Systemic therapies in breast cancer treatment can be broadly categorized into chemotherapy, targeted therapies, immunotherapies, as well as endocrine (hormone) therapies. Anthracyclines have long been the cornerstone treatment in breast cancer for decades [10]. Discovered in the 1940s, Daunorubicin was isolated from the bacterium Streptomyces and demonstrated to have clinical utility in the treatment of acute leukaemia. Doxorubicin, a precursor, was not long after discovered, with demonstrated utility beyond leukaemia in the treatment of solid organ tumours including breast cancer. Decades later, even with accelerated advancements and development of novel anti-cancer agents, anthracyclines remain one of, if not the most effective agents in the treatment of breast cancer and numerous other malignancies including acute leukaemia, bladder cancer, lung cancer and various others [11, 12]. Though these agents have been used for decades, their exact mechanism of action is not completely understood. One of the most widely accepted theories is that anthracyclines interact with topoisomerase II, an enzyme integral to DNA metabolism [13]. By binding to topoisomerase II, it prevents re-ligation of the double-stranded DNA breaks, thus inhibiting cell growth and promoting apoptosis. Other theories include excess oxidative stress from increased reactive oxygen species (ROS), DNA intercalation inhibiting DNA and RNA synthesis, and blockade of transcription factors that inhibit cell cycle progression.

Taxanes are another class of chemotherapeutic agents commonly used in the treatment of breast cancer. Paclitaxel was the first agent from this class that was approved for treatment of refractory ovarian cancer in the 1990s [14]. Since then, it is now widely utilized in various malignancies including but not limited to breast cancer, non-small cell lung cancer and prostate cancer. They primarily work through suppression of microtubule function, inhibiting a variety of processes including cell signalling, migration and division, eventually inducing cell death [15]. Anthracyclines in combination with taxanes are a common chemotherapy regimen for treatment of high-risk early breast cancer [16].

Cyclophosphamide is an alkylating agent that was derived from research into mustard gas used during World War II [17]. The metabolites of cyclophosphamide create cross-links between DNA strands, inhibiting protein synthesis and resulting in cellular apoptosis [18]. It is used in the treatment primarily of haematological malignancies as well as an immunosuppressive agent in several autoimmune conditions/organ transplants. In breast cancer, it represents an alternative agent for those patients at lower risk of spread, or for whom other agents are contraindicated [19].

Human epidermal growth factor receptor 2 (HER2) is a transmembrane protein belonging to the epidermal growth factor receptor (EGFR) family of tyrosine kinases [20]. It has been demonstrated to be overexpressed in a number of malignancies, including 20% of all breast cancer cases, as well as in a significant proportion of ovarian, gastric, and colorectal cancers, and has been associated with a poorer prognosis. Trastuzumab, a monoclonal antibody and the first HER2 inhibitor that entered clinical practice in the late 1990s, revolutionized treatment of HER2-positive breast cancers as a targeted therapy through binding to the specific receptors and preventing signal transduction, eventually inducing cell death [21]. Newer HER2 inhibitors have been engineered and significantly improved survival rates in patients with HER2-positive breast cancer.

Beyond this, there have also been newer therapies developed to target different molecules on the cancer cells. These include immunotherapy drugs that boost the patient’s immune system to remove cancer cells, such as the programmed cell death protein 1 (PD-1) and programmed cell death ligand 1 (PD-L1) inhibitors that are often used in triple-negative breast cancer [22, 23]. Pembrolizumab, a PD-1 inhibitor, is widely used in early-stage triple-negative breast cancer as well as those with metastatic, PD-L1 positive disease [24]. By binding to PD-1 on immune T cells, it prevents the binding of PD-L1 on cancer cells to PD-1, allowing the T cells to recognise and kill the cancer cells.

Since the late 19th century, there has been a growing understanding of the influence of the ovaries on breast cancer progression [25]. Though initially oophorectomy was performed for premenopausal women with breast cancer, growing research has demonstrated the prevalence of estrogen receptor (ER) and progesterone receptor (PR) in malignant cells. 80% of breast cancers are ER-positive, and of these 65% are also PR-positive; thus, most women with breast cancer are treated with endocrine therapy [26]. Tamoxifen is a selective ER modulator that acts as both an agonist and antagonist on ERs. In breast tissue, it competes with endogenous estrogen for binding and exerts antiestrogenic effects, inhibiting cell cycling of malignant cells. However, in other tissues such as bone, it stimulates estrogenic effects in regulating bone metabolism [27]. In post-menopausal women, aromatase inhibitors have shown superiority to tamoxifen by inhibiting the enzyme aromatase, thereby preventing the conversion of androgens to estrogen in peripheral tissues such as the adrenals, skin and adipose tissue (which are the primary source of estrogen post-menopause) [28].

