Author response: Risk of heart disease following treatment for breast cancer – results from a population-based cohort study
Year: 2022
Type: peer-review
Abstract: Article Figures and data Abstract Editor's evaluation Introduction Materials and methods Results Discussion Appendix 1 Appendix 2 Appendix 3 Data availability References Decision letter Author response Article and author information Metrics Abstract Background: There is a rising concern about treatment-associated cardiotoxicities in breast cancer patients. This study aimed to determine the time- and treatment-specific incidence of arrhythmia, heart failure, and ischemic heart disease in women diagnosed with breast cancer. Methods: A register-based matched cohort study was conducted including 8015 breast cancer patients diagnosed from 2001 to 2008 in the Stockholm-Gotland region and followed up until 2017. Time-dependent risks of arrhythmia, heart failure, and ischemic heart disease in breast cancer patients were assessed using flexible parametric models as compared to matched controls from general population. Treatment-specific effects were estimated in breast cancer patients using Cox model. Results: Time-dependent analyses revealed long-term increased risks of arrhythmia and heart failure following breast cancer diagnosis. Hazard ratios (HRs) within the first year of diagnosis were 2.14 (95% CI = 1.63–2.81) for arrhythmia and 2.71 (95% CI = 1.70–4.33) for heart failure. HR more than 10 years following diagnosis was 1.42 (95% CI = 1.21–1.67) for arrhythmia and 1.28 (95% CI = 1.03–1.59) for heart failure. The risk for ischemic heart disease was significantly increased only during the first year after diagnosis (HR = 1.45, 95% CI = 1.03–2.04). Trastuzumab and anthracyclines were associated with increased risk of heart failure. Aromatase inhibitors, but not tamoxifen, were associated with risk of ischemic heart disease. No increased risk of heart disease was identified following locoregional radiotherapy. Conclusions: Administration of systemic adjuvant therapies appears to be associated with increased risks of heart disease. The risk estimates observed in this study may aid adjuvant therapy decision-making and patient counseling in oncology practices. Funding: This work was supported by the Swedish Research Council (grant no: 2018-02547); Swedish Cancer Society (grant no: CAN-19-0266); and FORTE (grant no: 2016-00081). Editor's evaluation We feel that your work will be of interest to breast medical oncologists, cardiologists, and primary care providers who treat patients with breast cancer. We commend you for this study, which achieves its goal of identifying the incidence and hazard ratio of cardio-toxicity associated with breast cancer treatment within a general breast cancer population. The international nature of your collaborative study along with its large patient cohort size and long horizontal follow up are quite attractive features in solidifying previous findings and discovering future areas of exploration. https://doi.org/10.7554/eLife.71562.sa0 Decision letter Reviews on Sciety eLife's review process Introduction The use of adjuvant systemic therapies at least halves the risk of dying from breast cancer (Early Breast Cancer Trialists’ Collaborative Group (EBCTCG), 2005; Dowsett et al., 2015; Goldvaser et al., 2019; Gray et al., 2019). Nowadays, 80% of the breast cancer patients survive for at least 10 years and many will become long-term survivors. There are, however, concerns about therapy-associated late adverse health effects, including cardiovascular events (Khouri et al., 2012). Use of common (neo-)adjuvant therapies for breast cancer has been associated with an increased risk of heart diseases including heart failure, arrhythmias, and ischemic heart disease (Darby et al., 2013; Doyle et al., 2005; Harris et al., 2006; Hooning et al., 2007; Pinder et al., 2007; Taylor et al., 2017; Yeh and Bickford, 2009). However, this evidence mostly comes from studies focusing on specific subgroups of patients based on age, cancer stage, or treatment regimen. Although the benefits of radiotherapy far outweigh the risk of heart diseases, some studies have found increased incidence of and mortality due to heart disease in women subjected to radiotherapy techniques (Darby et al., 2013; Taylor et al., 2017). Knowledge on cardiotoxic effects of anthracycline-based chemotherapy regimens have led