https://orcid.org/0000-0001-6127-5401
Figures
Abstract
Background
With the increasing survival rate and overall lifespan of patients with breast cancer, increased mortality due to cardiovascular causes is a notable concern. Anthracyclines, which are commonly used and effective chemotherapeutic agents for breast cancer, are known for their cardiotoxicity. Existing studies vary in their findings owing to heterogeneous factors such as study population and design, underscoring the need for high-quality evidence to comprehensively examine the risk factors for anthracycline-induced cardiotoxicity. This study compared the incidence of cardiovascular disease, all-cause mortality, and their associated risk factors following adjuvant anthracycline therapy as the initial anticancer treatment, utilizing National Health Insurance claims data for a nationwide cohort of patients with breast cancer.
Methods
The study cohort comprised patients initially diagnosed with breast cancer between 2011 and 2013 who underwent mastectomy within 6 months of diagnosis and received anticancer therapy within 6 months of surgery. The primary outcomes were composite cardiovascular events and all-cause mortality during a follow-up period of up to 5 years. Secondary outcomes encompassed specific types of cardiovascular events such as congestive heart failure, heart failure and cardiomyopathy, coronary artery disease and cardiac arrest, and stroke. Incidence rates for these outcomes were presented as incidence rate ratios (IRR), and hazard ratios (HR) for individual risk factors were derived using the Cox proportional hazards model.
Results
The study cohort comprised 9,439 patients with breast cancer in both the anthracycline and non-anthracycline groups. There was no significant association between adjuvant anthracycline therapy and the composite cardiovascular events (adjusted HR 0.911 [95% CI 0.823–1.008]). In contrast, all-cause mortality was higher in the anthracycline group (IRR 2.155 [95% CI 1.892–2.455]; adjusted HR 2.160 [95% CI 1.882–2.480]). Adjuvant trastuzumab was significantly associated with an increased risk of the composite cardiovascular event (adjusted HR 1.256 [1.093–1.443]), although no trend was identified concerning adjuvant trastuzumab status in the subgroup analysis. Advanced age and hypertensive comorbidities were risk factors for all outcomes.
Citation: Lim Y-C, Lee YJ, Ha I-H, Lee Y-S (2025) Risk of cardiovascular disease and death in patients with breast cancer receiving anthracycline-based therapy: A retrospective cohort study. PLoS One 20(12): e0335083. https://doi.org/10.1371/journal.pone.0335083
Editor: Mohammad Zeeshan Najm, Apeejay Stya University, INDIA
Received: May 12, 2025; Accepted: October 6, 2025; Published: December 4, 2025
Copyright: © 2025 Lim et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The data underlying this study are not publicly available due to restrictions imposed by the Korean National Health Insurance Service (NHIS). Access to the NHIS health insurance claims database requires approval from the NHIS Institutional Data Access/Ethics Committee. Qualified researchers may request access to the data by contacting the NHIS through the following channels: Website: https://nhiss.nhis.or.kr/; Email: nhiss@nhis.or.kr; Phone: +82-1577-1000; Address: 32 Gungang-ro, Wonju-si, Gangwon-do 26464, Republic of Korea.
Funding: This research was supported by a grant from the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health and Welfare, Republic of Korea (grant number: HF20C0222).
Competing interests: The authors have declared that no competing interests exist.
Abbreviations: IRR, incidence rate ratios; HR, hazard ratios; CI, confidence intervals; BC, breast cancer; HER2, human epidermal growth factor receptor-2; CHF, congestive heart failure; HERA, Herceptin Adjuvant; CVD, cardiovascular disease; LVEF, left ventricular ejection fraction; NHIS, National Health Insurance Service; NHI, National Health Insurance; NAC, neoadjuvant chemotherapy; ICD-10, International Classification of Diseases 10th Revision; CAD, coronary artery disease; HF, heart failure; ST, stroke; CCI, Charlson comorbidity index; SERM, selective estrogen receptor modulators; AI, aromatase inhibitors; ER, estrogen receptor; PR, progesterone receptor; COPD, chronic obstructive pulmonary disease; RTx, radiotherapy
Background
Breast cancer (BC) is the most common malignancy affecting women worldwide. In recent decades, the development of early diagnostic techniques and neoadjuvant and adjuvant therapies have led to reduced mortality and increased survival of patients with BC [1,2]. With increasing survival and duration of disease, patients with BC are exposed to adjuvant chemotherapy regimens for a relatively long period, and several studies have been conducted to examine the side effects of the chemotherapeutic agents [3,4]. A representative example of the side effects of anticancer drugs with well-described causal relationship is anthracycline-induced cardiotoxicity [5,6]. Anthracyclines cause damage to myocardial cells via free radical generation [7]. As for the types of anthracycline-induced cardiotoxicity, acute cardiotoxicity occurs in <1% of the patients immediately after drug infusion, the early-onset type occurs in 1.6%–2.1% of patients during treatment or within the first year after treatment, and the late-onset type occurs in 1.6%–5% of patients at least 1 year after completion of chemotherapy [8]. Trastuzumab, which contributes to a remarkable increase in the survival rate of patients with human epidermal growth factor receptor-2 (HER2)-positive (+) early BC, is also among the widely known cardiotoxic agents [9,10]. A high incidence of severe congestive heart failure (CHF) was reported in the Herceptin Adjuvant (HERA) trial, one of the largest clinical studies on trastuzumab [11]. Severe cardiotoxicity following treatment with trastuzumab was reported in 3% of the patients in a meta-analysis of 29,000 women with BC [12]. Anthracyclines and trastuzumab have different cardiotoxic mechanisms; therefore, the concurrent use of the two agents has been reported to be associated with an increased risk of cardiotoxicity [13]. In a phase 3 trial on the comparative evaluation of the efficacy and safety of the standard chemotherapy alone (doxorubicin and cyclophosphamide) and standard chemotherapy plus trastuzumab in patients with metastatic BC, 27% of the patients in the combination therapy arm experienced cardiotoxicity, whereas only 8% of the patients in the monotherapy arm experienced cardiotoxicity [14]. Other studies have reported an increased risk of cardiac toxicity in some patients with the combined use of anthracyclines and trastuzumab [14,15]. Thus, the concomitant administration of the two agents is contraindicated and a sequential strategy of administration is considered in a limited number of cases [16]. In addition to anthracycline- or trastuzumab-induced cardiotoxicity, radiotherapy, diabetes mellitus, and hypertension are the major risk factors for cardiovascular disease (CVD) in patients with BC [17–19].
