The definition of chemotherapy-induced cardiomyopathy has not been precisely defined. Various trials have employed different definitions of this entity. The Cardiac Review and Evaluation Committee (CRCE) has established the following criterion to diagnose chemotherapy induced cardiac dysfunction [4]:

Anthracyclines still remain the cornerstone of many modern chemotherapeutic regimens, with doxorubicin being the most widely used agent. These agents intercalate between specific bases in the DNA and inhibit DNA and RNA synthesis. Cardiotoxicity with these agents was recognized very early and found to be dose-dependent [5]. Initial studies, performed in 1970s, recommended a maximum total doxorubicin dose 550 mg/m² [6]. However, these studies were limited by lack of evaluation for asymptomatic LV dysfunction. Recent studies report incidence of LV dysfunction to be as high as 50 % with long-term follow up [7]. Swain et al. [8] reported incidence of cardiomyopathy at 5 %, 16 % and 26 % respectively, with cumulative doses of 400, 500 and 550 mg/m². Hence, currently the cumulative dose is usually limited to 400–450 mg/m². However, there is no safe dose and cardiomyopathy has been reported with doses as low as 250 mg/m², especially in children [9].

Three different manifestations of anthracycline-mediated cardiotoxicity have been described [10]:

Various risk factors have been described for the development of anthracycline-mediated cardiomyopathy (Table 13.3) [11]. The most important risk factor is the total cumulative dose. Other major risk factors include extremes of age, pre-existing cardiovascular disease, concomitant use of other cardiotoxic drugs and chest irradiation.

Anthracyclines produce progressive and irreversible type I cardiomyopathy. Cardiac biopsy typically shows patchy areas of interstitial fibrosis, vacuolation and rarely frank necrosis [12]. The pathophysiology of anthracycline-medicated cardiomyopathy is not clearly understood and multiple mechanisms have been hypothesized [13]. The most commonly accepted hypothesis is generation of excessive reactive oxygen species (ROS) and free radical induced myocyte damage. Various mechanisms have been hypothesized for excess free radical production, including mitochondrial dysfunction, increased endothelial nitric oxide synthase production, iron dependent redox cycling, and NAD(P)H dependent mechanisms [13]. However, recent literature suggests that the ROS hypothesis may not fully explain the anthracycline-induced cardiotoxicity [13]. Other hypothesized mechanisms include inhibition of topoisomerase II, DNA cross-linking, decreased ATP production, direct damage to the mitochondria and cell membranes, and increased apoptosis.

Several attempts have been made to reduce doxorubicin cardiotoxicity:

Trastuzumab is a monoclonal antibody active against human epidermal growth factor (HER2) receptor, which is overexpressed in 25 % of breast cancers. However, trastuzumab significantly increases the risk of cardiomyopathy. A meta-analysis of randomized clinical trials with use of trastuzumab as an adjuvant chemotherapeutic agent showed a 1.6 % absolute increase in incidence of symptomatic heart failure and 7.2 % increase in LV systolic dysfunction [17]. Another trial in patients with metastatic breast cancer showed a 19 % increased risk of cardiomyopathy when trastuzumab was used in combination with anthracycline and cyclophosphamide, and a 12 % increased risk when added to paclitaxel [18]. Analysis of the SEER-Medicare database showed a 32.1 % incidence of cardiomyopathy in patients receiving trastuzumab and a 41.9 % incidence in patients receiving anthracycline plus trastuzumab, which is much higher than previously reported trials [19]. Most protocols recommend use of trastuzumab sequentially with anthracyclines. However, even sequential use is associated with an increased risk of cardiomyopathy, though much less than concomitant use [20] (See Table 13.4).

Trastuzumab produces type II cardiomyopathy, which is not dose dependent and potentially reversible. There is no visible myocyte damage on histology and changes are visible only on electron microscopy [21]. Some authors have re-challenged patients with trastuzumab without recurrence of cardiomyopathy in most patients [4]. However, many authors have questioned its reversible nature and reported a 20–40 % incidence of persistent LV dysfunction [22]. MRI studies have shown evidence of delayed gadolinium enhancement despite recovery of cardiac function, suggesting persistent myocardial damage [23]. Long-term studies are needed to better define the natural history of trastuzumab-induced cardiomyopathy.