to lowering of doses and less use of bolus injections to reduce peak concentrations of anthracyclines (Foukakis et al., 2016), but it is estimated that the risk of heart failure associated with anthracyclines remains increased for standard low-dose group as compared to non-users (Chung et al., 2020). While trastuzumab has been shown to reduce the risk of breast cancer mortality at 11 years of follow-up (Cameron et al., 2017), evidence on its cardiotoxicity is conflicting (Bowles et al., 2012; Papakonstantinou et al., 2020). In addition, recent evidence suggests that use of aromatase inhibitors in women with hormone receptor positive breast cancer may increase the risk of heart failure, compared to tamoxifen (Khosrow-Khavar et al., 2020). Risk assessment of immediate and later occurring heart disease events following breast cancer is important for the planning of cardiac surveillance programs and possible prophylactic pharmacotherapy. Here, we report the risks of heart diseases in a cohort representative of the general breast cancer population with long-term follow-up. We specifically aimed to assess risks of heart diseases by time since diagnosis, and according to adjuvant treatments. Materials and methods Breast cancer cohort Request a detailed protocol This study took advantage of the Stockholm-Gotland Breast Cancer Register comprising all women diagnosed with primary invasive breast cancer between 2001 and 2008 in the Stockholm-Gotland region. The Stockholm-Gotland Breast Cancer Register has about 99% completeness and provides detailed information on tumor and treatment characteristics, as well as routine follow-up on locoregional recurrences and distant metastases (Colzani et al., 2011; Wigertz et al., 2012). Detailed description of the breast cancer cohort can be found elsewhere (Holm et al., 2016). We included all patients diagnosed with non-metastatic breast cancer (stages I–III) and without prior diagnosis of heart disease at age 25–75 years (N = 8015). To compare the risk of heart diseases after breast cancer diagnosis, we randomly sampled up to 10 women from the general female population in Stockholm-Gotland region matched on year of birth (Appendix 1—figure 1). Each reference individual was alive and free of breast cancer on the date of the matched patient’s diagnosis of breast cancer (the index date). The matched cohort was linked through the unique personal identity number to the Swedish Cancer Register, Patient Register, Cause of Death Register and Migration Register, and follow-up started from the index date until the date of heart disease diagnosis, emigration, death, breast cancer relapse, or end of follow-up (December 31, 2017), whichever occurred first. We also performed linkage with the Prescribed Drug Register which contains data on all drugs dispensed from Swedish pharmacies from July 2005 onward to validate the use of hormone therapy. The study was approved by the Regional Ethical Review Board in Stockholm. Heart diseases Request a detailed protocol We identified the following heart diseases according to relevant ICD (International Classification of Disease) codes in the Swedish Patient Register and the Cause of Death Register (Appendix 2—table 1): heart failure (ICD-10: I50, ICD-9: 428A, 428B, 428X), arrhythmias (ICD-10: I47–I49, ICD-9: 427), and ischemic heart disease (ICD-10: I20–I25, ICD-9: 410–414). We included both inpatient and outpatient diagnoses, as well as cause of death records in our outcome definition, except for myocardial infarction, which was based on inpatient and cause of death records solely. To ensure specificity of the outcomes studied, only primary, and not underlying, diagnoses were considered for analyses. Breast cancer treatment specifics Request a detailed protocol We extracted the treatment data from the Stockholm-Gotland Breast Cancer Register. As ~90% of HER-2 positive cancers were treated with trastuzumab between 2005 and 2008 in the Stockholm-Gotland region and the Swedish Prescribed Drug Register does not cover data on treatment with trastuzumab, HER-2 positivity was used as a proxy when no registry data on trastuzumab was available during this time period (30% of the HER-2 positive patients had missing information on trastuzumab). Data on adjuvant endocrine therapy was verified against the Prescribed Drug Register and categorized into tamoxifen and/or aromatase inhibitor use. Since radiotherapy