Cardiovascular Impact of Near Complete Estrogen Deprivation in Premenopausal Women with Breast Cancer: The CROWN Study

Abstract

Survival with operable breast cancer has improved markedly in recent decades, however, treatment-related cardiovascular toxicities threaten to offset these gains. Ovarian function suppression paired with aromatase inhibition, for premenopausal women with hormone receptor (HR)-positive breast cancer, is a newer widely adopted therapy with the potential for significant long-term cardiovascular toxicity. Abrupt estrogen deprivation for non-cancer reasons is associated with accelerated coronary artery disease. Women with breast cancer treated with aromatase inhibition in addition to ovarian function suppression experience a dual hit with regards to estrogen exposure. The CaRdiac Outcomes With Near-complete estrogen deprivation (CROWN) study seeks to understand the early, subclinical natural history of cardiovascular compromise in young women undergoing near-complete estrogen deprivation (NCED) therapy. It is critical to understand the early subclinical development of cardiovascular disease to identify a window for therapeutic intervention before overt cardiovascular events occur. This three-site regional study (Atrium Health Wake Forest, Duke, and Virginia Commonwealth University) uses serial stress cardiac magnetic resonance (CMR) imaging and cardiac computed tomography angiography (CCTA) obtained during the initial two years of NCED therapy to study myocardial prefusion reserve (MPR), large cardiovascular vessel changes, left ventricular function, and other cardiovascular parameters. The CROWN cohort will consist of 90 premenopausal women with breast cancer, 67 with HR-positive disease receiving NCED and 23 comparators with HR-negative disease. Participants will undergo three annual CMR scans and two CCTA scans during the two-year study period. After initial activation hurdles, accrual has been brisk, and the study is expected to complete accrual in December 2024. Efforts are in place to encourage participant retention with the study primary outcome, change in MPR between the two groups, to be reported in 2026–2027. The results of this study will enable premenopausal women with breast cancer to balance the health burdens of cancer at a young age and treatment-related cardiovascular morbidity. Finally, the tools developed here can be utilized to study cardiovascular risk across a range of cancer types and cancer therapies with the ultimate goals of both developing generalizable risk stratification tools as well as validating interventions which prevent overt cardiovascular compromise.

Background

Survival after a diagnosis of breast cancer has improved markedly in recent decades.¹ This has been attributed to both wide-spread screening as well as therapeutic advances; however, the cardiovascular morbidity associated with some aspects of anti-neoplastic therapy threaten to offset breast cancer survival gains.^2–13 Here, we describe the CaRdiac Outcomes With Near-complete estrogen deprivation (CROWN) study (NCT05309655) which seeks to understand the subclinical evolution of cardiovascular toxicity associated with a newer, more effective therapeutic modality used for premenopausal women with hormone receptor (HR) positive breast cancer.

Management of HR-positive breast cancer in premenopausal women, a highly prevalent disease, remains a formidable challenge in oncology. In 2023, an estimated 297,700 women will be diagnosed with breast cancer in the US.¹ In 2022, 17% of women in the US diagnosed with breast cancer were age 49 years or younger and 3% were younger than 40 years of age.¹⁴ With approximately 80% of all breast cancers being HR-positive,¹⁴ each year tens of thousands of premenopausal women in the US, and many more globally, will face this diagnosis. Furthermore, in recent years the incidence of HR-positive disease in women <50 years has been increasing, in contrast to HR-negative breast cancer which has been stable or declining in incidence.¹⁵ In addition, younger women have inferior survival outcomes relative to postmenopausal women with HR-positive breast cancer.^16–20 Moreover, in this subtype of breast cancer, the risk of recurrence persists at a steady rate over decades.^{21, 22}

