Due to recent advances in BC treatment, survival rates have been increasing, leading to a growing population of BC long-term survivors [21]. Currently among BC survivors cardiovascular disease is one of the leading cause of morbidity and mortality [21]. These cardiotoxicity effects are becoming increasingly recognized and cardio-oncology is emerging as a multidisciplinary field to advance our ability to prevent, diagnose, and treat these potentially life-threatening side effects [22].
In our study, women with BC showed an increased risk for cardiovascular events, in particular for MI compared to healthy women, which is aligned with previous studies [3, 14, 22, 23]. Potential reasons for this increased risk could be due to different factors directly or indirectly correlated to BC [24]. The extent to which cardiovascular risk may be involved depends on patient related factors (biological mechanisms, modifiable risk factors, and potential genetic predispositions) and/or specific treatments [21].
The main factor to determine this increased risk of cardiovascular events has been hypothesized to result from the effects of the anti-cancer treatments. They may have adverse effects on myocardial cells causing necrosis, apoptosis, oxidative stress, damage to endothelial cells, and inflammation [11, 14]. In different clinical settings, it has been shown that many forms of chemotherapy and radiation are associated with increased risks of developing cardiovascular complications [9–12, 25]. Cardiovascular complications associated with chemotherapy and radiation therapy can include hypertension, arrhythmias, coronary artery disease, heart failure, valvular disease, thromboembolic disorders, peripheral vascular disease, stroke, pulmonary hypertension, and pericardial complication [10, 26]. Even though the effects of chemotherapy and radiation therapy have been investigated for years, including in the setting of BC, there is still a lack of data regarding the cardiovascular side effects of these treatments and their contribution to cardiac toxicity [3, 14, 15, 23].
In trying to investigate the cardiotoxicity impact of the different treatments for BC, we analysed the various contributions of chemotherapy and radiation on the risk of MI.
It was observed that both chemotherapy and radiotherapy increase the risk of MI but with a different weight. It is very important to consider the different impacts of these therapies for future clinical considerations.
Our study analysed a large number of subjects and we found that chemotherapy almost doubled the risk for MI, and the effect was even more evident when we considered the effect of chemotherapy as a single treatment.
These data confirms the well-established evidence found in several tumours that different chemotherapy agents may induce cardiotoxicity through a wide variety of mechanisms [27–29]. The myocardium consists of cells that have limited regenerative capability, which may render the heart susceptible to effects from chemotherapeutic agents [11]. Such myocardial toxicity and the consequent hypercoagulable states and inflammation, encompasses a heterogeneous group of disorders ranging from relatively benign arrhythmias to potentially lethal conditions such as myocardial ischemia/infarction and cardiomyopathy [11, 29]. Studies on BC confirm the higher risk of cardiovascular disease among chemotherapy treated patients [13, 14, 29–31]. In BC treatment, the traditional third generation polychemotherapy regimens are the most widely used treatment and consist of anthracycline and taxanes administered sequentially [3]. Anthracyclines such as daunorubicin and mitoxantone are the most well-studied and generate dose dependent cardiomyopathy and congestive heart failure [29, 32–34]. The mechanisms of anthracycline-induced cardiotoxicity remain unclear but may involve direct pathways for reactive oxygen species (ROS) generation and topoisomerase 2 and other indirect pathways [28, 35]. Additionally, taxanes such as paclitaxel and docetaxel can primarily cause arrhythmias, bradycardia and myocardial ischemia [36, 37]. Moreover, a number of studies on the risk of cardiovascular disease in BC patients confirmed the cardiotoxic effects of trastuzumab [38–41]. In addition, the combination of anthraciclines and transtuzumab appears to increase the incidence of cardiac side effects, which mediated cardiac failure in a direct dose-dependent manner [40, 42].
Similar to chemotherapy, radiotherapy is also associated with a variety of cardiovascular complications involving the pericardium, myocardium, valves, coronary arteries and conduction systems [10]. Radiotherapy irradiation of the heart has been shown to be associated with long-term cardiac toxicity such as heart failure, coronary artery disease, myocardial infarction, and cardiovascular death [33, 43]. At a cellular level radiotherapy effects seem to induce cardiac injuries mediated by reactive oxygen species and myocardial fibrosis [23]. The risk of cardiovascular damage due to radiotherapy seems to be higher especially in left-sided cancers that generally receive a higher dose of radiation to the heart than those with right-side irradiation, as well as in women with pre-existing cardiac risk factors [8, 33, 43]. Among the past epidemiological studies there are a number of different and contrasting results on the contribution of radiotherapy to cardiovascular events [43]. The differences between some epidemiological studies could be probably due to the variety of radiation therapy regiments in the history of BC: indeed, the range of doses to the heart has changed over the past few decades.
