Concomitant Discontinuation of Cardiovascular Therapy and Adjuvant Hormone Therapy Among Patients With Breast Cancer

Key Points Question Do patients with breast cancer tend to discontinue their cardiovascular therapy at the same time they discontinue adjuvant hormone therapy? Findings In this population-based cohort study of 5493 patients, the highest risk of discontinuing cardiovascular therapy was observed at the time when patients discontinued adjuvant hormone therapy. Furthermore, patients who discontinued adjuvant hormone therapy were at a higher risk of both breast cancer mortality and cardiovascular mortality. Meaning These findings suggest that multiple disease−focused interventions are warranted to prevent discontinuation of treatment for other diseases and to improve overall survival in patients with breast cancer.


Introduction
Breast cancer is the most common female cancer, with 684 996 women dying of this disease in 2020 worldwide. 1Adjuvant hormone therapy reduces breast cancer mortality by 31%. 2 Despite the known benefit, discontinuation of adjuvant hormone therapy is common, with a 5-year discontinuation rate ranging from 31% to 73%, 3,4 which could subsequently lead to therapy failure and poor breast cancer outcomes. 5,6proved survival in patients with breast cancer has brought increased incidence of comorbidities.Cardiovascular disease is one of the most frequent comorbidities among patients with breast cancer. 7,8Except death due to breast cancer, cardiovascular disease is the first leading cause of death in patients with breast cancer in Sweden. 91][12] Consequently, a large proportion of patients with breast cancer concurrently use multiple medications.[15][16][17][18] Despite previous studies having identified a few shared risk factors for discontinuation of adjuvant hormone therapy and cardiovascular therapy, [13][14][15] little is known about whether discontinuation of these 2 therapies happens simultaneously.
We hypothesized that patients who discontinue adjuvant hormone therapy were more likely to discontinue cardiovascular therapy and thus had higher mortality due to cardiovascular disease.Specifically, our study aims to (1) investigate whether patients with breast cancer who discontinue adjuvant hormone therapy tend to discontinue their cardiovascular therapy concomitantly; (2)   examine breast cancer-specific mortality and cardiovascular disease-specific morality by discontinuation of adjuvant hormone therapy; and (3) identify shared and unshared factors associated with discontinuation of adjuvant hormone therapy and cardiovascular therapy.

Methods
The regional Ethical Review Authority in Stockholm, Sweden, approved this cohort study.Informed consent is not required for large-scale, registry-based studies in Sweden, provided that the study is deemed ethical by the Ethical Review Authority.This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

Data Source and Study Population
Using unique personal identification numbers, 19 we linked 8 Swedish health registers: the Stockholm-Gotland Quality Register for Breast Cancer, 9,20 Prescribed Drug Register, 21 Mammography Screening Database, 22 National Patient Register, 23 Longitudinal Integrated Database for Health Insurance and Labour Market Studies (LISA), 24 Multi-Generation Register, 25 National Cancer Register, 26 and Cause of Death Register. 27The Stockholm-Gotland Quality Register for Breast Cancer (the Stockholm-Gotland Breast Cancer Register [1976-2007] and the National Quality Register for Breast Cancer [2008 onward]) includes all patients with breast cancer diagnosed in the Stockholm-Gotland region and provides detailed information on tumor characteristics and treatments, with 99.9% of incident breast cancer cases being recorded. 9,20The Prescribed Drug Register contains prescription records nationwide since July 2005, with 99.7% completeness. 21The Cause-of-Death Register (1952 onward) assesses the underlying cause of death according to strict rules by the patient's usual physician or the physician last seeing the patient before death and records a specific underlying cause of death for 96% of individuals. 27rough linkage between the Stockholm-Gotland Quality Register for Breast Cancer and the Prescribed Drug Register, we identified 13 053 women diagnosed with nonmetastatic and estrogen receptor-positive breast cancer at 40 to 74 years of age from July