Several years ago, long-awaited clinical trials reported that rendering premenopausal women with HR-positive breast cancer menopausal, combined with aromatase inhibition, significantly improved survival outcomes in these women.^23–25 Near-complete estrogen deprivation (NCED) combines medical or surgical ovarian function suppression with aromatase inhibition which blocks peripheral conversion of androgens into estrogen. Guidelines in the United States now recommend NCED for premenopausal women with intermediate to high-risk HR-positive breast cancer.²⁶ Recent European guidelines suggest expanding the use of NCED to almost all premenopausal women with HR-positive breast cancer.²⁷

This therapeutic approach has been adopted into clinical practice, although the long-term health sequela from NCED remain unknown. This is a significant concern for these young women who are at risk for both breast cancer recurrence and treatment-related toxicity.^{28, 29} Among the most concerning potential adverse sequela of NCED are later cardiovascular events including myocardial infarction, need for coronary revascularization, or cardiac death. These may not become clinically apparent for many years, at which point irreversible injury may have occurred.³⁰ Notably, hypoestrogenemia in premenopausal women was strongly associated with coronary artery disease in the WISE study³¹ and other studies evaluating the impact of oophorectomy for non-cancer indications in premenopausal women.^32–34 In addition, aromatase inhibitor therapy prevents estrogen production elsewhere in the body³⁵ further decreasing estrogen beyond that seen with oophorectomy alone. Trials and meta-analyses assessing the impact of aromatase inhibitor therapy across the full range of patient age have been associated with increased cardiovascular risk.^36–39 Further, underlying cardiovascular disease is often an exclusion criteria for clinical trials leaving open the possibility that the true cardiovascular impact from these agents may be higher in the real world, as was recently reported in a recent population-based review.⁴⁰

The single-site pilot ESPRIT Study (Estrogen Suppression and Perfusion Reserve with Aromatase-Inhibitor Treatment in Premenopausal Women with Breast Cancer Study) (NCT03505736) investigated this question. ESPRIT enrolled 21 premenopausal women (16 Caucasian, 5 Black; median age 44.7 years), 14 with HR-positive breast cancer within 3 years of initiating NCED (median 8 months) and 7 with TNBC within 3 years of chemotherapy (median 10 months) for comparators. This study, which utilized paired non-contrasted cardiac magnetic resonance imaging (CMR) 3–6 months apart for each patient, found that global myocardial perfusion reactivity (MPR) to adenosine stress declined in women on NCED during the 3–6 month study interval which was both clinically and statistically significant from changes observed in comparators (p=0.02).⁴¹ Left ventricular function remained unchanged during the study interval for both groups (p>0.05). In ESPRIT, the cardiovascular stress tests identified two women with HR-positive breast cancer (14%) with abnormal results during stress imaging who were sent for further cardiovascular evaluations. There were no such findings for women in the comparator cohort. The CROWN Study, funded by the US National Heart Lung and Blood Institute, seeks to confirm and extend the findings of ESPRIT. This three-site, five-year study of women receiving NCED, paired with a comparison cohort of women with HR-negative breast cancer, uses serial reproducible assessments of cardiovascular dysfunction to understand the subclinical evolution of cardiovascular injury. Combining this cardiovascular assessment with biomarker and demographic correlates, the ultimate goal of this study is to develop tools to assess and mitigate cardiovascular risk for this at-risk population.

Methods

Study design and outcomes

The CROWN study utilizes novel, sophisticated imaging tools to acquire outcomes data, with the primary outcome of determining the 24-month difference in stress myocardial blood flow during adenosine stress CMR in premenopausal women treated with NCED for high-risk HR-positive breast cancer and in premenopausal women treated without NCED for HR-negative breast cancer. Additional secondary outcome measures include determining the 12-month difference in stress myocardial blood flow during adenosine stress CMR between the groups, 12-month and 24-month difference in aortic stiffness, the association of stress CMR myocardial blood flow with total coronary plaque burden from coronary computed tomography angiography (at baseline and 24-month difference) as well as difference in variability in these measures. Secondary objectives include developing predictive models using data measured in this study (i.e., study group, demographics, etc.) to predict the change in myocardial perfusion reserve at 12-months and again at 24-months. The approach to developing these predictive models will follow the approach we used in the ESPRIT⁴¹ study. Multiple exploratory outcomes are also embedded in the study.