To monitor and prevent cardiotoxicity, which has a serious impact on mortality rate and the quality of life of BC survivors, various recommendations have been proposed in clinical practice guidelines [2,20–22]. A recommendation is to limit the cumulative total lifetime dose of anthracyclines, which induces dose-dependent cardiotoxicity [23]. For early detection, patients should undergo routine electrocardiogram and left ventricular ejection fraction (LVEF) monitoring with echocardiography [24]. Furthermore, the underlying risk factors in individual patients and preventive treatments for high-risk populations should be considered [25]. Although various strategies are used as recommended by these clinical guidelines, further studies are needed to verify and confirm their effectiveness in preventing cardiotoxicity in real-world practice [26]. In addition, extensive research and better understanding are required for cases with little clinical manifestation of the symptoms or for the risk of late-onset, long-term cardiotoxicity [27–29]. Therefore, there is a pressing need for the accumulation of more clinical evidence that would enable the identification of patients at a higher risk of CVD and for subsequently establishing a treatment strategy to reduce the risk of developing CVD.
For CVD, which is the leading cause of mortality among patients with BC other than the cancer itself, a number of previous studies investigated the incidence and clinical predictors (including exposure to a high cumulative doxorubicin dose, trastuzumab use, and chest radiotherapy) of anticancer agent-induced cardiotoxicity in treatment of BC [21,30]. However, according to a systematic review on the risk of death from CVD following BC, there were limitations in the generalizability of findings because published studies were heterogeneous in terms of their study design, population, study period, age of patients included in the analysis, and cancer stage of the patients included in the analysis. The number of included studies was also small [31]. Therefore, in this study, a claims database, which includes a nationwide cohort of patients with BC in Korea, was used for a detailed analysis. For the study population of all patients with BC in Korea, long-term follow-up monitoring was conducted for outcomes of CVD events and all-cause mortality following anthracycline therapy, and risk factors influencing the occurrence of the outcomes were evaluated.
Methods
Data source
This nationwide, retrospective cohort study used healthcare claims data retrieved from the National Health Insurance Service (NHIS) of Korea. The database covers approximately 50 million people (99% of the South Korean population) from January 1, 2002, to December 31, 2020. The recorded information includes details on sex, age, inpatient and outpatient medical services, diagnostic codes, drug prescriptions, and healthcare providers. This database is nationally representative and includes the dates of death of the population from Korean Statistics. Cause-of-death information was not available in the claims extract; thus, mortality analyses were limited to all-cause deaths. The related materials and metadata are publicly available on the National Health Insurance Data Sharing Service homepage (http://nhiss.nhis.or.kr (accessed on October 6, 2023)). The Institutional Review Board of Gachon University granted a formal waiver for consent (no.: 1044396–202101-HR-001–01).
Study cohort
The National Health Insurance (NHI) claims data from January 1, 2002, were used, and patients who were first diagnosed with BC (International Classification of Diseases 10th Revision [ICD-10] code: C50) between January 1, 2011, and December 31, 2013, were included in the study cohort; the date of the first diagnosis of BC was defined as the entry date. To control for the BC stage and disease-related characteristics of the patients included in the study cohort, those who underwent mastectomy within 6 months after the first diagnosis of BC and received chemotherapy within 183 days after mastectomy were included. The exclusion criteria were as follows: [1] patients who underwent treatment that may be considered BC treatment (anticancer therapy [chemotherapy], mastectomy, radiotherapy) during the year preceding the entry date; [2] patients who received neoadjuvant chemotherapy; [3] patients with a history of cancer other than BC or CVD from 1 year preceding the entry date to the start of follow-up; and [4] patients with missing data in the study variables and male patients (Figs 1, 2).
ICD-10: International classification of diseases 10th revision.
NAC: neoadjuvant cancer therapy; CCI: Charlson comorbidity index.