Discontinuation of Adjuvant Hormone Therapy and Cardiovascular Therapy
Discontinuation of adjuvant hormone therapy and cardiovascular therapy was defined as failure to refill a prescription of the corresponding therapy within 6 months after a previous prescription. 3edish pharmacies dispense a maximum of 3 months of prescribed drugs

Breast Cancer Mortality and Cardiovascular Disease-Specific Mortality
The main underlying cause of death was used to identify death due to breast cancer through ICD-10 code C50 and death due to cardiovascular disease through ICD-10 codes I00 to I99 in the Cause-of-Death Register.Patients were followed up until death, emigration, or end of the study period (December 31, 2019; last date cause of death was available), whichever came first.The median follow-up time was 6.5 years.

Factors Associated With Discontinuation of Adjuvant Hormone Therapy and Cardiovascular Therapy
Information on divorce and income at diagnosis was retrieved from the LISA register. 24

Discussion
Our findings suggest that patients with breast cancer who discontinued adjuvant hormone therapy were also more likely to discontinue their cardiovascular therapy.In particular, patients with breast cancer tended to discontinue these 2 therapies at the same time.Moreover, those who discontinued adjuvant hormone therapy were at a higher mortality risk due not only to breast cancer but also cardiovascular disease.
Our findings further suggest that more than 2 in 5 patients with breast cancer concurrently used cardiovascular therapy and adjuvant hormone therapy.Many patients with breast cancer, especially elderly patients, present with multiple chronic diseases and concomitantly use several long-term medications. 30Elderly patients now account for most patients who survive breast cancer, and the number is projected to double by 2040. 31,32Such a large number highlights the clinical importance of monitoring use of cardiovascular medications among patients with breast cancer.
We also found that among concomitant users of adjuvant hormone therapy and cardiovascular therapy, the discontinuation of these 2 therapies tends to happen concomitantly.More than 1 in 9 patients with breast cancer discontinued their cardiovascular therapy when they discontinued their adjuvant hormone therapy.This finding suggests that, once oncologists detected discontinuation of adjuvant hormone therapy, they should ask patients about their adherence to cardiovascular therapy.
That is, oncologists should shift from single-to multiple-disease focus to prevent discontinuation of treatment by patients with breast cancer, which is likely to be achieved through an extended cardiooncology approach and cooperation.
In addition to concomitant discontinuation, we found that patients who discontinued adjuvant hormone therapy were also more likely to discontinue cardiovascular therapy in other time periods, both before and after adjuvant hormone therapy discontinuation.This result is consistent with results of a previous study 33,34 showing that nonadherence to cardiovascular therapy before breast cancer diagnosis is associated with nonadherence to adjuvant hormone therapy.
As expected, we found that adjuvant hormone therapy discontinuation was associated with an increased mortality not only due to breast cancer but also cardiovascular disease, after adjusting for tumor characteristics, treatments, prediagnosis major cardiovascular event, and prediagnosis use of cardiovascular therapy.Our result suggests that the concomitant discontinuation is highly likely to translate to higher mortality.This aligns with previous studies 35,36 indicating that nonadherence to  Furthermore, while the association with cause-specific mortality did not significantly differ by hormone therapy type, it is important for future studies with a larger sample size to replicate this analysis considering the difference in efficacy and adverse effects profile between tamoxifen and aromatase inhibitors. 37,38scontinuation of adjuvant hormone therapy and cardiovascular therapy may share common risk factors and underlying mechanisms.In our study, we found that nonparticipation in mammography screening, divorce, low income, and use of symptom-relieving drugs are associated with both discontinuation of adjuvant hormone therapy and discontinuation of cardiovascular therapy.These shared factors may serve as potential targets for interventions that may simultaneously improve adherence to more than 1 life-saving medication, thus having greater costeffectiveness.For example, we found that patients who used symptom-relieving drugs were at higher risk of discontinuing both therapies, suggesting that only prescribing symptom-relieving drugs seems to be not enough to prevent therapy discontinuation once patients have therapyrelated symptoms.Other additional interventions, such as better clinician-patient communication, are needed to improve adherence for both adjuvant hormone therapy and cardiovascular therapy.
Despite that, we also identified several unique factors specifically associated with discontinuation of adjuvant hormone therapy or discontinuation of cardiovascular therapy.For example, women with a family history of breast cancer were less likely to discontinue their adjuvant hormone therapy, but no difference was found for discontinuation of cardiovascular therapy.These