Patient population

Eligible participants are identified in the oncology clinics of the three sites, Atrium Health Wake Forest Baptist (AHWFB), Virginia Commonwealth University (VCU) and Duke University. Identification and recruitment strategies include a focus on both minority populations and an approach targeting enriched cohorts identified through presentation to cancer centers disease-oriented teams. The disease-oriented team meeting as well as tumor board meetings are used to engage treating oncologists. Additionally, a dedicated institutional review board approved patient flyer describing CROWN was developed and is available for study team members to educate potential patients as they approach the eligibility window.

Given the disparities seen in both cancer and cardiovascular outcomes for historically underserved populations, CROWN represents an opportunity to address both health concerns and is actively seeking to enroll participants from minority communities. North Carolina and Virginia both have populations which are greater than 20% African American and approximately 33% non-white. Strategies to ensure accrual mirrors the catchment area population, include emphasizing the importance of this to the CROWN investigator team, harnessing existing cancer center resources, as well as site specific tools. For example, the AHWFB Office of Community Outreach and Engagement has a signature population health navigation program that currently focuses on Hispanic, African American, rural populations and adolescent and young adult (AYA) populations. Each underserved population has been assigned a non-nurse navigator who was strategically selected to provide culturally and linguistically concordant care, with specific skill sets for each community. The program helps provide education around the importance of clinical trials for patients they serve in addition to the financial and psycho-social support they may need to enroll and continue with the longitudinal trial.

Patients in both the HR-positive and HR-negative breast cancer groups are eligible within a narrow time window after completing any chemotherapy, surgery and radiation (Table I). Once registered, participants begin two years of serial imaging, laboratory and clinical assessments, as well as quality of life and exercise measurements (Figure 1, Table 2). Clinical variables including patient demographics, treatment information and cardiovascular risk factors at baseline and throughout the study period are also collected (Supplemental Table I). Patients will then be followed for up to an additional three years, tracking cardiovascular health and cancer outcomes. Participant accrual is anticipated to occur over a 30-month period (Figure 2). Data acquisition and analysis will be on-going throughout the study. Final acquisition of data on the primary endpoint, will be obtained and analyzable after the final patient completes Year 2 of study procedures.

Table I:

CROWN Study Inclusion and Exclusion Criteria

Inclusion Criteria
	Planned breast cancer treatment with NCED therapy that includes aromatase inhibitor therapy with medically or surgically induced menopause within 3 months of initiating NCED (HR-positive tumor) or, for the cohorts not receiving NCED therapy, within three months of planned chemotherapy, surgery or radiation.
	Women age ≤55 years of age who were premenopausal at the time of diagnosis of breast cancer.
	Women with human epidermal growth factor-2 (HER2) negative and women with human epidermal growth factor-2 (HER2) positive breast cancer are eligible.
	Treatment with CDK-inhibitor, PARP inhibitor, immunotherapy or biologic (non-chemotherapy) agent as part of anti-neoplastic treatment plan is allowed.
	Treatment with selective-estrogen receptor degrader rather than aromatase inhibitor is allowed.
	Diagnosed with Stage I-III breast cancer.
	Eastern Cooperative Group performance status of 0–2.
	Normal renal function.
	Patients with concurrent malignancies are eligible as long as therapies and disease course for these are reasonably expected to not impact cardiovascular function. (Examples of eligible malignancies include papillary/follicular thyroid cancer, basal cell carcinoma of the skin, squamous cell carcinoma of the skin, in-situ and early-stage cervical cancers, etc.).
	Patients with prior COVID-19 are eligible if they have recovered from the illness and are free of COVID-related symptoms other than allowable persistent symptoms: loss of taste and smell and/or grade 1 fatigue.
	Ability to understand and the willingness to sign an IRB-approved informed consent document.
Exclusion Criteria
	History of allergic reactions attributed to compounds of similar chemical or biologic composition to adenosine.
	Active asthma or COPD currently requiring medications or active wheezing.
	Contraindication to MRI such as some breast expanders, ferromagnetic cerebral aneurysm clips, pacemakers, defibrillators or other implanted electronic devices.
	Uncontrolled intercurrent illness including, but not limited to ongoing or active infection, symptomatic congestive heart failure, unstable angina pectoris, cardiac arrhythmia, or psychiatric illness/social situations that would limit compliance with study requirements.
	Pregnant women are excluded from this study. Because some methods of birth control are not 100% reliable, a pregnancy test is required, unless the patient has undergone either a bilateral oophorectomy, hysterectomy or both.
	Coronary revascularization in the past 6 months or known severe multi-vessel CAD previously determined to be not amendable to mechanical intervention.
	Ongoing, unrelieved symptoms thought to represent cardiac ischemia.
	Allergy or prior sensitivity to gadolinium or other contrasting agents or their excipients.
	Men with breast cancer.