to the left breast has, in particular, been implicated in heart complications, radiotherapy was categorized according to tumor laterality (left vs. right). Bilateral tumors were coded separately in this analysis. Chemotherapy was coded as anthracycline-based, anthracycline plus taxane-based, cyclophosphamide, methotrexate, and fluorouracil (CMF)-based regimens. Covariates Request a detailed protocol The Stockholm Breast Cancer Register contains data on date of diagnosis, menopausal status at diagnosis, and type of surgery (breast conserving surgery vs. mastectomy). Tumor characteristics were also retrieved from this register, including tumor size (T), regional lymph node involvement (N), and presence of metastases (M), all from pathology records and summarized in TNM stage as defined according to the American Joint Committee on Cancer, 2010. Information on inpatient comorbidities at diagnosis was extracted through the Swedish Patient Register and summarized into the Charlson comorbidity index (CCI) score, a widely used method for classifying chronic comorbid conditions (Charlson et al., 1987). To account for the potential confounding effect from tobacco abuse, chronic pulmonary disease, and hypertension in the associations, we further identified associated diagnoses before cancer using ICD codes from the patient register (Appendix 2—table 1). Statistical analyses Request a detailed protocol We compared the risk of heart diseases in breast cancer patients with that observed in the matched cohort, using flexible parametric model (FPM) with time since index date as underlying time scale. The FPM is similar to the Cox proportional hazards model in that it provides a hazard ratio (HR) as measure of association. In our analysis, a restricted cubic spline with four internal and two boundary knots (five degrees of freedom) placed at quintiles of the event times was used in the FPM for the baseline hazard. The key advantage of FPM is that non-proportional hazards can easily be fitted by adding a second spline for the interaction with time. Considering the correlation within the matched clusters, a shared frailty term (as random effects) was incorporated into the model and the maximum (penalized) marginal likelihood method was used to estimate the regression coefficients and the variance for the frailty. Aalen-Johansen estimation was used to assess the cumulative incidences of heart diseases in breast cancer patients and matched reference individuals, while other causes of death were considered as competing events. Next, we studied the association of adjuvant breast cancer therapy with heart disease risk in breast cancer patients using Cox proportional hazards models. We adjusted these analyses for age and year of diagnosis (model 1), and additionally for menopausal status at diagnosis, cancer stage, type of surgery, CCI score, hypertension, chronic pulmonary disease, and tobacco abuse. All treatment-specific models were mutually adjusted for adjuvant therapies. Considering the possible selection bias in the administration of radiotherapy (Wadsten et al., 2018), the analysis for radiotherapy only included patients receiving radiotherapy, making a comparison between left-sided, right-sided, and both-sided breast cancer. Multiple imputation with chained equations was used to deal with the treatment categories with missing information. We replaced the missing data with 10 rounds of imputations and all the covariates were included in the imputation model. Considering the time-dependent effect of treatment, we separated the analysis according to different follow-up periods, within 10 years after breast cancer diagnosis and beyond, respectively. All statistical analyses were performed using STATA version 15.1. Results Descriptive characteristics of the study population are described in Table 1. The median age at breast cancer diagnosis was 59 years, with 74.3% of patients aged less than 65 years (Table 1). Approximately 40% of all patients received adjuvant chemotherapy, with anthracyline-based regimens being most frequently administered. Endocrine therapy was received by more than 80% of patients (constituting mainly of tamoxifen). Of all patients diagnosed between mid-2005 and 2008, 13% were coded as receivers of trastuzumab. Table 1 Descriptive characteristics of the study population. CharacteristicsOverall (N = 8015)Year of diagnosis % (N)2001–200225.6 (2072)2003–200424.4 (1952)2005–200625.8 (2064)2007–200824.0 (1927)Age at diagnosis % (N)<50 years23.0 (1842)50–65 years51.3 (4111)>65 years25.7 (2062)Menopausal status % (N)Premenopausal30.4 (2289)Postmenopausal69.6 (5251)Missing (N)475Stage % (N)Stage I48.2 (3690)Stage II39.4 (3014)Stage III12.4 (948)Missing (N)363Tumor size % (N)≤2 cm25.7 (2000)2–5 cm44.0 (3418)> 5 cm30.3 (2351)Missing (N)246Comorbidity % (N)None90.1 (7222)15.5 (439)≥24.4 (354)History of hypertensionNo94.6 (7579)Yes5.4 (436)History of chronic pulmonary disease or tobacco abuseNo96.6 (7746)Yes3.4 (269)Surgery % (N)No1.0 (79)Yes, breast-conserving60.7 (4852)Yes, mastectomy38.3 (3058)Missing (N)26Radiotherapy % (N)No22.5 (1774)Yes, left-sided37.5 (2962)Yes, right-sided39.1 (3088)Yes, both-sided0.85 (67)Missing (N)124Chemotherapy % (N)No58.5 (4604)Yes41.5 (3262)Missing (N)149Hormone therapy % (N)No18.10 (1424)Yes, tamoxifen53.0 (4247)Yes, aromatase inhibitors19.3 (1550)Yes, type unknown8.1 (645)Missing (N)149Trastuzumab therapy % (N) *No87.3 (2180)Yes12.7 (316)Missing (N)1497 * Missingness on individual variables is less than 5%, except for menopausal status (5.9%, N = 488). Treatment-specific analysis of trastuzumab was restricted to patients diagnosed between 2005 and 2008 (missingness = 37.5%, N = 1497). Over a median follow-up of 10.8 years (interquartile range = 6.5 years), arrhythmias were the most frequently reported heart disease (n = 570), followed by ischemic heart disease (n = 307) and heart failure (n = 243). The cumulative incidence of arrhythmias, ischemic heart disease, and heart failure in the breast cancer patients cohort was 11.0%, 5.7%, and 4.8% after 15 years of follow-up, while the cumulative incidence was 8.2%, 5.9%, and 3.8% in the matched cohort, respectively (Appendix 2—table 2 and Appendix 3—figure 1). Figure 1 shows the time-dependent risks of heart diseases in breast cancer patients compared to the reference population. A short-term increase in risks of arrhythmia and heart failure was found in breast cancer patients (Table 2, Figure 1, HR at first year for arrhythmia = 2.14; 95% CI = 1.63–2.81, for heart failure = 2.71; 95% CI = 1.70–4.33, respectively). An elevated risk of ischemic heart disease was also observed in the first year after diagnosis (HR = 1.45; 95% CI = 1.03–2.04), followed by a decline with increasing follow-up time. Interestingly, increased risks for arrhythmia and heart failure were noted beyond 10 years (Table 2, Figure 1, HR for arrhythmia = 1.42; 95% CI = 1.21–1.67; HR for heart failure = 1.28; 95% CI = 1.03–1.59). Figure 1 Download asset Open asset Time-dependent hazard ratio of heart diseases in breast cancer patients compared to age-matched women. In all models, time since index date was the underlying time scale and a restricted cubic spline with four internal and two boundary knots (five degrees of freedom) placed at quintiles of the event times was used for the baseline hazard. Time-dependent effects were modelled by adding interaction terms with time using a second spline with three degrees of freedom. Figure 1—source data 1 Summary data for Figure 1. https://cdn.elifesciences.org/articles/71562/elife-71562-fig1-data1-v2.xlsx Download elife-71562-fig1-data1-v2.xlsx Table 2 Hazard ratios for heart diseases in breast cancer patients compared to the matched cohort. ArrhythmiaHeart failureIschemic heart diseaseNo.HR (95% CI)No.HR (95% CI)No.HR (95% CI)Time since diagnosis<1 year642.14 (1.63–2.81)222.71 (1.70–4.33)381.45 (1.03–2.04)1–2 years341.08 (0.76–1.53)192.07 (1.27–3.37)341.12 (0.79–1.61)2–5 years1071.07 (0.88–1.30)381.14 (0.82–1.59)720.84 (0.66–1.07)5–10 years2041.13 (0.98–1.30)781.02 (0.81–1.29)1040.82 (0.67–1.00)10–17 years1611.42 (1.21–1.67)861.28 (1.03–1.59)590.79 (0.61–1.03) Abbreviations: No. = number of cases. HR = hazard ratio. CI = confidence interval. The HRs are estimated using flexible parametric model and conditioned on matching criteria (year of birth). In all models, time since index date was the underlying time scale and a restricted cubic spline with four internal and two boundary knots (five degrees of freedom) placed at quintiles of the event times was used for the baseline hazard. All analyses were stratified by time since index date. Statistically significant results with p-value<0.05 are bolded. Table 3 presents the HRs in breast cancer patients by adjuvant therapy. Patients treated with chemotherapy and trastuzumab