Radiation treatment represents the only other form of localised therapy along with surgery in the treatment of breast cancer. It aims to damage the DNA in cancer cells (though inadvertently in surrounding healthy cells as well) through ionising radiation that promotes cell death [29]. It can either be given to the entire breast tissue or to part of it in patients at low risk of spread [30]. It is widely used in breast cancer (as well as in several other cancers) and has helped achieve higher cure rates in those without metastatic spread, with over 50% of breast cancer patients now receiving radiotherapy sometime during the course of their treatment [31].

One method of cardiotoxicity surveillance has been through monitoring of cardiac biomarkers before, during, and after treatment [62]. The most commonly used markers are cardiac-specific troponins (cTnT, cTnI) and natriuretic peptides. Cardiac troponins are regulatory proteins found predominantly, or in the case of troponin I exclusively, in the heart that assist in the interaction between actin and myosin to facilitate coordinated contraction [63]. As cardiomyocytes are irreversibly damaged, they release troponins into the systemic circulation, making them a sensitive and specific marker of myocardial injury [64]. They are the gold-standard of diagnosis of myocardial injury in acute coronary syndromes, including myocardial infarction.

Natriuretic peptides are short-chain proteins from the heart that function as hormones that regulate cardiovascular homeostasis through their vasodilatory, natriuretic, and diuretic effects [65]. There are three key types: atrial natriuretic peptide (ANP), b-type (formerly brain) natriuretic peptide (BNP), and c-type natriuretic peptide (CNP), produced by atrial cardiomyocytes, ventricular cardiomyocytes, and endothelial cells, respectively [62]. Among these, BNP is typically preferred as a marker given its reflection of ventricular pressure and load. When ventricular cardiomyocytes are stimulated to produce biologically active BNP from its precursor proBNP, N-terminal pro-B-type natriuretic peptide (NT-proBNP, a biologically inactive peptide) is released into the systemic circulation in equal amounts [66]. NT-proBNP is the preferred marker over BNP given it is a more stable compound, has fewer interactions with cardiac medications, and levels do not fluctuate with the circadian rhythm [62].

Cardiac biomarkers represent a sensitive marker of early cardiotoxicity given the changes in serum concentration in subclinical injury [67]. In addition, there have been studies that have looked at their utility in risk stratification for the development of cancer therapy-related cardiac dysfunction (CTRCD), which can alter an Oncologist’s choice of treatment regimen [68]. In addition to this, changes in troponin and NT-proBNP have been shown to correlate with the degree of LV systolic dysfunction. Beyond surveillance of cardiotoxicity, they also provide an assessment of the degree of cardiac dysfunction [68]. While troponin and NT-proBNP represent sensitive markers of myocardial injury and ventricular stretch/strain, respectively, they carry significant limitations as markers of cardiotoxicity. Both markers lack standardization of measurement assays across centres, which can confound potentially clinically significant changes [69]. Furthermore, though sensitive for any injury or strain to cardiomyocytes, their specificity for CTRCD is limited as numerous medical conditions including ischaemic heart disease, cardiac arrhythmias, pulmonary emboli and sepsis can cause elevations in these markers. While these markers can be elevated in the acute setting during or immediately after treatment, numerous studies have highlighted their poor predictive value in monitoring for late-onset cardiotoxicity, a common manifestation from treatment with anthracyclines or HER2 inhibitors [70].

While alterations in cardiac biomarkers reflect structural myocardial changes in the form of cardiac injury or strain, cardiac imaging provides a new avenue for surveillance of CTRCD as a functional assessment of the heart. Transthoracic echocardiography (TTE) is a readily available ultrasound imaging tool that can provide a comprehensive structural and functional assessment of the chambers of the heart, its valves, and the surrounding vasculature. It provides standardized measurements that are relatively consistent across countries and frequently updated by numerous expert cardiac bodies throughout America, Europe, and Asia [71–73]. It is the most commonly used modality for monitoring for cancer therapy-related cardiotoxicity, with most major bodies defining CTRCD as a reduction in LVEF of > 10% from baseline, to a value less than 53% [74]. LVEF is a critical marker of overall cardiac health and one of the best predictors of cardiovascular outcomes across all patient groups [75]. However, it has limited sensitivity, with reductions in LVEF often a late manifestation of cardiotoxicity and may reflect a missed opportunity for early intervention [76]. Further limiting its sensitivity, reductions in LVEF where it remains > 45% are not associated with increased mortality [77]. Other limitations include its significant inter-observer variability between operators and its lack of assessment of left ventricular diastolic dysfunction, which can precede systolic dysfunction in breast cancer patients treated with cardiotoxic chemotherapy [78, 79].