Abbreviations: NCED, near-complete estrogen deprivation; HR, hormone receptor; CDK, cyclin-dependent kinase; PARP, poly-ADP ribose polymerase; COPD, chronic obstructive pulmonary disease; MRI, magnetic resonance imaging; CAD, coronary artery disease.

Figure 1. Study Schema for CROWN.

Table II:

CROWN Study Procedures

	Study Time Point
	Baseline	Year 1	Year 2	Years 3–5
Stress CMR	✓	✓	✓
CT Angiogram	✓		✓
Electrocardiogram	✓	✓	✓
Study Labs	✓	✓	✓
Health Status		✓	✓	✓
IPAQ Survey (quarterly)	✓	✓	✓
PROMIS Global Health Survey	✓	✓	✓	✓

Abbreviations: CMR, cardiovascular magnetic resonance imaging; CT, computed tomography, IPAQ, International Physical Activity Questionnaire, PROMIS, Patient-Reported Outcomes Measurement Information System.

Figure 2. CROWN Study Timeline.

Participant Retention

Given the study duration, participant retention will be a vital component of study success. Multiple strategies have been implemented to maximize participant retention. These include (1) careful patient selection, with a focus on patients not likely to leave the area during the study period, (2) use of cognitive/behavioral management strategies to structure a positive testing environment (study staff regularly review the participants’ data related to follow-up attendance and identify any who need additional reminders, cues to action and/or in-person meetings), (3) incentives ($50 per participant/imaging to defray associated visit costs), and (4) use of telephone interviews if in-person visits are not feasible. The investigator team has also sought to help CROWN participants appreciate the contribution they are making to advancing cardiovascular health in the breast cancer community and to similarly feel that they are part of a larger CROWN and cardiovascular trial community. Toward this end, CROWN participants receive quarterly IRB-approved newsletters with study updates and broader information on cardiovascular health. Participants also receive study related items of limited monetary value (i.e., magnets). A patient advocate has also been engaged for regular input regarding study design and conduct. While the breast cancer community is a diverse group, many of these patients are particularly motivated around issues of cardiovascular health in cancer survivorship.

Study endpoints

The CROWN Study brings together highly diverse expertise in cardiology, medical oncology, and cardiovascular imaging and then utilizes sophisticated, quantitative imaging techniques to identify subclinical decrements when an interventional window may exist early in survivorship. Each participant will complete 2 imaging studies within 60 days of registration: CMR and coronary computed tomography angiography (CCTA). Each imaging assessment will be blinded to patient identifiers and the corresponding imaging results with 15% receiving double reading to calculate intra- and inter-observer variability. We will assess both visual quality and quantitative drift quarterly to monitor for longitudinal drift and bias. Quantitative myocardial perfusion imaging will measure the myocardial blood flow (MBF, ml/g/min) following intravenous adenosine administration in each voxel at rest and stress states in 3 short axis slices using established acquisition imaging parameters.⁴² Quantitative MBF is more objective than visual assessments and semi-quantitative techniques^{43, 44} and can better detect globally reduced MBF in the setting of multi-vessel disease or microvascular dysfunction.^{42, 45} For MBF analysis, arterial input function model profiles, electrocardiogram and respiratory gating success, and image quality will be reviewed for each acquisition. All contouring and quantitation are performed in Circle CVI42 software (v5.16, Calgary, CAN; Figure 3).

