [47]. Six patients developed cardiac toxicity of grade 2 or more: four patients with grade 2 (cumulative doxorubicin 75, 150, 300 and 300mg/m2, respectively), one with grade 3 (total doxorubicin 591mg/m2), and one with grade 4 (total doxorubicin 420mg/m2).

Up today, there is no clear evidence on how the anti angiogenic agents may cause heart disease and the mechanisms of VEGF inhibitors-related hypertension remain poorly understood. Serum aldosterone, catecholamine, and renin levels would not be affected by VEGF inhibitors [48]. Other authors suggested that VEGF targeting could be responsible for cholesterol emboli syndrome which may account for bevacizumab-induced acute cardiovascular complications, including hypertension [49]. It was demonstrated that bevacizumab treatment resulted in endothelial dysfunction and capillary rarefaction, which are could be responsible for the hypertension observed in patients treated with VEGF inhibitors [50]. VEGF would enhance endothelial nitric oxide (NO) synthase activity, which is implicated in the response pattern to angiogenic growth factors. VEGF blockade and NO inhibition may result in several nonspecific cardiovascular effects. Preliminary results suggested that bevacizumab would enhance radiation response in animal tumor model systems [51]. Blocking the radiation-mediated increase in VEGF with bevacizumab may produce additive antitumor activity [52]. In vitro data demonstrated that the anti-VEGF treatment potentiates radiation-mediated lethality of human umbilical vein endothelial cells, suggesting that the induction of VEGF by irradiation contributes to the protection of tumor blood vessels from radiation-induced cytotoxicity [53]. As long as we have little prospective data and no long-term follow-up, it is important to remain careful when assessing new therapeutic approaches combining bevacizumab with therapies that may contribute to cardiac toxicity, including anthracyclines, trastuzumab, and radiation therapy.

3.3. Multitargeted tyrosine kinase inhibitors (TKIs) 

Sunitinib (Sutent; Pfizer) and sorafenib (Nexavar; Bayer Pharmaceuticals Corporation) are oral inhibitors of growth factor receptors, most important of which are the vascular endothelial growth factor (VEGF) receptor, platelet-derived growth factor receptor, and stem cell factor TKI receptor. Both have been proven effective as single-agent therapies in renal cell carcinoma (RCC), with potent antiangiogenic and antitumor activity. By targeting hypoxia-inducible (HIF) gene pathway and subsequent inhibition of HIF-induced gene products, both agents affect tumor angiogenesis and tumor cell proliferation [54]. Sorafenib affects angiogenesis and cell proliferation by targeting the MAPK pathway at the level of Raf kinase and/or tumor cell apoptosis induction [55]. Sorafenib also inhibits VEGF receptors VEGFR-1, VEGFR-2, VEGFR-3, and platelet-derived growth factor receptor-β tyrosine kinase autophosphorylation [56]. Targeting multiple Raf isoforms with sorafenib may overcome resistance to other systemic agents. Moreover, the ability of TKIs to induce apoptosis may increase the cytotoxicity of irradiation. Preclinical studies demonstrated activity in preclinical models, encouraging further clinical assessment [57]. These properties provide a strong rationale for investigating the use of VEGF inhibitors in combination with other agents in breast cancer patients, including radiotherapy. Sunitinib was evaluated in 64 patients with metastatic breast cancer, 50mg/day in 6-week cycles (4 weeks on, then 2 weeks off treatment) [58]. Seven (11%) patients achieved a partial response. Median time to progression and overall survival were 10 and 38 weeks, respectively. Interestingly, responses occurred in triple negative tumors and HER2-positive, trastuzumab-treated patients. Although most side effects were of mild-to-moderate severity, ten patients (16%) developed grade hypertension, four of whom had grade 3 hypertension (6%). Preliminary studies reported potential efficacy of combining sorafenib and docetaxel in breast cancer patients [59]. Sorafenib is also currently investigated in combination with anastrozole [60]. However, safety data from RCC indicate that cardiac damage from TKI would be largely underestimated. Khakoo et al. retrospectively conducted a study during a 1-year period on patients who received sunitinib, and reported that 6 of 224 (2.7%) patients developed heart failure. Symptomatic heart failure occurred soon after initiation of sunitinib and was associated with decline in cardiac function and elevations in blood pressure. In most patients, this was not completely reversible [61].