Measurement of strain by speckle tracking echocardiography is a relatively new technique that provides a new method of assessment of left ventricular function. While LVEF relies on changes in LV volumes in systole and diastole, myocardial strain measures systolic function by changes in tissue deformation across multiple segments of the LV, averaged to obtain a single result often denoted as a negative value (representing myocardial fibre shortening in systole) [80]. LV global longitudinal strain (GLS) reflects the longitudinal contractility of the LV myocardial fibres and has been shown to have better intra- and inter-observer variability compared to LVEF [81]. It is a more sensitive marker of subclinical systolic dysfunction, with reduction in LV GLS noted prior to overt changes in LVEF, and allows risk stratification in those with LVEF > 45% [77]. It has also been demonstrated as a superior predictor of cardiovascular outcomes and all-cause mortality compared to LVEF [81]. In cancer patients treated with cardiotoxic chemotherapy, reduced baseline LV GLS demonstrated significant predictive value for the development of cardiotoxicity during treatment and demonstrated utility in detection of subclinical LV systolic dysfunction on follow-up TTEs [80]. While some studies have shown benefits of early intervention with cardioprotective therapy in those patients with ‘subclinical’ cardiac dysfunction, further research is required to validate these results [82]. However, this is now a class I recommendation by the European Society of Cardiology (ESC) to be performed at baseline and follow-up of patients at risk of cancer therapy-related cardiotoxicity [83]. Given the risk of late-onset cardiotoxicity, particularly with anthracyclines, some studies recommend longer 5- and 10-year follow-up TTE in patients at higher risk, including those who received higher cumulative doses of anthracyclines, those over 60 years of age at the time of treatment, and those with baseline cardiovascular risk factors [84]. The ESC has now incorporated LV GLS as a marker of CTRCD, with a reduction in LV GLS > 15% from baseline even in the presence of normal LVEF representative of mild CTRCD. However, LV GLS is load-dependent and thus its accuracy is limited in haemodynamic overload [85]. Furthermore, measuring LV GLS requires acquiring images at high frame rates, requires an operator experienced with strain imaging, and has limited accuracy with poor image quality [86]. Similar to LVEF, it does not reflect changes in diastolic function; with HFpEF, the more prevalent phenotype at long-term follow-up in breast cancer patients compared to HFrEF [87].

There are several novel echocardiographic parameters that are being explored as potential markers of early cardiotoxicity. LV global circumferential strain (GCS) and right ventricular free wall strain (RV FWS) measure LV torsional contraction and RV longitudinal contraction, respectively, both using speckle tracking echocardiography. They have been shown in studies to identify a greater number of cancer patients with subclinical cardiac dysfunction compared to traditional parameters [88]. Left atrial phasic strain (particularly reservoir and conduit strain) has also been investigated as markers of early or ‘subclinical’ diastolic dysfunction superior to traditional Doppler-derived parameters [89]. Reductions in left atrial strain have been identified in cancer patients treated with cardiotoxic therapies prior to changes in LVEF or symptoms, and have also been identified in a greater number of patients compared to those with significant reduction in LV GLS, consistent with prior studies highlighting diastolic dysfunction as an earlier marker of cardiotoxicity [32, 89]. Myocardial work, an echocardiographic measure derived from LV GLS and brachial systemic blood pressure, has also been demonstrated as a potential marker of early cardiotoxicity in cancer patients [90]. Studies have highlighted its sensitivity, potentially more so than LV GLS, and also its lack of influence from loading conditions. While most patients are typically monitored using TTE to measure LVEF and LV GLS, cardiac biomarkers are often utilised as an adjunct given their sensitivity to myocardial injury in high-risk patients with pre-existing cardiovascular disease [67].

Cardiac magnetic resonance imaging (CMRI) has emerged as a potential surveillance modality for cardiotoxicity. It provides the most accurate assessment of chamber structure and function, including strain, and can more reliably identify marginal changes in left ventricular function compared to traditional TTE. Beyond TTE, it can also better delineate between and characterise endocardial, myocardial, epicardial and pericardial tissue, identifying areas of oedema, inflammation or fibrosis to better understand the pathophysiology of cardiotoxicity [91]. However, CMRI is quite costly, time-consuming, and much less readily accessible when compared to TTE [92]. In addition, many patients are precluded from the scan due to the presence of metallic foreign bodies or implants. For these reasons, its use in clinical practice for surveillance of cardiotoxicity in cancer patients has been limited.

Surveillance for treatment-related cardiotoxicity in breast cancer patients is crucial in reducing morbidity and mortality. Through early recognition of ‘subclinical’ myocardial injury using more sensitive biomarkers and imaging tools, early cessation of cardiotoxic treatment, longer surveillance in select patients at higher risk of late-onset cardiotoxicity, and prompt initiation of cardioprotective medications can prevent irreversible myocardial injury and reduce mortality [33].