Figure 3. Case CROWN Quantitative Perfusion CMR Images.

A quad bolus adenosine stress CMR perfusion examination is performed at Baseline, Year 1, and Year 2 study visits. Motion corrected perfusion images are captured at rest (A, left column) and stress (A, right column) at the basal, mid, and apical short axis views. Post-processing is performed to quantify the myocardial blood flow (MBF) at rest (B, left column) and stress (B, right column). Finally, polar plots are displayed in C showing the rest MBF, stress MBF, rest relative MBF, stress relative MBF, relative myocardial perfusion reserve (MPR), and MPR corresponding to each American Heart Association 16-segment. All analyses will be performed offline in Circle CVI42 (v5.16).

CCTA can identify obstructive coronary artery disease as well as nonobstructive atherosclerotic plaque extent and composition and is superior for risk stratification in women.⁴⁶ Quantitative analyses will be performed in accordance with established protocols^47–49 on atherosclerotic plaque for those with one or more segments of nonobstructive or obstructive disease. The total plaque, calcified plaque, noncalcified plaque, and low-attenuation (<30 Hounsfield units) plaque volumes will be measured (Figure 4). For women without evidence of disease, the quantified volume and burden will be expressed as a value of zero. Finally, stenoses will be visually assessed and scored as normal, nonobstructive, or obstructive. To address radiation risk/benefit, CCTA scans will be collected with prospective cardiac gating at accredited centers using lowest dose protocols. Additionally, CCTA scans will only be collected at two of the three study visits.

Figure 4: Case CROWN CT Angiography Images.

Contrasted computed tomographic angiography images are acquired at Baseline and Year 2 study visits and processed in Circle CVI42 software (v5.16) for calcium score and plaque burden (calcified, non-calcified, and low-density plaques). Left anterior descending (LAD) coronary artery shown in three cross-sections (Panels A-C) with red centerline. Pixels identified as having plaque are color coded in an overlay in Panels A-C according to calcified plaque (CP, yellow), non-calcified plaque (NCP, orange), and low-density NCP (LD-NCP, red). A 3D model shown in Panel D highlights processed coronary arteries and branches in blue with the active LAD artery highlighted in red. A curved reconstruction of the LAD with plaque overlays is shown in Panel E with three cross sections (insets) at dashed line locations. Final quantitative plaque analysis and 3D model inset are shown in Panel F.

To assess the aortic stiffness, a 2D phase-contrast scan of the ascending and descending thoracic aorta in a para-axial view at the plane of the pulmonary artery will be performed for calculation of the pulse wave velocity (PWV) (Figure 5). A cine series will be acquired in a matching imaging plane for calculation of aortic distensibility. PWV will be calculated as previously described by our group.^50–52 Briefly, the ascending and descending aortic luminal areas are contoured and the mean velocity for each phase is extracted. The PWV is then calculated as transit distance, drawn midline on the aorta in a parasagittal view, divided by the transit time difference between the velocity curves. For calculation of the aortic distensibility, we will utilize published methods^50–53 to measure the aortic lumen maximum and minimum areas in the ascending and descending aorta and calculate the distensibility given the blood pressure taken during image acquisition. These analyses will be performed in either Circle CVI42 software (v5.16, Calgary, CAN) or custom-written MATLAB software (The Mathworks, Natick, MA) which has been utilized in multiple cardio-oncology studies.^50–53 Additional steady state free precession cine imaging of the short- and long-axis views of the left ventricle (LV) will also be collected for the measurement of LV end-systolic and end-diastolic volumes, LV mass, LV ejection fraction, and LV strain (Figure 6).