had an increased risk of heart failure compared to patients not receiving these treatments (for patients receiving anthracyclines-based chemotherapy, HR = 1.74, 95% CI = 1.20–2.52; for patients receiving trastuzumab, HR = 2.34, 95% CI = 1.05–5.22) (Table 3). The risk of heart failure in patients receiving anthracyclines potentially persisted for 10 years beyond the diagnosis of breast cancer (HR = 1.66, 95% CI = 0.86–3.19, Appendix 2—table 3). Receipt of aromatase inhibitors was associated with risk of ischemic heart disease (HR = 1.52, 95% CI = 1.03–2.26 compared to patients without hormonal therapy), but not with risk of heart failure or arrhythmias (Table 3). A direct comparison between left-sided and right-sided radiotherapy showed no evidence of an association of radiotherapy with heart disease except for a somewhat increased risk of ischemic heart disease in women with left-sided breast cancer, particularly after 10 years after cancer diagnosis, although not statistically significant (Appendix 2—table 3, HR = 1.16, 95% CI = 0.89–1.51; HR for the risk beyond 10 years = 1.29, 95% CI = 0.70–2.37). Table 3 Hazard ratios for heart diseases in breast cancer patients by different adjuvant therapies. Total No.HR (95% CI) for arrhythmiaHR (95% CI) for heart failureHR (95% CI) for ischemic heart diseaseTreatment variablesNModel 1Model 2NModel 1Model 2NModel 1Model 2RadiotherapyRight-sided2948216REF (1.00)REF (1.00)80REF (1.00)REF (1.00)99REF (1.00)REF (1.00)Left-sided30752130.93 (0.77–1.13)0.93 (0.77–1.13)871.03 (0.76–1.40)1.07 (0.79–1.45)1191.14 (0.87–1.49)1.16 (0.89–1.51)Both-sided6740.65 (0.24–1.76)0.63 (0.23–1.70)41.81 (0.66–4.95)1.75 (0.63–4.85)31.08 (0.34–3.40)0.98 (0.31–3.11)ChemotherapyNo.4604376REF (1.00)REF (1.00)133REF (1.00)REF (1.00)207REF (1.00)REF (1.00)Anthracyclines-based1426831.18 (0.97–1.45)1.08 (0.84–1.39)512.30 (1.70–3.10)1.74 (1.20–2.52)441.31 (0.99–1.72)1.29 (0.92–1.81)Anthracyclines + taxanes286101.15 (0.63–2.10)1.01 (0.53–1.92)134.71 (2.57–8.63)3.09 (1.55–6.14)61.36 (0.63–2.93)1.29 (0.57–2.95)CMF9661.08 (0.55–2.14)1.04 (0.51–2.12)31.21 (0.36–4.00)0.90 (0.27–3.04)21.11 (0.49–2.51)1.04 (0.46–2.36)Hormone therapyNo.142489REF (1.00)REF (1.00)49REF (1.00)REF (1.00)43REF (1.00)REF (1.00)REF (1.00)REF (1.00)REF Total No. to the number of patients. N events to the number of observed cases. HR = hazard CI = confidence = cyclophosphamide, methotrexate, and reference women in the matched controls from the general population. Hazard ratios are estimated from Cox proportional hazards models with time since diagnosis as underlying time scale. Hazard ratios for model 1 are adjusted for age and Hazard ratios for model 2 are adjusted including age at diagnosis, year of diagnosis, menopausal Charlson comorbidity stage, tumor type of surgery, of hypertension, chronic pulmonary disease and tobacco abuse, and all treatment variables in the * Treatment-specific analysis of trastuzumab was restricted to patients diagnosed between 2005 and Discussion In a we that the incidence of heart disease in breast cancer patients was significantly than the incidence observed in matched reference from the general population. The risks of arrhythmia and heart failure were increased beyond 10 years after diagnosis. Receipt of trastuzumab, as well as administration of was associated with an increased risk of heart failure, while of aromatase inhibitor therapy was associated with an increased risk of ischemic heart disease. We found an increased risk of arrhythmia and heart failure in breast cancer patients as compared with the matched reference from the general which is similar to the risk of heart failure reported by a previous study et al., the of our findings to However, as patients in our cohort were aged between and years, is when these findings to who may have more by time since diagnosis revealed long-term increased risks of arrhythmia and heart failure following breast cancer diagnosis, that a cardiac be to cardiac health in breast cancer patients. As the long-term risk was observed for heart failure but not ischemic heart disease, the cardiotoxic effect of chemotherapy be mainly on the by the effect of through but not the cardiac et al., The that risk of ischemic heart disease in breast cancer patients was only elevated after diagnosis is not the of with a cancer diagnosis in the which may to short-term of ischemic heart disease et al., 2012; et al., 2016). In