Limitations
This study has several limitations.First, using prescription data from the register may underestimate the discontinuation rate, because prescription refill does not necessarily guarantee that a patient used the drug.Second, misclassifying patients who used aspirin as nonusers is possible because the Swedish Prescribed Drug Register does not capture over-the-counter medications.However, given the fact that low-dose and/or long-term use of aspirin is prescribed, we anticipate this misclassification is likely minimal.Third, although we censored patients at disease progress of both breast cancer and cardiovascular disease and adjusted for Charlson Comorbidity Index score at diagnosis and cancer treatments, we cannot rule out the possibility that the concomitant discontinuation of adjuvant hormone therapy and cardiovascular therapy could have been a result of clinical decision-making.While this may partially contribute to the observed association in our study, it is unlikely to substantially account for all associations we have reported.Fourth, the study did not have sufficient power to conduct a stratification analysis by timing of initiating cardiovascular therapy due to limited sample size for patients who initiated cardiovascular therapy after breast cancer diagnosis.Fifth, caution is needed when generalizing our findings to other countries, since our study population included only patients with breast cancer diagnosed in Sweden.

Conclusions
In this cohort study of patients with breast cancer concomitantly using adjuvant hormone therapy

Figure 2
Figure 2 shows factors associated with discontinuation of adjuvant hormone therapy and cardiovascular therapy.Shared risk factors included nonparticipation in mammography screening, low income, divorce, and use of symptom-relieving drugs including analgesics, gastrointestinal tract drugs, antidepressants, and hypnotics and/or sedatives.Unshared risk factors included age at diagnosis, family history of breast cancer, family history of major cardiovascular events, use of hormone replacement therapy, baseline Charlson Comorbidity Index score, and prediagnosis major cardiovascular events.

Figure 1 .
Figure 1.Cause-Specific Mortality by Discontinuation of Adjuvant Hormone Therapy Among Patients With Breast Cancer

Figure 2 .
Figure 2. Factors Associated With Discontinuation of Cardiovascular Therapy and Adjuvant Hormone Therapy Among Patients With Breast Cancer