Our results showed that radiotherapy appears to have less impact on MI risk than chemotherapy. The increase in cardiovascular risk found after radiotherapy in BC patients is not statistically significant both in comparison with healthy subjects and in patients with cancer but not treated with therapies.
Our evidence is aligned with the findings that modern techniques of radiotherapy, radiotherapy, in particular 3D conformal volume-based radiation therapy, which relies on computed tomography treatment volume definition and subsequent radiotherapy planning, is able to selectively irradiate target volumes, concurrently sparing organs at risk, thus increased the therapeutic window for radiotherapy [44]. A recent analysis by the Danish Brest Cancer Group reported on a 10-year cumulative risk of cardiac event of 2 − 1% (1.8–2.4) for left-sided breast cancer patients, irradiated with computed- tomography based radiotherapy. The incidence rate ratio for cardiac events in left-sided vs right- sided patients was 0.9 (0.69–1.16), with no trend towards worse outcome within the first 10 years [45]. In the last 20 years, new possibilities emerged in breast cancer RT allowing for targeted solutions and personalized approaches. The increased adoption of hypofractionation, the selective use of the boost to the lumpectomy cavity, the reduction in treatment volume with partial breast irradiation, the introduction of volume-based target volume definition and selection, and the integration with primary systemic therapy strategies, all enlarged the therapeutic portfolio of the radiation oncologist to decrease the treatment-related toxicity profile and to reduce the harm to the cardiovascular system [46].
Since the 1980s, there has been a significant decrease in radiation dose to the heart, decreasing the areas of fibrosis and edema after radiotherapy treatments [47].
Our research showed that the cardiotoxic effects of chemotherapy and radiation are distinct from any potential cardiotoxic effects of adjuvant treatments. Indeed our results showed that different hormonal therapy (both tamoxifen and aromatase inhibitors) do not contribute on the risk of MI. Recently, a number of studies considered the effects of adjuvant therapies, in particular of tamoxifen and aromatase inhibitors on cardiovascular risk, but the existing studies are still conflicting, hence supplementary research is needed [48, 49].
Finally, our study wanted to investigate the risk of stroke after treatment for BC. While there are a number of studies evaluating the risk of MI, evidence of an association between treatment for BC and stroke is limited [29, 50]. Prior studies have found increased risk of stroke in patients with BC who were given radiotherapy [50]. More specifically it has been found that radiation to the supraclavicular lymphnodes gives a significant dose of radiation to the proximal carotid artery, which increases the risk of carotid stenosis and ischaemic stroke [50].
Our data indicated that women with BC showed an increased risk for stroke compared to healthy women, as previously reported for MI. While it is very clear the effect of therapies (in particular of chemotherapy) on MI, the risk for stroke seems to be independent from specific treatments, including both chemotherapy and radiotherapy. Indeed, the higher risk for stroke observed comparing BC (any therapy) with healthy women disappeared when comparing therapies among BC patients. The results are the same considering both haemorrhagic and ischemic stroke. This data validates other studies that not detect any association between stroke risk and specific chemotherapy regimens [29]. It was known that MI and stroke share a lot of common risk factors but our intriguing results suggest that in the pathogenesis of stroke other different factors could be involved indirectly as consequences of tumour effects [24]. Further studies could clarify if genetic susceptibility, as well as other factors such as stress, smoking, and hypertension, could determine a higher risk of stroke independent from specific therapies [29].
Even with a number of intriguing findings, the present study has several potential limitations. The main limitation of the study is that we cannot identify patients with tumours on the left side and on the right side, as cardiac radiation doses are higher on the left side in BC patients with left-sided tumours. Moreover, it was not possible to obtain information on the type and the dose of chemotherapy used. Another important limitation is that there isn’t adjustment for some important factors such as BMI and alcohol consumption, both associated with BC and MI and therefore possible confounders. However, we estimated the amount of residual confounding using the EPIC-Turin study and it seemed to be small, if present. Finally, the short follow-up period of this study, from 2011 to 2017, is a significant limitation because the late effect of therapies could need more time to show its damage. Nevertheless, this study is one of the largest studies on this topic, based on the entire population of Piedmont region (more than 4 million inhabitants), where nearly 20,000 women with BC have been found in the follow-up period.