Exposure
In this study, adjuvant anthracycline therapy as anticancer therapy for BC was defined as exposure. The study cohort design consisted of an anthracycline group for patients who received at least one prescription of anthracycline therapy within 1 year from the date of the first adjuvant anticancer therapy (defined as the index date) and a non-anthracycline group for patients who did not receive anthracycline therapy. The treatment window was defined as 1 year considering the treatment period of anthracyclines and other agents used as anticancer therapy for BC [32] with the aim of evaluating the risk of early onset and late-onset chronic progressive cardiotoxicity, excluding acute cardiotoxicity cases [8]. Anthracyclines included chemotherapeutic agents with doxorubicin or epirubicin that were listed for NHI reimbursement during the study period. Anthracycline exposure was defined as any prescription of doxorubicin or epirubicin within 365 days of the index date; cumulative dose (e.g., mg/m²) and dose intensity were not available in the claims data and were therefore not incorporated. Owing to the increased risk of heart failure (HF) in cases of concomitant administration with anthracyclines, concurrent administration of trastuzumab was contraindicated. However, since trastuzumab, whose sequential administration following the adjuvant anthracycline therapy may be considered in clinical practice, is a major influencing factor for the evaluation of outcomes in this study, subgroup analysis was performed by grouping the patients according to adjuvant trastuzumab status.
Study outcomes and covariates
CVD events were defined as cases of NHI claims with diagnostic codes of coronary artery disease (CAD) and cardiac arrest, CHF, HF and cardiomyopathy, or stroke (ST) within the 5th-listed secondary diagnosis during the follow-up period. The diagnostic codes defined as the outcomes of this study were based on the ICD-10, and the details are provided in S1 Table. The primary outcomes of interest were the incidence of the composite cardiovascular events and all-cause death during the study period, and the evaluation of risk factors affecting the incidence of these events. The primary composite cardiovascular outcome was defined as time to the first occurrence of any of the following subtypes: coronary artery disease and cardiac arrest, congestive heart failure, heart failure and cardiomyopathy, or stroke; follow-up ended at the earliest of the first CVD event, death, or end of follow-up. For secondary outcomes, we conducted separate time-to-event analyses for each subtype; occurrences of other CVD subtypes did not censor follow-up for the subtype under analysis, and recurrent events within the same subtype were not counted.
From the nationwide cohort database in this study, measurable sociodemographic characteristics (age and household income quintile), Charlson comorbidity index (CCI) scores, and comorbidities related to the incidence of outcomes were included as covariates [33–35]. Information on the age and household income quintiles of patients with BC was collected from NHI claims issued on the index date, and comorbidities and CCI scores were evaluated from claims issued during the year preceding the index date. The status of adjuvant trastuzumab, which is associated with cardiotoxicity, within 1 year from the index date [10] and the history and frequency of radiotherapy from the entry date (date of the first diagnosis of BC) to 1 year from the index date were included as covariates in this study [17]. In addition, the frequency of adjuvant anticancer therapy administered within 1 month of the index date for each patient group was documented and is presented in the supplementary.
Statistical analysis
Differences in baseline characteristics between the anthracycline and non-anthracycline groups were compared using the Chi-square test. The incidence rate (per 100 person-years) was derived by measuring the incidence of the defined outcomes for each patient group and the time to onset for each incidence. The incidence rate ratio (IRR) and 95% confidence interval (CI) were also calculated. Follow-up was defined as monitoring up to the earliest time point of the occurrence of a CVD event, death, or the end of the 5-year follow-up period. Accordingly, follow-up began 365 days after the index date (the first adjuvant anticancer therapy) and continued until the earliest occurrence of a composite cardiovascular event, death, or 5 years. Participants without an event were censored at 5 years or at the end of available data, whichever came first. For secondary outcomes, we conducted separate cause-specific time-to-first-event analyses for each subtype; events of other CVD subtypes did not censor follow-up for the subtype under analysis, allowing a single patient to contribute at most one first event to multiple different subtypes. For analysis of risk factors for CVD and mortality of the patients with BC, the Cox proportional hazards model was used to calculate hazard ratios (HR) and 95% CIs. Proportional hazards were assessed by visual inspection of Kaplan–Meier curves for the exposure groups; no major departures (e.g., curve crossing) were observed. In the Cox model, exposure to adjuvant anthracycline therapy, CCI score, household income quintile, presence of adjuvant trastuzumab, comorbidities, and frequency of radiotherapy were included as covariates. Patients were grouped into subgroups of those who underwent adjuvant trastuzumab therapy within 1 year from the index date and those who did not, and a subgroup analysis was performed to evaluate the difference in the trend of CVD event incidence. A 1:1 greedy propensity score matching was performed for age group (exact variable), household income quantile, CCI score, and number of radiotherapies. Differences in characteristics between the patient groups before and after propensity score matching were analyzed using Chi-square tests. Statistical significance was defined as two-sided p < 0.05. All the statistical analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC, USA).
Results
Baseline characteristics
From a cohort of 56,618 patients who were first diagnosed with BC between 2011 and 2013, 9,439 patients were included in both the anthracycline and non-anthracycline groups after a final screening based on the inclusion and exclusion criteria. (Table 1) The age group included as an exact matching variable in the propensity score matching showed an identical distribution between the two groups, and the distribution of other variables showed similarity between the two groups after matching. However, statistically significant between-group differences in these variables were still observed. In terms of comorbidities, higher proportions of other CVDs and sleep disorders occurred in the anthracycline group (4.29% non-anthracycline vs. 5.37% anthracycline; 12.71% non-anthracycline vs. 14.30% anthracycline, respectively), whereas osteoporosis showed higher prevalence in the non-anthracycline group (30.59% non-anthracycline vs. 25.46% anthracycline). Most other comorbidities showed similar trends between the two groups (Table 2; extended variables in S2 Table). In the case of the non-anthracycline group, as the adjuvant anticancer therapy for the first year, selective estrogen receptor modulators (SERM) (64.11% vs. 44.76%), aromatase inhibitors (AI) (30.58% vs. 25.01%), and methotrexate (20.18% vs. 0.19%) were used by higher percentages whereas taxanes (1.33% vs. 38.73%) were used by a small percentage of the patients (S3 Table).