a
Categorized as low and high using the median as a cutoff.b Defined as having a diagnosis of ischemic heart diseases (International Statistical Classification of Diseases and Related Health Problems, Tenth Revision [ICD-10] codes I20-I25) or cerebrovascular diseases (ICD-10 codes I60-I69) before breast cancer diagnosis.c Defined as having a female first-degree relative with breast cancer before index patient's breast cancer diagnosis.d Defined as having a first-degree relative with ischemic heart diseases (ICD-10 codes I20-I25) or cerebrovascular diseases (ICD-10 codes I60-I69) before index patient's breast cancer diagnosis.eAnalysis was restricted to postmenopausal women.
To further investigate the association between discontinuation of adjuvant hormone therapy and cardiovascular therapy, we used a matched cohort.For each patient who discontinued adjuvant hormone therapy, we randomly sampled 1 patient who continued the therapy at the first patient's discontinuation date and used the same cardiovascular medication.Those who discontinued and continued medications were individually matched on year of breast cancer diagnosis and age at diagnosis (±3 years).We used χ 2 tests to compare baseline characteristics between those who continued and discontinued adjuvant hormone therapy.Poisson regression with clustered sandwich estimator was used to estimate the association, adjusting for matching variables, tumor size, lymph node status, tumor grade, progesterone receptor status, surgery type, chemotherapy, radiotherapy, baseline hormone therapy type, prediagnosis use of cardiovascular therapy, prediagnosis major cardiovascular event, and Charlson Comorbidity Index score at diagnosis.This analysis was stratified by 4 time bands in relation to adjuvant hormone therapy discontinuation: −12 to −3 months, −3 months, 3 months, and 3 to 12 months.Kaplan-Meier curves were further constructed for breast cancer-specific and cardiovascular disease-specific mortality for those who discontinued adjuvant hormone therapy and matched patients who continued therapy.Marginal Cox proportional hazards regression models with clustered sandwich estimator were used to compare cause-specific mortality between patients who discontinued and matched patients who continued therapy, adjusting for matching variables, tumor size, lymph node status, tumor grade, progesterone receptor status, surgery type, chemotherapy, radiotherapy, baseline hormone therapy type, prediagnosis use of cardiovascular therapy, prediagnosis major cardiovascular event, and Charlson Comorbidity Index score at diagnosis.All the aforementioned analyses were stratified by baseline type of adjuvant hormone therapy.Finally, we identified shared and unshared factors associated with discontinuation of adjuvant hormone therapy and cardiovascular therapy among patients who concomitantly used these 2 therapies.Marginal Cox proportional hazards regression models with clustered sandwich estimator were used, adjusting for age at breast cancer diagnosis, diagnosis calendar period, tumor size, lymph node status, tumor grade, progesterone receptor status, surgery type, chemotherapy, radiotherapy, baseline hormone therapy type, prediagnosis use of cardiovascular therapy, prediagnosis major cardiovascular event, and Charlson Comorbidity Index score at diagnosis.All analyses were performed using SAS, version 9.4 (SAS Institute Inc); Stata, version 17.0 (StataCorp LLC); or R, version 1.4.1106(R Project for Statistical Computing) at a 2-tailed α = .05.

Table 1 .
Baseline Characteristics of Women Diagnosed With Breast Cancer Who Concomitantly Used Cardiovascular Therapy in Stockholm, Sweden, 2005 to 2020 a Discontinuation of Cardiovascular Therapy and Hormone Therapy for Breast Cancer ). Table1shows baseline characteristics, comorbidities, clinical features, and treatment information by discontinuation of adjuvant hormone therapy in the matched cohort.Most variables presented in Table1had a low level of missingness (<5%) and the missingness was not associated with adjuvant hormone therapy discontinuation, except for tumor grade, with data missing for 395 patients (7.2%).Among women who concomitantly used both therapies, the 5-year cumulative discontinuation rate was 42.1% for adjuvant hormone therapy and 47.6% for cardiovascular therapy.JAMA Network Open | OncologyJAMA Network Open.2023;6(7):e2323752.doi:10.1001/jamanetworkopen.2023.23752(Reprinted) July 17, 2023 5/13 Downloaded from jamanetwork.comby Linkopings Universitet Liu Periodica Online user on 03/01/2024 Concomitant Discontinuation

of Cardiovascular Therapy and Adjuvant Hormone Therapy
statins, and 42 of 256[16.4%]for aspirin).The discontinuation rate of cardiovascular therapy was the highest at the time of discontinuing adjuvant hormone therapy compared with other periods of follow-up, with an indcidence rate ratio (IRR) of 197.63 (95% CI, 169.11-230.97).Stratified analyses by baseline type of adjuvant hormone therapy yielded consistent results, with IRRs of 199.35 (95% CI,

Table 1 .
Baseline Characteristics of Women Diagnosed With Breast Cancer Who Concomitantly Used Cardiovascular Therapy in Stockholm, Sweden, 2005 to 2020 a (continued) d Categorized as low and high using the median as a cutoff.eDefinedashaving a female first-degree relative with breast cancer before the index patient's breast cancer diagnosis.Analysis was restricted to patients who were registered in the Swedish Multi-Generation Register.fDefinedas having a first-degree relative with g Analysis was restricted to postmenopausal women.hDefined as at least 1 prescription of cardiovascular