addition, surgery after breast cancer diagnosis increase the risk of et al., which myocardial infarction, and the effect appears to 1 year after diagnosis et al., 2017; et al., 2019). The long-term risk of ischemic heart disease in breast cancer patients compared to age-matched women be by the of in breast cancer and ischemic heart disease. with age at and age at were associated with increased risk of breast cancer, while risk of ischemic heart disease were found these women Group on in 2012; et al., 2020). The findings that administration of as well as trastuzumab, is associated with increased risks of heart failure are in with previous studies (Bowles et al., 2012; et al., 2013; Doyle et al., 2005; et al., et al., et al., and et al., in specific subgroups of breast cancer patients. While is to to heart failure through a of that and Bickford, may also the and of and The cardiotoxic effect of trastuzumab may be by of the HER-2 in which the through of and which and et al., Yeh and Bickford, 2009). As is no cardiac for chemotherapy in routine and cardiac assessment is only performed prior to and during the treatment period for HER-2 positive patients in a cardiac be for these patients. estimates of incidence of ischemic heart disease following treatment with aromatase inhibitors have not been reported in the et al., et also found an increased risk of myocardial in patients treated with aromatase inhibitors, as compared to treated with tamoxifen et al., 2016). has been that this increase in risk of ischemic heart disease in be to risk of heart disease following tamoxifen treatment et al., 2011; et al., that in the study, an increased risk of ischemic heart disease was observed in patients treated with aromatase inhibitors compared to patients not receiving hormone treatment, the effect from tamoxifen not the reported risk increase of ischemic heart disease in patients treated with aromatase While previous studies have shown that radiotherapy for breast cancer increase the risk of ischemic heart disease and heart failure (Darby et al., 2013; Harris et al., 2006; Pinder et al., some particularly including patients diagnosed in recent have not an increase et al., et al., et al., 2020). the time of heart and the use of heart with doses of may in doses of to the heart et al., 2017; Taylor and 2015; Taylor et al., 2009). for results in the be that patients with a left-sided breast cancer who are to cardiovascular diseases are less to be for radiotherapy (Darby et al., 2013; Taylor and in our study, we compared risk of ischemic heart disease in patients treated with radiotherapy for left-sided to treated for right-sided breast cancer. We found increased risk of ischemic heart disease, this risk (as more in the follow-up period after 10 our results only risk of heart disease due to radiotherapy in women treated in after year studies with detailed information on the heart of or cumulative are to and more to this A of our study is that we were to risks of heart disease by time since diagnosis and adjuvant The risk estimates in this study an into the short-term and long-term effects of adjuvant breast cancer therapy in routine all breast cancer patients were for a long The nature of the study and linkage to data further the risk of selection and information We The Swedish Patient Register has for heart failure, arrhythmia, and ischemic heart disease positive between and et al., et al., by diagnoses However, of heart diseases may have In addition, comorbidities extracted from the patient register may not patients with we not have data on trastuzumab during the study period and on HER-2 status as which may have in some The of heart disease after treatment with trastuzumab and were also in our study, and have in or when identifying an association between and the relevant cardiac events. the Stockholm-Gotland Breast Cancer Register only records treatment, not patients received these therapies. However, the between the and breast cancer treatment in has been reported to be about 95% et al., 2019). it be noted that cancer patients are to increased medical in the period after diagnosis. This the increased of heart disease events within the first year of follow-up, although
Source: eLife
Language: English
Subfield: Cardiology and Cardiovascular Medicine
Field: Medicine
Domain: Health Sciences
Open Access status: gold
Estimated APC paid: $2,500
OpenAlex ID: W4220752019