Left ventricular dysfunction after TKI treatment has been reported and is hypothesized to be partially due to direct cardiomyocyte toxicity and exacerbated by hypertension [62]. Targeted by both angiogenesis inhibitors, HIF-1-related gene products are physiologic mediators of myocardial remodeling, acute and chronic ischemia and vascular permeability [63], [64]. Chu et al. examined sunitinib's effects in preclinical models and demonstrated that TKI may induce apoptosis in cardiomyocytes. They observed that mice treated for 12 days with sunitinib had striking abnormalities of cardiomyocytes, including mitochondrial swelling and degenerative changes. It was also shown that sunitinib may target mitochondria in cultured cardiomyocytes, leading to cytochrome c release into the cytosol, which is an activator for cell death and apoptosis [65]. Moreover, the disruption of the VEGF signaling pathway is involved in capillary density, contractile dysfunction, fibrosis, and CHF [66].

In metastatic RCC patients, the use of sunitinib resulted in a decline of LVEF by 10% [67]. Cardiac ischemia was observed in 3% [68]. In gastrointestinal stromal tumors, 11% of patients given sunitinib developed cardiovascular events, with CHF recorded in 8%. Hypertension was observed in 47% [65].

When analyzing the outcome of 74 patients intended for sunitinib or sorafenib treatment for RCC, 33.8% experienced a cardiac event, 40.5% had ECG changes and 18% were symptomatic. Although seven patients (9.4%) were seriously compromised, all of them recovered after cardiovascular management and were considered eligible for TKI continuation [54]. Telli et al. reported a greater than 10% incidence of symptomatic grade 3/4 heart failure after delivering sunitinib for either RCC or gastrointestinal stromal tumor [69]. Rarely, sunitinib may prolong the QT interval, resulting in an increased risk of ventricular arrhythmias. This evidence together suggests that blood pressure and the LVEF should be closely monitored in those patients with a previous history of coronary artery disease or cardiac risk factors who are treated with a TKI. For patients without cardiac risk factors, a baseline evaluation of the ejection fraction should be considered [70]. Caution should also be used in patients with a history of QT prolongation and in those taking antiarrhythmics, or patients with electrolyte disturbances.

4.1. A need for new treatment modalities 

In a few years, there is no doubt that new targeted therapies will be assessed in breast cancer patients, including in the adjuvant setting. There might be addition of these toxicities with that from adjuvant systemic therapies or RT [71]. Long-term data in the Hodgkin disease area showed that the risk of severe cardiac toxicity following doxorubicin and mediastinal RT was greater than that reported in patients treated with RT alone [72]. In breast cancer patients, anthracycline-based adjuvant chemotherapy has been associated with the development of a dilated cardiomyopathy, left-sided congestive heart failure, but also subclinical cardiomyopathies [73]. Although data from adjuvant trials were not significant, it was also suggested that aromatase inhibitors may be associated with a slightly greater incidence of cardiovascular events [74]. Every effort should be made to avoid irradiating the heart in order to substantially decrease the risk of death from cardiac disease associated with radiotherapy. However, data regarding the interaction of breast irradiation and targeted therapies with potential toxicity are limited since these agents have been in use a shorter time and thus lack long-term outcome data. Improvements in irradiation modalities, including sophisticated techniques minimizing cardiac dose, such as gating or intensity-modulated irradiation, may improve the set-up of left-sided tumors and might be critical in preventing cardiac disease [75], [76], [77]. The emerging principle of radiotherapy for breast cancer should be to minimize the dose to the heart as much as possible. Although the use of CT-planning for breast cancer has only recently become more widespread in clinical practice, 3D planning may be particularly useful for minimizing cardiac disease because it allows for accurate quantification of heart dose and greater planning flexibility to adapt treatment planning to each patient's geometry. In this setting, Image Guided Radiation Therapy (IGRT) may be also of value for improving patient's repositioning accuracy. However, heart dosimetric data are scarce, and it is unknown which measures of the heart dose or volume would be most relevant to subsequent heart disease [78]. Most guidelines recommend that less than 30% of the heart should receive more than 40Gy, which leads to a probability of 5% complications within 5 years. However, these data are discussed and there is evidence that low doses can also cause heart disease. A substantial amount of relevant information should be required in further clinical trials, including doses delivered to cardiac structures, especially the coronary arteries. Conformal planning or intensity-modulated radiotherapy (IMRT) techniques reduce cardiac irradiation while maintaining efficient target coverage. Preliminary results suggested that IMRT plans would reduce the maximal dose to the left ventricle by a mean of 30% and the average heart volume exposed to more than 30Gy from 45cm3 to less than 6cm3 in patients with left-sided breast cancer [79]. Considering the hazard for cardiac toxicity, three-dimensional CT-planning and IMRT optimization should be considered in further trials for more cardiac sparing. Proton therapy is another highly conformational radiotherapy device, which may theoretically contribute decreasing the dose delivered to the critical organs. Due to isodose conformation, Helical Tomotherapy (HT) provides a potential tool to decrease the risk of cardiac toxicity and to better spare the heart from high-dose irradiation [80]. However, low doses are distributed to larger heart volumes, with unknown long-term consequences on cardiac function. Substantial reductions of radiation doses to heart can also be achieved using breathing adaptation of adjuvant radiotherapy following conservative surgery for breast cancer [81].

