Cardiovascular Events and Mortality in Patients Undergoing Adjuvant Radiotherapy for Breast Cancer: a Systematic Review

Hong Kong J Radiol 2021 Sep;24(3):148-66

 

 

¹ Odette Cancer Centre, Sunnybrook Health Sciences Centre, University of Toronto, Toronto, Ontario, Canada

² Department of Oncology, Princess Margaret Hospital, Hong Kong

³ Centro Oncológico AZ do Noroeste, Patos de Minas, Minas Gerais, Brazil

 

  Full Article in PDF

Breast cancer is the most frequently diagnosed cancer and the leading cause of cancer death in females worldwide.[1] It has been shown through randomised trials that adjuvant radiotherapy (RT) following breast-conserving surgery substantially reduces breast cancer recurrence and reduces the absolute breast cancer mortality rate.[2] [3] RT administered to breast cancer patients usually exposes the heart, an organ at risk, to some radiation. Cheng et al[4] conducted a literature review and meta-analysis on this topic, including studies published prior to January 2015, and found that breast cancer RT was associated with an absolute increase of 76.4 cases of coronary heart disease (95% confidence interval [CI]=36.8-130.5) and 125.5 cases of cardiac death (95% CI=98.8-157.9) per 100 000 person-years, respectively. In order to create optimised and tailored treatment plans, the current relationship between adjuvant breast RT and cardiovascular risks must be studied so that physicians and patients may appropriately consider the benefits of reduced breast cancer mortality with the potential long-term cardiovascular risks. We performed a systematic review to assess the risk of cardiovascular events (CVEs) and cardiovascular mortality (CVM) and its correlation with breast/chest wall RT for women with breast cancer (including breast cancer and ductal carcinoma in situ) following breast-conserving surgery/mastectomy (for node-positive or involved resection margin disease), as well as disease laterality. This will allow radiation oncologists to better inform their patients about the risks and benefits of adjuvant RT so that patients may make a more informed decision.

 

A literature search was completed using MEDLINE, Embase, and Cochrane Central Register of Controlled Trials from inception through to July 2020. Search terms for breast cancer included ‘breast cancer’, ‘breast neoplasm’, ‘breast carcinoma’, and ‘breast tumour or tumour’ (online supplementary Appendix). RT terms included ‘radiotherapy’, ‘radiation’, ‘irradiation’, and ‘radiation injury’. CVE terms included ‘heart disease’, ‘heart infarction’, ‘myocardial infarction (MI) or heart attack’, ‘angina pectoris’, ‘congestive heart failure (CHF)’, ‘coronary artery disease (CAD)’, ‘coronary artery obstruction’, ‘heart or cardio or cardiovascular disease (CVD)’, ‘ischemic heart disease (IHD)’, ‘dosage risk’, ‘cardiovascular risk’, and ‘cardiovascular death’.

 

Screening was first done based on the title and abstract independently by two authors (P Taylor, S Chan), with discrepancies being resolved through discussion between the two authors. Then, full-text screening was conducted independently by the two authors. Inclusion criteria were reports of the clinical cardiovascular outcomes, including CVEs and/or CVM as defined above. Specifically, studies were included if they reported comparisons in clinical cardiovascular outcomes between patients who received RT and those who did not receive RT and/or between patients who received left-sided RT and those who received right-sided RT. Exclusion criteria included any studies that investigated cancers other than breast cancer and the effects of irradiation on organ systems other than the cardiovascular system. Studies employing brachytherapy, partial breast irradiation, or boost to the tumour bed alone were excluded. Full-length papers, including cohort studies, case-control studies, and randomised controlled trials published as original papers written in English, were considered. Any case reports and non-original articles such as systematic reviews were excluded.

 

Data extraction was conducted independently by two authors (P Taylor, S Chan). Both authors engaged in a discussion regarding any discrepancies between the extracted data and came to a consensus. The following data were extracted from the papers: publication year, geographical location, sample size, mean/median follow-up, number of CVEs, number of cardiovascular deaths, laterality, hazard ratios (HRs), incidence ratios, risk ratios (RRs), mortality ratios (MRs), and associated measures of variance for all categories of outcomes.

 

The search identified 7363 publications, of which 1063 were duplicates and excluded (Figure). A further 6132 articles were excluded because they did not meet the inclusion criteria. The remaining 168 articles underwent full-text screening. Of them, 92 were excluded for failure to meet the inclusion criteria, leaving 76 studies that were analysed in this systematic review.

 

Figure. PRISMA flow diagram.