Figure 5: Case CROWN Aorta CMR Images.

CMR images related to the aortic structure and function are collected at Baseline, Year 1, and Year 2 study visits. The aortic assessments include aortic distensibility (A), wall thickness (B), and pulse wave velocity (C and D). All analyses will be performed offline in either Circle CVI42 (v5.16) or custom-written MATLAB software.

Figure 6: Case CROWN Left Ventricular (LV) Cine CMR Images.

The CMR examinations at Baseline, Year 1, and Year 2 study visits also captures steady state free precession cine images to assess LV structure and function. Two-chamber (A), four-chamber (B), and three-chamber (C) long axis cines are acquired as well as a short-axis cine stack (D) for assessment of LV volumes, mass, ejection fraction, and strain. All analyses will be performed offline in Circle CVI42 (v5.16).

Statistical Methods

The primary statistical analysis for aims 1 and 2 of this study will assess the difference between and within patient groups, premenopausal women with HR-positive disease treated with NCED compared to premenopausal women with HR-negative disease not treated with NCED. With data collected longitudinally over 24 months (0, 12, and 24 months), we will use a longitudinal linear model framework to model each outcome of interest (e.g. MPR) and assess within- and between-group effects over time. Our models will include fixed effects for the patient group and time point at which the outcome measure is assessed. Additional baseline patient risk factors and comorbidities may be included as fixed effects in an adjusted analysis. Repeated measurements over time will be nested within patients via random intercepts. An interaction term between group and time will be included to first test the hypothesis that there is a differential rate of change in the outcome being assessed between patient groups. In the presence of a significant interaction, contrasts will be used to compare patient groups at time-points of interest, including the primary follow-up time point at 24 months. Separately, within-group changes over time for each patient group will be assessed separately in linear mixed models, including baseline characteristics and time as fixed effects. Covariates included in each model will be assessed for potential interaction effects with patient group based on specified hypotheses of scientific interest and known risk factors (e.g. age).

For the third aim of the study, we will assess performance of regression models for predicting our primary outcome of interest, MPR, based on baseline patient characteristics. Through this methodology, we will identify baseline patient variables that best yield a select, yet informative predictive model for follow-up MPR measures. Upon completion of the study, we can develop predictive models to determine what characteristics best predict perfusion deficit in participants at their final visit. We can fit a general linear model that examines baseline characteristics to predict the level of perfusion deficit. Using this approach, we will also use a 100-fold cross validation procedure to examine the performance of the prediction model. In addition, we will calculate the ACC/AHA pooled cohort cardiovascular disease risk for all patients to calculate base (non-cancer therapy related) risk for cardiovascular disease.⁵⁴

In total, 90 women will be recruited, 67 premenopausal HR-positive women treated with NCED and 23 HR-negative women not treated with NCED. Allowing for a 10% loss to follow-up, with 80 patients total (60 HR-positive NCED vs 20 HR-negative), this study achieves 80% power at a 5% significance level to detect a between group difference in MPR of 2.8% assuming a standard deviation of 4% based on preliminary data and assuming a correlation of 0.3 between baseline and follow-up measurements of MPR. More generally, this sample size yields a detectable effect size of 0.69 standard deviations with 80% power.

Study status

Initial study start-up procedures including site subaward contracting, institutional review board approvals, and activation took longer than projected and six participants were accrued from August 2022 to March 2023. However, accrual has now increased markedly and study completion on or near the original timeline is anticipated; as of submission, a total of n=16 have been accrued to the study. While still early, the study appears to be accruing minority populations at target rates (25%; 20 participants) with 2 Black and 2 Hispanic participants on CROWN.