4.2. A need for new tools of assessment 

A Cardiac Guidelines Consensus Committee was established to provide oncologists with some guidance regarding the management and prevention of cardiac toxicity during adjuvant trastuzumab therapy [82]. Given the similarities between trastuzumab- and lapatinib-induced cardiac toxicity, Moy and Goss suggested that these consensus guidelines should serve as a basis for assessing patients treated with lapatinib [41].

We believe that this principle of precaution should be extended to patients with multitargeted tyrosine kinase or VEGF inhibitors in the adjuvant setting, particularly in combination with radiotherapy. Pretreatment assessment of cardiac function, which may be obtained by multigated acquisition scan or echocardiogram, is crucial. Cardiac surveillance in asymptomatic patients is more controversial, but close surveillance remains necessary during treatment and a long time after its completion. High-risk combinations should be avoided in patients with a subnormal LVEF. To be able to determine which patients should be targeted with cardioprotective strategies, there is a need for improvement in cardiac assessment. Appraisal of tomorrow's treatments implicates new risks for tomorrow's cardiac toxicity, justifying the design of a more comprehensive approach to cardiac toxicity assessment. Traditionally, chemotherapy-induced cardiotoxicity has been detected by measuring changes in LVEF, which is insensitive to subtle changes in myocardial function. There is growing evidence that myocyte damage occurs with chemotherapy at lower dosages than can be appreciated by declines in the LVEF values [83]. Regarding that the myocardium must have undergone sufficient damages to exhibit a decrease in LVEF, unchanged values in LVEF is not equated to a lack of cardiotoxicity.

While cardiac biomarkers may appear necessary to detect cardiac toxicity at earliest stage, the incorporation of a combination of biomarkers that reflect myocardial cell damage, ventricular function, renal function, and inflammation to a model with established risk factors may add substantial information with respect to the risk of subsequent cardiac toxicity following radiotherapy [84]. Recent data in 54 disease-free breast cancer survivors suggested that brain natriuric peptide (BNP) measurement might be of value to identify patients requiring intensive cardiac follow-up [85]. From a median follow-up of 2.7–6.5 years, median BNP was raised almost three-fold (p<0.001). Although breast irradiation does not commonly result in elevations of serum Troponin T or Creatine kinase-MB, the prognostic value of subtle changes in those biomarkers shortly after left breast irradiation should be assessed. New techniques for assessment of cardiac function may also help in predicting the cardiac toxicity of tomorrow's combination therapies. Those may include computed tomography of the coronary arteries. Other newer echocardiographic function estimators may be based on automated border detection algorithms and ultrasonic integrated backscatter analysis [86]. Finally, peripheral blood telomeres were shown to be correlated with mortality from cardiac disease [87]. Loss of telomere function is associated with increased radiation sensitivity and loss of cellular viability and renewed potential. Further assessment of telomere status would be of interest to identify a potential relationship between increased telomere shortening and cardiac toxicity.