 

Of the 76 studies, 35 investigated the risk of CVM with respect to RT laterality,[5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] 28 studies investigated the risk of CVEs with respect to RT laterality,[7] [10] [11] [12][16] [18] [20] [22] [24] [33] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] 26 studies investigated the risk of CVM with respect to RT compared with no RT,[5] [7] [10] [14] [21] [27] [29] [30] [32] [38] [54] [58] [59] [60] [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71] [72] and 22 studies investigated the risk of CVEs with respect to RT compared with no RT.[7] [36] [44] [45] [48] [49] [54] [59] [61] [62] [63] [67] [68] [69] [73] [74] [75] [76] [77] [78] [79] [80] Several studies overlapped between the categories and investigated the risk of CVEs and/or CVM with respect to RT and/or RT laterality. Results from the largest studies, based on study population size, will be highlighted for each of the four categories. The results for all studies are also reported in Tables 1 2 3 4.

 

Of 35 studies investigating the risk of CVM with respect to RT laterality, seven (20%) found a significant increased risk of CVM in patients who received left-sided RT compared with patients who received right-sided RT, all of which included study periods that started prior to 1985 (Table 1).[8] [10] [21] [26] [28] [30] [39] An additional six studies found a significant association between laterality and CVM only in subgroup analysis.[5] [11] [23] [32] [34] [36] Of these six studies, three had subgroup analyses based on time period stratification, where in general older study periods before 1980-1990 were significant while more recent time periods were not (Table 1).[5] [23] [34]

 

Table 1. Studies assessing the risk of cardiovascular mortality in patients who received left-sided RT compared with those who received right-sided RT.

 

Of 28 studies investigating the risk of CVEs with respect to RT laterality, eight (29%) found a significant increased risk of CVEs in patients who received left-sided RT compared with patients who received right-sided RT (Table 2).[11] [18] [40] [41] [46] [49] [51] [56] Of these eight studies, five included study periods that started prior to 1985.[11] [18] [40] [41] [46] An additional three studies found a significant association between laterality and CVEs only in subgroup analysis.[22] [47] [55] These subgroup analyses were based on different treatment types and differences in types of CVEs.

 

Table 2. Studies assessing the risk of cardiovascular events in patients who received left-sided RT compared with those who received right-sided RT.

 

Of 26 studies investigating the risk of CVM for patients that received RT compared with those who did not receive RT, seven (27%) found a significant increase in CVM (Table 3).[29] [30] [32] [63] [64] [71] [72] Of these seven studies, six included study periods that started prior to 1985. An additional five studies found a significant association between RT and CVM only in subgroup analysis.[5] [10] [21] [68] [70] These subgroup analyses were based on different treatment types and differences in specific causes of CVM, such as death from cardiac diseases compared with death from vascular diseases.

 

Table 3. Studies assessing the risk of cardiovascular mortality in patients who received RT compared with those who did not receive RT.

 

Of 22 studies investigating the risk of CVEs for patients that received RT compared with those who did not receive RT, five (23%) found a significant association between RT and CVEs (Table 4).[36] [73] [75] [79] [80] An additional three studies found a significant association between RT and CVEs only in subgroup analysis.[74] [77] [78] These subgroup analyses were based on different treatment types and the presence of pre-existing cardiovascular risk factors.

 

Table 4. Studies assessing the risk of cardiovascular events in patients who received RT compared with those who did not receive RT.

 

This review summarises the cardiovascular morbidity and mortality risk associated with breast adjuvant RT and the laterality of the RT. When comparing patients who received left-sided RT with those who received right-sided RT, we found that 7 of 35 studies found a significant increase in the risk of CVM and 8 of 28 studies found a significantly increased risk of CVEs. For patients who received RT compared with those who did not receive RT, 7 of 26 studies found a significantly increased risk of CVM and 5 of 22 studies found a significant increased risk of CVEs.

 

A previous meta-analysis conducted by Cheng et al[4] examined studies of breast cancer patients from 1966 to 2015. The authors[4] found that patients who received RT had an increased risk of coronary heart disease (RR=1.30, 95% CI=1.13-1.49) and cardiac mortality (RR=1.38, 95% CI=1.18-1.62) compared with patients who did not receive RT. They also found that patients who received left-sided RT experienced an increased risk of developing coronary heart disease compared with patients receiving right-sided RT (RR=1.29, 95% CI=1.13-1.48).[4] Patients receiving left-sided RT also experienced an increased risk of cardiac death compared with patients receiving right-sided RT (RR=1.22, 95% CI=1.08-1.37).[4] In contrast to Cheng et al,[4] in the present review, we found newer studies in which there was no significant increased risk of CVEs or CVM in patients who received RT compared with patients who did not receive RT. We also found more studies in which patients who received left-sided RT had no significant increased risk of CVEs or CVM compared with patients who received right-sided RT. These differences likely reflect the fact that the present review includes many new studies since 2015 in which the study populations received modern RT techniques. However, because we did not conduct a meta-analysis in the present review, it remains unclear whether our findings represent a significantly different association between breast cancer RT and cardiovascular risk compared with that reported by Cheng et al.[4]

 