Analysis of incidence
The incidence rate of the composite cardiovascular events was 1.864 per 100 person-years for the anthracycline group and 1.958 per 100 person-years for the non-anthracycline group, and the IRR was 0.952 (95% CI 0.866–1.046), confirming no significant between-group differences. As a result of analyzing the incidence rates of the four types of CVD events in the anthracycline group, CHF (IRR 1.185 [95% CI 1.013–1.386]) and HF and cardiomyopathy (IRR 1.197 [95% CI 1.008–1.421]) showed high incidence rates, and CAD and cardiac arrest (IRR 0.900 [95% CI 0.794–1.021]) and ST (IRR 0.842 [95% CI 0.686–1.033]) showed relatively low incidence rates, but no statistically significant differences were confirmed. In contrast, the incidence of all-cause mortality was 1.434 per 100 person-years in the anthracycline group and 0.684 per 100 person-years in the non-anthracycline group, indicating a higher incidence of all-cause mortality in the anthracycline group (IRR 2.155 [95% CI 1.892–2.455]) (Table 3).
Analysis of hazards
Similar to the results from the analysis of incidence rates, the results of Cox proportional hazards analysis showed that adjuvant anthracyclines administered in the first year did not have a significant impact on the risk of the composite cardiovascular events (adjusted HR 0.911 [95% CI 0.823–1.008]); however, the adjuvant anthracycline therapy increased the risk of all-cause mortality (adjusted HR 2.160 [95% CI 1.882–2.480]). Older age, high CCI scores, comorbidities of other CVDs, anemia, renal failure, and hypertension were identified as factors contributing to an increased risk of the composite cardiovascular events and all-cause mortality. Patients in the high household income quintile showed a lower risk of both outcomes. When the patients underwent adjuvant trastuzumab therapy in the first year, the risk of the composite cardiovascular events was high (aHR, 1.256 [95% CI, 1.093–1.443]) (Fig 3). Adjuvant trastuzumab increased the risk of CHF (aHR 1.715 [95% CI 1.395–2.108]) and HF and cardiomyopathy (aHR 1.871 [95% CI 1.498–2.335]), but the agent had no significant impact on the risk of CAD and cardiac arrest or ST (Table 4).
HR, hazard ratio; CCI, Charlson comorbidity index; COPD: Chronic obstructive pulmonary disease; RTx, radiotherapy.
As a result of subgroup analysis by grouping the patients into subgroups according to the use of adjuvant trastuzumab as the first year adjuvant anticancer therapy, among patients without adjuvant trastuzumab (n = 22,603 eligible), the propensity score–matched cohort comprised 9,093 patients per group (total 18,186), in whom mortality was significantly higher in the anthracycline group (aHR 2.335 [95% CI 2.027–2.689]). Advanced age and renal failure increased the risk of most outcomes, showing a similar trend as in the main analysis; however, there was a difference in that the risk of outcomes increased with higher income quintile of the patients (S4 Table). Among patients who received adjuvant trastuzumab (n = 3,815 eligible), the propensity score–matched cohort included 386 patients per group (total 772); within this matched set, adjuvant anthracyclines were not a significant predictor for any outcome. Similar to the results of the other analyses described above, advanced patient age increased the risk of most outcomes (S5 Table).
Discussion
In this study, the outcomes of the composite cardiovascular events and all-cause mortality were compared between patients administered and those not administered anthracycline treatment among those who underwent adjuvant anticancer therapy within 1 year after the first diagnosis of BC. In addition, the risk factors affecting the incidence of the outcomes were evaluated. The incidence rates of CHF, HF and cardiomyopathy, and all-cause mortality were higher in the anthracycline group. Advanced age, high CCI score, anemia, and renal failure were common major risk factors for the composite cardiovascular events and mortality. Adjuvant trastuzumab therapy was a risk factor for CHF and HF and cardiomyopathy; however, when subgroup analysis was performed considering the presence of adjuvant trastuzumab, adjuvant anthracyclines were not a significant risk factor for all outcomes. Given the large sample size, small differences may achieve statistical significance; therefore, we emphasized effect sizes with 95% CIs and clinical relevance over p-values and reported absolute incidence rates to aid interpretation.
The prognosis of BC with both hormone receptors (estrogen receptor [36] or progesterone receptor [PR]) positive (ER + /PR+) is better than that with only one hormone receptor positive or both hormone receptors negative, as the tumor phenotypes in the latter cases are more aggressive and have a more unfavorable prognosis [37]. In the case of ER + /PR + BC, endocrine therapy such as AI or SERM is considered the first-line therapy. Thus, there are more patients with hormone receptor-positive BC in the non-anthracycline group, which resulted in a lower mortality rate. In this study, the frequency of the concomitant administration of different anticancer agents between the two groups showed that a high proportion of patients in the non-anthracycline group received AI and SERM therapies (S3 Table). Many of the patients in the anthracycline group also underwent endocrine therapy, and these are presumed to be patients with metastatic BC who had metastasis after endocrine therapy or whose treatment failed due to resistance to the endocrine therapy [38], which may have contributed to increasing the risk of mortality in the anthracycline group [37]. Beyond differences in endocrine therapy, residual confounding by indication remains a plausible explanation: anthracyclines are often prescribed for biologically higher‐risk disease, and unmeasured tumor characteristics (e.g., stage, grade, burden) not captured in claims data may have adversely influenced prognosis despite propensity score matching and adjustment [39]. Additionally, subclinical or non-coded cardiotoxicity—manifesting as asymptomatic declines in left-ventricular ejection fraction—may prompt treatment modification without being captured by ICD-coded cardiovascular endpoints, thereby indirectly contributing to poorer cancer outcomes [40].