Table 2 .
IRR of Discontinuing Cardiovascular Therapy Before and After Discontinuation of Adjuvant Hormone Therapy in Patients With Breast Cancer a cardiovascular comorbidities, death, emigration, completion of 5-year adjuvant hormone therapy, or end of the study period (August 31, 2020), whichever came first.bAdjusted for matching variables, tumor size, lymph node status, tumor grade, progesterone receptor status, surgery type, chemotherapy, radiotherapy, baseline hormone therapy type, prediagnosis use of cardiovascular therapy, prediagnosis major cardiovascular event, prediagnosis menopausal symptoms, and Charlson Comorbidity Index at diagnosis.
Patients who discontinued adjuvant hormone therapy were 1:1 matched to those who continued therapy by year of breast cancer diagnosis, age at diagnosis (±3 years), and type of cardiovascular therapy.Patients were followed up until death, emigration, or end of the study period (December 31, 2019), whichever came first.Hazard ratios (HRs) were adjusted for matching variables, tumor size, lymph node status, tumor grade, progesterone receptor status, surgery type, chemotherapy, radiotherapy, baseline hormone therapy type, prediagnosis use of cardiovascular therapy, prediagnosis major cardiovascular event, and Charlson Comorbidity Index score at diagnosis.cardiovascular medications following breast cancer diagnosis is associated with a higher level of low-density lipoprotein and an increased risk of a cardiac event.To potentially prevent these unnecessary deaths, interventions focusing on multiple medications adherence are urgently needed.
Shared factors were statistically significant for discontinuation of both therapies with same direction of association; unshared factors were statistically significant only for discontinuation of one of the therapies.Hazard ratios (HRs) were adjusted for age at diagnosis, calendar period of diagnosis, tumor size, lymph node status, tumor grade, progesterone receptor status, surgery type, chemotherapy, radiotherapy, baseline hormone therapy type, prediagnosis use of cardiovascular therapy, prediagnosis major cardiovascular event, and Charlson Comorbidity Index score at diagnosis.
Discontinuation of Cardiovascular Therapy and Hormone Therapy for Breast Cancer unique factors suggest that discontinuation of different therapies also has its own unique mechanism and thus should be investigated separately to build a full picture of discontinuation of specific therapy.
JAMA Network Open.2023;6(7):e2323752.doi:10.1001/jamanetworkopen.2023.23752(Reprinted) July 17, 2023 9/13 Downloaded from jamanetwork.comby Linkopings Universitet Liu Periodica Online user on 03/01/2024 and cardiovascular therapy, our results call for extra vigilance for concomitant discontinuation of adjuvant hormone therapy and cardiovascular therapy in clinical practice.Shared risk factors identified in our study may help to develop targeted interventions to simultaneously improve adherence to both therapies and subsequently improve overall survival among patients with breast cancer.37. Early Breast Cancer Trialists' Collaborative Group (EBCTCG).Aromatase inhibitors versus tamoxifen in early breast cancer: patient-level meta-analysis of the randomised trials.Lancet.2015;386(10001):1341-1352. doi:10.1016/S0140-6736(15)61074-1 38.Perez EA.Safety profiles of tamoxifen and the aromatase inhibitors in adjuvant therapy of hormoneresponsive early breast cancer.Ann Oncol.2007;18(suppl 8):viii26-35.doi:10.1093/annonc/mdm263Incidence Rate Ratio and 95% CI of Discontinuing Cardiovascular Therapy Before and After Discontinuation of Adjuvant Hormone Therapy in Breast Cancer Patients, Stratified by Baseline Type of Adjuvant Hormone Therapy eTable 2. Cause-Specific Mortality by Discontinuation of Adjuvant Hormone Therapy in Breast Cancer Patients, Stratified by Baseline Type of Adjuvant Hormone Therapy