The mechanism of trastuzumab-induced cardiomyopathy is not well understood, but inhibition of ErbB2 is thought to be the main mechanism. ErbB2 is a critical component of multiple anti-apoptotic pathways and is necessary for myocyte survival and repair. Trastuzumab binds to the ErbB2 on cardiac myocytes and blocks the cardioprotective ErbB2-ErbB4 signaling pathway [24]. Removal of trastuzumab leads of resumption of these pathways and recovery of cardiac damage. The synergistic cardiotoxic effects of anthracycline and trastuzumab can be explained by myocyte damage due to anthracyclines and blockage of repair mechanisms by trastuzumab [25].

A few approaches to reduce trastuzumab-induced cardiotoxicity are under investigation:

Cyclophosphamide is activated in the liver and its active metabolite crosslinks DNA, disrupting cell division. It has been associated with increased risk of cardiomyopathy, especially in combination with an anthracycline or cisplatin [30]. Other risk factors include advanced age and mediastinal irradiation. The mechanism of cardiotoxicity is not very well established and appears to be related to the strength of the individual dose, rather than cumulative dose [30]. Ifosamide has also been rarely associated with cardiomyopathy [31].

Paclitaxel, alone, has not been implicated as a cause of cardiomyopathy. However, it reduces the elimination of doxorubicin and increases its toxicity [32]. Paclitaxel should be avoided immediately before doxorubicin administration and within the next hour. The most common cardiovascular side-effect of paclitaxel is transient asymptomatic bradycardia [33].

5-Flourouracil (5-FU) and its oral prodrug capecitabine have been associated with cardiotoxicity. The most common cardiac side effect is myocardial ischemia, likely related to coronary vasospasm [34]. Cardiomyopathy from these agents is rare and limited to a few case reports. It appears to have a type II pattern, with recovery of cardiac function in most cases after cessation of the agent [35].

Bevacizumab is a recombinant monoclonal antibody against the vascular endothelial growth factor (VEGF) receptor. It has primarily been associated with an increased risk of thrombotic events. The risk of cardiomyopathy is very low and follows the reversible type II pattern [36]. The most probable hypothesis is the loss of protective effects of VEGF against endothelial dysfunction caused by excess oxidative stress.

The most common cardiovascular effect of tyrosine kinase inhibitors is hypertension. Sunitinib has been associated with a 6–8 % incidence of HF and up to a 19 % incidence of cardiomyopathy, especially in patients with pre-existing coronary artery disease and cardiac risk factors, like hypertension [37]. Imatinib, an inhibitor of the Bcr-Abl protein, is used in the treatment of chronic myelogenous leukemia and has been associated with HF as well [38]. It is hypothesized to induce cardiotoxicity by the activation of the endoplasmic reticulum stress response pathways. Electron microscopy of cardiac biopsies demonstrates membrane whorls, pleomorphic mitochondria, effaced cristae, glycogen accumulation, lipid droplets, and vacuoles [38]. However, data is limited to animal studies and a few small case series.

Traditionally, the detection of cardiotoxicity has relied on detection of reduction in LVEF. Multiple-gated acquisition (MUGA) scan and transthoracic echocardiography are the two most commonly used modalities. Baseline LVEF should always be obtained before initiation of chemotherapy. The values obtained from different modalities are not interchangeable and, hence, the same modality should continually be used throughout the chemotherapy protocol for an assessment and comparison of LVEF.

Reduction in LVEF is a late development in the cascade of development of cardiomyopathy and early identification is critical. Several cardiac biomarkers have been studied to identify early cardiac damage. However, utmost caution must be exercised as a false positive result can withhold lifesaving therapy. A good biomarker must be easy to measure, accurate, reproducible, and most importantly, should have high specificity to limit likelihood of false positive results. Biomarkers should be used as an adjunct to the previously described modalities of cardiac assessment [49].

Troponin I is the most studied biomarker as a predictor of development of cardiomyopathy. Elevation of troponin I levels can predict cardiac damage earlier than currently used modalities [50]. One study involving 703 patients receiving anthracyclines showed a 30 % incidence of troponin I elevation and 9 % patients had persistent elevation even at 1 month [51]. Biomarker measurements were done before starting chemotherapy and immediately after. The testing was repeated at 12, 24, 36 and 72 h and again at 1 month. Cardiovascular endpoints were seen in 1 %, 37 % and 84 % patients, respectively, in troponin I negative, transient positive and persistent troponin I positive patients at 1 month. The positive and negative predictive value of troponin I was 84 % and 99 %, respectively. However, this has not been validated in repeated larger trials. A consensus regarding optimal timing of troponin I measurement has also not been reached.