In the present study, based on the analysis of real-world data, the unadjusted incidence rates of CHF and HF and cardiomyopathy in patients with BC treated with adjuvant anthracyclines were found to be high; however, Cox proportional hazard analysis adjusted for several covariates showed that the effect of adjuvant anthracyclines on the outcomes was not significant. Anthracycline-induced cardiotoxicity, a side effect of chemotherapy agents, was reported several decades ago, and methods of active monitoring and prevention of chemotherapy-induced cardiotoxicity have been under development [2,5,6,24]. Thus, recent studies on the treatment of patients with BC report that the side effects do not progress to severe cardiotoxicity. In the patient cohort of the present study, CVDs other than cardiotoxicity (CAD and cardiac arrest, and ST) did not show a specific trend. Unlike adjuvant anthracyclines, adjuvant trastuzumab therapy has a significant impact on the incidence of CHF and HF and cardiomyopathy. Trastuzumab-induced cardiotoxicity is also a commonly reported side effect [11], and the findings of this study also showed consistency in this aspect; however, the results of the subgroup analysis based on subgroups depending on the adjuvant trastuzumab status showed no statistical significance in terms of the risk of cardiotoxicity (S4 and S5 Tables). Subgroup analysis can be considered as an analysis based on the types of receptor expression between groups, consisting of the HER-2 overexpression group, for whom adjuvant trastuzumab is recommended, and the group of the other patients without HER-2 overexpression [41]. In each subgroup, adjuvant anthracycline therapy was not a significant risk factor for incidence rates of CHF or HF and cardiomyopathy. Although the sample size was small, the administration of anthracycline and trastuzumab (concomitant or sequential) in the first year of adjuvant therapy was not significantly associated with the risk of CHF and HF and cardiomyopathy.
With increased overall survival after treatment, CVD is now considered one of the leading causes of death in patients with BC. A previous study reported that BC-related and other-cause deaths increased with age and in patients with CVD [42], which is consistent with the findings of this study. Although few studies have been published on renal disease in patients with BC, the findings of this study show that renal disease, one of the major risk factors for CHF, is associated with an increased risk of cardiovascular events and mortality [43]. This study also showed that adjuvant radiotherapy was associated with a reduced risk of all-cause mortality, which is similar to that reported in previous studies [44,45]. On the contrary, a previous study reported that adjuvant radiotherapy induced cardiovascular complications and increased mortality from cardiovascular causes [46], and the difference between the findings of this study and those of the previous study may be because of the fact that the duration of follow-up in this study was not sufficiently long for evaluation of radiotherapy-induced CVDs. In addition, although not included in the analysis, active research is underway on cardioprotective therapies (e.g., beta-blockers, statins, or dexrazoxane) that may reduce the risk of CVD and death in patients with BC [47].
With advancements in diagnostic techniques and the development of personalized treatments for BC, the treatment patterns of individual real-world patients with BC are becoming highly diverse [48]. Thus, even if studies use the same outcomes for assessment, patient characteristics are heterogeneous in each prior study, resulting in heterogeneous results among these studies and posing difficulties in comparisons between studies. The significance of this study lies in the comparison of incidences of the CVD events and all-cause mortality following adjuvant anthracycline therapy for patients with BC who underwent mastectomy during the early period from diagnosis without neoadjuvant anticancer therapy, as well as risk factors for the outcomes, which contributed to establishing clinical evidence that can be used as basic reference data for BC treatment in real-world clinical settings. In the future, using data sources across multiple countries and those containing a wide range of patient information such as cancer stage and laboratory test results, follow-up studies are warranted to obtain clinical evidence considering the various characteristics of real-world patients with BC.
This study analyzed the incidence of outcomes in patients with BC in Korea using NHI claims data. By ensuring a follow-up period of up to 5 years per patient, which allowed the monitoring of long-term effects, late-onset cardiotoxicity in patients with BC was evaluated. The incidence of the composite cardiovascular events, rather than a decrease in LVEF, an indicator of early cardiotoxicity, was used as the outcome, which enabled the evaluation of the incidence of severe cardiotoxicity. Furthermore, the incidences of CAD and cardiac arrest, and ST, which are CVDs other than cardiotoxicity, were presented, which allowed the detection of any bias from the characteristics of the patient cohort.