A systematic review conducted by Drost et al[81] found that the mean heart dose steadily decreased from 4.6 Gy in 2014 to 2.6 Gy in 2017 (p = 0.003). Combining this with the dose-dependent relationship between major cardiac events and mean dose to the heart by Darby et al,[40] it is likely that the decrease in mean heart dose owing to improved contemporary RT techniques has led to improved outcomes in breast cancer patients in recent years. Our findings are also in support of this hypothesis since all studies that found a significant association between RT laterality and CVM included treatment groups that started prior to 1985. This is consistent with the systematic review by Cheng et al,[4] which found an increased risk of cardiovascular death and coronary heart disease associated with RT among studies in which the breast cancer patients were diagnosed and irradiated before 1980 (RR=1.45, 95% CI=1.14-1.89) compared with women diagnosed and irradiated after 1980 (RR=1.15, 95% CI=0.92-1.44; p = 0.04). Similarly, Giordano et al[23] found that in 1979, the HR for ischaemic heart disease mortality in left-sided compared with right-sided disease was 1.50 (95% CI=1.19-1.87), but this HR declined by 6% with each succeeding year between 1979 and 1988 (HR=0.94; 95% CI=0.91-0.98).

 

As such, newer research has started to evaluate whether the use of modern linear accelerator machines instead of ⁶⁰Co fields and various contemporary radiation techniques reduces the cardiac radiation dose and subsequent cardiac toxicities. One example is intensity-modulated RT, which allows a more conformal target coverage without exposing organs at risk to as much radiation.[82] Other notable advancements in cardiac sparing techniques include the use of deep inspiration breath hold, enhanced patient positioning, and heart blocking.[83] Deep inspiration breath hold and respiratory gating rely on the principle that during inspiration, the diaphragm flattens, and the lungs expand, causing the heart to be pulled away from the chest wall and thus decrease the radiation dose to the heart and the left anterior descending artery.[84] A 2019 study conducted by Simonetto et al[85] found that the use of deep inspiration breath hold reduced the risk of estimated mean heart dose by 35%, compared with free breathing. Furthermore, to reduce the heart dose in breast cancer patients, prone positioning can be used to increase the planning target volume to heart distance by displacing cardiac structures and substructures out of irradiated volumes.[86] [87] [88] Other common methods include multileaf collimator modification during RT planning.[89] [90] However, an important pitfall is that it may shield part of the breast tissue, which needs to be irradiated; thus, a balance must be achieved in order to maximise the heart shielding while minimising the target volume missed.[89] In addition to these RT techniques, the omission of internal mammary chain lymph node irradiation and rib inclusion for chest wall RT has been utilised in early-stage breast cancer patients to reduce the dose to the normal tissue. However, long-term studies are needed to investigate the effect of contemporary RT planning techniques in minimising radiation exposure to nearby normal tissue and the heart.

 

In addition to the use of modern RT techniques, considerations must be made in terms of whether RT is being combined with chemotherapeutic agents as common chemotherapeutic agents have known cardiotoxic effects.[91] [92] [93] [94] A 10-year cohort study of breast cancer patients receiving concomitant RT and chemotherapy found that there was no significant association between CVEs and RT laterality (HR=2.38, 95% CI=0.80-7.11; p = 0.12).[55] However, there was a significant increase in CVEs for patients receiving left-sided RT with a doxorubicin-equivalent dose ≥250 mg/m² compared with patients receiving right-sided RT with a cumulative doxorubicin-equivalent dose <250 mg/m² (HR=5.22, 95% CI=1.67-21.15; p = 0.006).[55] Similarly, a 2016 study found that there was no significant increase in the risk of CVEs for patients who received RT compared with those who received surgery alone (HR=0.97, 95% CI=0.62-1.51; p = 0.882).[77] However, there was a significant increase in CVEs in patients who received RT and chemotherapy compared with those who received surgery only (HR=1.84; 95% CI=1.34-2.53; p < 0.001).[77]

 

There are several limitations to this systematic review. First, the heterogeneity of the data made it difficult to make direct comparisons among studies. Many of the studies did not report data on the individual types of CVEs or the specific cause of CVM, and so we used a composite outcome of CVE and CVM, which included myocardial infarction, coronary artery disease, conduction abnormalities, congestive heart failure, and other cardiovascular diseases. Heterogeneity also exists because of the lack of detail on RT techniques, RT volume, dose, and fractionation. In addition, variability in morbidity and mortality assessment, as well as in the follow-up time of the studies, is another source of heterogeneity. Second, the dose-dependent relationship of cardiovascular risk cannot be evaluated since radiation doses were not available for all studies. Lastly, because this was a systematic review, and a meta-analysis was not conducted, studies were not weighted based on the number of patients. In the future, a meta-analysis would aid in determining whether there is a significant increase in the risk of CVEs and CVM associated with the use of RT and/or RT laterality. Other confounding variables may also be investigated and stratified, such as menopausal status, types of adjuvant chemotherapy and hormonal agents, which may impact cardiotoxicity.

 

Although modern RT techniques seem to have minimised the cardiac exposure in breast cancer patients receiving RT, more comprehensive studies with longer follow-up periods must be conducted to investigate any associated cardiovascular risk.