This study also has several limitations. First, data such as BC stage, laboratory data, and lifestyle factors could not be collected for each patient because these characteristics cannot be collected from the NHI claims data. To mitigate selection bias, we restricted the cohort to patients who underwent mastectomy within 6 months of the first BC diagnosis without neoadjuvant therapy, and who initiated adjuvant anticancer therapy within 6 months after surgery. Baseline characteristics were balanced using PSM. Some residual imbalance remained; however, in large samples, small absolute differences may remain statistically significant. We further adjusted for covariates using multivariable Cox models in the matched cohort, but residual and unmeasured confounding cannot be ruled out. Furthermore, formal proportional-hazards diagnostics were not conducted (e.g., Schoenfeld residuals), and although Kaplan–Meier curves did not demonstrate gross crossing, visual inspection could not exclude modest non-PH. Consequently, reported HRs should be interpreted as average effects over follow-up [49]. Subsequent analyses will incorporate formal PH tests and, where appropriate, alternatives such as time-varying Cox specifications or restricted mean survival time. Second, the absence of tumor stage and molecular subtype—key determinants of treatment and prognosis—may introduce confounding by indication and limit risk stratification and generalizability. In lieu of receptor data, we used adjuvant trastuzumab prescriptions as a pragmatic proxy for HER2 status; although its positive predictive value is likely high, this indirect measure may miss some HER2-positive patients not treated with trastuzumab (e.g., older or comorbid individuals who did not receive trastuzumab), leading to misclassification [50]. Subgroup findings should therefore be interpreted with caution, and linkage to tumor registries or EHRs to confirm biomarkers is warranted. Third, anthracycline exposure was classified as a binary status (ever/never) within the 1-year treatment window; cumulative dose (e.g., mg/m²) and dose intensity were not captured. Given the dose–response relationship for anthracycline cardiotoxicity, this may have attenuated dose-dependent associations, and estimates should be interpreted with caution. In addition, secondary subtype analyses were modeled separately without a competing-risks or multi-state framework; other CVD subtypes were not treated as competing events, and multiple subtype-specific tests raise concerns about multiplicity. Lastly, due to the retrospective design and prespecified treatment window, anticancer therapies initiated or continued outside the window could not be fully accounted for. This prespecified 1-year window was chosen to harmonize exposure ascertainment across regimens and to align with the intensive adjuvant period during which anthracyclines are typically administered, thereby enhancing comparability and minimizing time-related ambiguities in claims data. Accordingly, our estimates primarily reflect risks in the early and early-chronic post-therapy periods; future studies using time-varying exposure frameworks can build on these findings to evaluate longer horizons.
Conclusion
Analysis of real-world data from a nationwide cohort of patients with BC in Korea showed that the incidence rates of CHF, HF and cardiomyopathy, and all-cause mortality were higher among patients who received adjuvant anthracycline therapy as the initial anticancer therapy. However, the results of the survival analysis adjusted for related covariates showed that adjuvant anthracyclines were not a significant risk factor for the composite cardiovascular events specified in this study. Older patients and those with high CCI scores and comorbidities of renal failure may have higher risks of cardiotoxicity and mortality, indicating that more thorough and well-planned monitoring is required for patients with these characteristics.
Supporting information
S1 Table. Study information (outcome definitions and ICD-10 codes).
https://doi.org/10.1371/journal.pone.0335083.s001
(DOCX)
S2 Table. Baseline characteristics (extended).
https://doi.org/10.1371/journal.pone.0335083.s002
(DOCX)
S3 Table. Frequency of adjuvant anticancer therapy administered within 1 month of the index date.
https://doi.org/10.1371/journal.pone.0335083.s003
(DOCX)
S4 Table. Multivariable Cox regression analysis for cardiovascular events in patients not treated with trastuzumab.
https://doi.org/10.1371/journal.pone.0335083.s004
(DOCX)
S5 Table. Multivariable Cox regression analysis for cardiovascular events in patients treated with trastuzumab.
https://doi.org/10.1371/journal.pone.0335083.s005
(DOCX)
References
- 1. Miller KD, Nogueira L, Mariotto AB, Rowland JH, Yabroff KR, Alfano CM. Cancer treatment and survivorship statistics, 2019. CA: A Cancer J Clin. 2019;69(5):363–85.
- 2. Nicolazzi MA, Carnicelli A, Fuorlo M, Scaldaferri A, Masetti R, Landolfi R, et al. Anthracycline and trastuzumab-induced cardiotoxicity in breast cancer. Eur Rev Med Pharmacol Sci. 2018;22(7):2175–85. pmid:29687878
- 3. Shapiro CL, Recht A. Side effects of adjuvant treatment of breast cancer. N Engl J Med. 2001;344(26):1997–2008. pmid:11430330
- 4. Azim HA Jr, de Azambuja E, Colozza M, Bines J, Piccart MJ. Long-term toxic effects of adjuvant chemotherapy in breast cancer. Ann Oncol. 2011;22(9):1939–47. pmid:21289366
- 5. Von Hoff DD, Layard MW, Basa P, Davis HL Jr, Von Hoff AL, Rozencweig M, et al. Risk factors for doxorubicin-induced congestive heart failure. Ann Intern Med. 1979;91(5):710–7. pmid:496103
- 6. Jones RL, Swanton C, Ewer MS. Anthracycline cardiotoxicity. Expert Opin Drug Saf. 2006;5(6):791–809. pmid:17044806
- 7. Bird BRJH, Swain SM. Cardiac toxicity in breast cancer survivors: review of potential cardiac problems. Clin Cancer Res. 2008;14(1):14–24. pmid:18172247
- 8. Curigliano G, Cardinale D, Suter T, Plataniotis G, de Azambuja E, Sandri MT, et al. Cardiovascular toxicity induced by chemotherapy, targeted agents and radiotherapy: ESMO clinical practice guidelines. Ann Oncol. 2012;23 Suppl 7:vii155-66. pmid:22997448
- 9. Ewer SM, Ewer MS. Cardiotoxicity profile of trastuzumab. Drug Saf. 2008;31(6):459–67. pmid:18484781
- 10. Pinder MC, Duan Z, Goodwin JS, Hortobagyi GN, Giordano SH. Congestive heart failure in older women treated with adjuvant anthracycline chemotherapy for breast cancer. J Clin Oncol. 2007;25(25):3808–15. pmid:17664460
- 11. Piccart-Gebhart MJ, Procter M, Leyland-Jones B, Goldhirsch A, Untch M, Smith I, et al. Trastuzumab after adjuvant chemotherapy in HER2-positive breast cancer. N Engl J Med. 2005;353(16):1659–72. pmid:16236737
- 12. Mantarro S, Rossi M, Bonifazi M, D’Amico R, Blandizzi C, La Vecchia C, et al. Risk of severe cardiotoxicity following treatment with trastuzumab: a meta-analysis of randomized and cohort studies of 29,000 women with breast cancer. Intern Emerg Med. 2016;11(1):123–40. pmid:26712595
- 13. Barish R, Gates E, Barac A. Trastuzumab-induced cardiomyopathy. Cardiol Clin. 2019;37(4):407–18. pmid:31587782
- 14. Slamon DJ, Leyland-Jones B, Shak S, Fuchs H, Paton V, Bajamonde A, et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med. 2001;344(11):783–92. pmid:11248153
- 15. Tarantini L, Cioffi G, Gori S, Tuccia F, Boccardi L, Bovelli D, et al. Trastuzumab adjuvant chemotherapy and cardiotoxicity in real-world women with breast cancer. J Card Fail. 2012;18(2):113–9. pmid:22300778
- 16. Rayson D, Richel D, Chia S, Jackisch C, van der Vegt S, Suter T. Anthracycline-trastuzumab regimens for HER2/neu-overexpressing breast cancer: current experience and future strategies. Ann Oncol. 2008;19(9):1530–9. pmid:18480068
- 17. Lenneman CG, Sawyer DB. Cardio-oncology: an update on cardiotoxicity of cancer-related treatment. Circ Res. 2016;118(6):1008–20. pmid:26987914
- 18. McGale P, Darby SC, Hall P, Adolfsson J, Bengtsson N-O, Bennet AM, et al. Incidence of heart disease in 35,000 women treated with radiotherapy for breast cancer in Denmark and Sweden. Radiother Oncol. 2011;100(2):167–75. pmid:21752480
- 19. Pinder MC, Duan Z, Goodwin JS, Hortobagyi GN, Giordano SH. Congestive heart failure in older women treated with adjuvant anthracycline chemotherapy for breast cancer. J Clin Oncol. 2007;25(25):3808–15. pmid:17664460
- 20. Chlebowski RT. Adriamycin (doxorubicin) cardiotoxicity: a review. West J Med. 1979;131(5):364–8. pmid:394479
- 21. Armenian SH, Lacchetti C, Barac A, Carver J, Constine LS, Denduluri N, et al. Prevention and monitoring of cardiac dysfunction in survivors of adult cancers: american society of clinical oncology clinical practice guideline. J Clin Oncol. 2017;35(8):893–911. pmid:27918725
- 22. Curigliano G, Lenihan D, Fradley M, Ganatra S, Barac A, Blaes A, et al. Management of cardiac disease in cancer patients throughout oncological treatment: ESMO consensus recommendations. Ann Oncol. 2020;31(2):171–90. pmid:31959335
- 23. Gottdiener JS, Mathisen DJ, Borer JS, Bonow RO, Myers CE, Barr LH, et al. Doxorubicin cardiotoxicity: assessment of late left ventricular dysfunction by radionuclide cineangiography. Ann Intern Med. 1981;94(4 pt 1):430–5. pmid:7212498
- 24. Truong J, Yan AT, Cramarossa G, Chan KKW. Chemotherapy-induced cardiotoxicity: detection, prevention, and management. Can J Cardiol. 2014;30(8):869–78. pmid:25064580
- 25. Jerusalem G, Lancellotti P, Kim S-B. HER2+ breast cancer treatment and cardiotoxicity: monitoring and management. Breast Cancer Res Treat. 2019;177(2):237–50. pmid:31165940
- 26. Lee M, Chung W-B, Lee J-E, Park C-S, Park W-C, Song B-J, et al. Candesartan and carvedilol for primary prevention of subclinical cardiotoxicity in breast cancer patients without a cardiovascular risk treated with doxorubicin. Cancer Med. 2021;10(12):3964–73. pmid:33998163
- 27. Marinko T, Dolenc J, Bilban-Jakopin C. Cardiotoxicity of concomitant radiotherapy and trastuzumab for early breast cancer. Radiol Oncol. 2014;48(2):105–12. pmid:24991199
- 28. Di Nardo P, Lisanti C, Garutti M, Buriolla S, Alberti M, Mazzeo R, et al. Chemotherapy in patients with early breast cancer: clinical overview and management of long-term side effects. Expert Opin Drug Saf. 2022;21(11):1341–55. pmid:36469577
- 29. Bansal N, Adams MJ, Ganatra S, Colan SD, Aggarwal S, Steiner R, et al. Strategies to prevent anthracycline-induced cardiotoxicity in cancer survivors. Cardiooncology. 2019;5:18. pmid:32154024
- 30. Lotrionte M, Biondi-Zoccai G, Abbate A, Lanzetta G, D’Ascenzo F, Malavasi V, et al. Review and meta-analysis of incidence and clinical predictors of anthracycline cardiotoxicity. Am J Cardiol. 2013;112(12):1980–4. pmid:24075281
- 31. Gernaat SAM, Ho PJ, Rijnberg N, Emaus MJ, Baak LM, Hartman M, et al. Risk of death from cardiovascular disease following breast cancer: a systematic review. Breast Cancer Res Treat. 2017;164(3):537–55. pmid:28503723
- 32. Waks AG, Winer EP. Breast Cancer Treatment: A Review. JAMA. 2019;321(3):288–300. pmid:30667505
- 33. Lee YS, Lim YC, Yeo J, Kim SY, Lee YJ, Ha IH. Risk of lymphedema and death after lymph node dissection with neoadjuvant and adjuvant treatments in patients with breast cancer: an eight-year nationwide cohort study. Healthcare. 2023.
- 34. Zhang M, Yang H, Xu C, Jin F, Zheng A. Risk factors for anthracycline-induced cardiotoxicity in breast cancer treatment: a meta-analysis. Front Oncol. 2022;12:899782. pmid:35785172
- 35. Kim YA, Cho H, Lee N, Jung S-Y, Sim SH, Park IH, et al. Doxorubicin-induced heart failure in cancer patients: a cohort study based on the Korean national health insurance database. Cancer Med. 2018;7(12):6084–92. pmid:30453386
- 36. Latimer NR, Adler AI. Extrapolation beyond the end of trials to estimate long term survival and cost effectiveness. BMJ Med. 2022;1(1):e000094. pmid:36936578
- 37. Punglia RS, Kuntz KM, Winer EP, Weeks JC, Burstein HJ. The impact of tumor progesterone receptor status on optimal adjuvant endocrine therapy for postmenopausal patients with early-stage breast cancer: a decision analysis. Cancer. 2006;106(12):2576–82. pmid:16703595
- 38. Barrios CH, Sampaio C, Vinholes J, Caponero R. What is the role of chemotherapy in estrogen receptor-positive, advanced breast cancer?. Ann Oncol. 2009;20(7):1157–62. pmid:19221149
- 39. Hubbard RA, Gatsonis CA, Hogan JW, Hunter DJ, Normand S-LT, Troxel AB. “Target Trial Emulation” for observational studies - potential and pitfalls. N Engl J Med. 2024;391(21):1975–7. pmid:39588897
- 40. Camilli M, Cipolla CM, Dent S, Minotti G, Cardinale DM. Anthracycline cardiotoxicity in adult cancer patients: JACC: cardioOncology state-of-the-art review. Cardio Oncology. 2024;6(5):655–77.
- 41. Nahta R, Esteva FJ. HER-2-targeted therapy: lessons learned and future directions. Clin Cancer Res. 2003;9(14):5078–84. pmid:14613984
- 42. Patnaik JL, Byers T, DiGuiseppi C, Dabelea D, Denberg TD. Cardiovascular disease competes with breast cancer as the leading cause of death for older females diagnosed with breast cancer: a retrospective cohort study. Breast Cancer Res. 2011;13(3):R64. pmid:21689398
- 43. Russo G, Cioffi G, Di Lenarda A, Tuccia F, Bovelli D, Di Tano G, et al. Role of renal function on the development of cardiotoxicity associated with trastuzumab-based adjuvant chemotherapy for early breast cancer. Intern Emerg Med. 2012;7(5):439–46. pmid:22714882
- 44. Fisher B, Anderson S, Bryant J, Margolese RG, Deutsch M, Fisher ER, et al. Twenty-year follow-up of a randomized trial comparing total mastectomy, lumpectomy, and lumpectomy plus irradiation for the treatment of invasive breast cancer. N Engl J Med. 2002;347(16):1233–41. pmid:12393820
- 45. Early Breast Cancer Trialists’ Collaborative Group (EBCTCG), Darby S, McGale P, Correa C, Taylor C, Arriagada R, et al. Effect of radiotherapy after breast-conserving surgery on 10-year recurrence and 15-year breast cancer death: meta-analysis of individual patient data for 10,801 women in 17 randomised trials. Lancet. 2011;378(9804):1707–16. pmid:22019144
- 46. Zagar TM, Cardinale DM, Marks LB. Breast cancer therapy-associated cardiovascular disease. Nat Rev Clin Oncol. 2016;13(3):172–84. pmid:26598943
- 47. Kalam K, Marwick TH. Role of cardioprotective therapy for prevention of cardiotoxicity with chemotherapy: a systematic review and meta-analysis. Eur J Cancer. 2013;49(13):2900–9. pmid:23706982
- 48. Zardavas D, Irrthum A, Swanton C, Piccart M. Clinical management of breast cancer heterogeneity. Nat Rev Clin Oncol. 2015;12(7):381–94. pmid:25895611
- 49. Hernán MA. The hazards of hazard ratios. Epidemiology. 2010;21(1):13–5. pmid:20010207
- 50. Reeder-Hayes K, Peacock Hinton S, Meng K, Carey LA, Dusetzina SB. Disparities in use of human epidermal growth hormone receptor 2-targeted therapy for early-stage breast cancer. J Clin Oncol. 2016;34(17):2003–9. pmid:27069085