Discussion

The CROWN Study seeks to understand the natural history of women embarking on NCED to lower the risk of breast cancer recurrence. In studying myocardial blood flow during stress, as well as coronary plaque serially in these premenopausal patients, we seek to identify cardiovascular pathology subclinically before cardiovascular events occur. Complete results of CROWN are anticipated in 2026–2027, though some information (i.e., differences in baseline to Year 1) will be available earlier. As sophisticated cardiovascular imaging is unlikely to become a routine screening tool, an important outcome of CROWN will be to identify prognostic models or simpler, less expensive tools that can predict which individuals of the tens of thousands of women who start this therapy each year will be at increased risk of cardiovascular compromise. This type of outcome prediction could be used to inform preventive medical or lifestyle strategies or it could be used to guide further cardiovascular risk assessment or surveillance strategies. A similar tool used in this space is a bone density scan. Women on NCED are at risk for significant bone loss during therapy, though while some experience significant decline others have minimal bone impact from this therapy. A bone density scan assesses bone health before a fracture occurs and allows for treatment to mitigate this risk.

Early challenges for the CROWN Study have included activation and accrual ramp-up issues. These have been mitigated by a series of interventions, with accrual improving markedly in recent months. These interventions include development and use of patient flyers explaining the trial, opening of the study at cancer center satellite clinics which see large volumes of patients, and engaging clinical teams at these sites. Additionally, issues on data transfer during study conduct have arisen and have been resolved. This study requires complex cardiovascular imaging at multiple sites with significant data to not only be transferred to a single site for core lab analysis, but also to have imaging results with standardized imaging protocols align across sites. Participant retention will become an important study metric as the study progresses and while processes are in place to maximize participant retention, unanticipated issues could arise.

Further open questions still remain with regards to the cardiovascular toxicity of anti-neoplastic treatments for premenopausal women with breast cancer. CROWN could offer important clues here. For example, does NCED differentially impact patients who have received anthracyclines or radiation therapy, particularly left-sided radiation therapy? A handful of these patients will see newer breast cancer treatments such as cyclin-dependent kinase (CDK) 4/6 inhibitors, poly-ADP ribose polymerase (PARP) inhibitors, or selective estrogen receptor degraders. Detailed cardiovascular imaging over time in these women could offer signals of sustained cardiovascular health, which would be reassuring, or alternatively suggest subclinical changes that could require further investigation. The CROWN Study will provide vital information for this population of younger women who face the dual risks of cancer recurrence and cardiovascular complications from therapy. Future opportunities to learn from the CROWN cohort, and other similar cohorts, will also build our knowledge in this space.

Importantly, successful completion of CROWN demonstrates both feasibility and efficacy in identifying early cardiovascular decline in a relatively young at-risk population. It will also provide a framework for future studies seeking to assess subclinical cardiovascular impact of new therapeutic options. Such a framework would have applications in other breast cancer therapies, in other cancer types, and also beyond cancer treatment, such as in estrogen deprivation therapies for non-cancer endocrinopathies and in transgender health. Additionally, future work may consider alternative non-invasive measurements of cardiovascular disease such as carotid plaque imaging or retinal scans and the inclusion of assessments of extra-cardiac vascular bed effects in tissues such as the kidneys and/or brain. Funding for prolonged follow-up of the CROWN cohort will be sought as cardiovascular outcomes related to early atherosclerotic deposits and myocardial perfusion deficits may occur many years after treatment. Additionally, a pending National Cancer Institute funded Phase III trial, NRG-BR009: A Phase III Adjuvant Trial Evaluating the Addition of Adjuvant Chemotherapy to Ovarian Function Suppression plus Endocrine Therapy in Premenopausal Patients with pN0–1, ER-Positive/HER2-Negative Breast Cancer and an Oncotype Recurrence Score ≤ 25 (OFSET), will examine whether NCED can replace chemotherapy in some women. This trial, which is anticipated to activate in 2023, will enroll over 4,000 women. OFSET will provide a tremendous opportunity to understand the cardiovascular health of women on this therapy and to better understand the relative impact of chemotherapy.

Finally, while understanding risk and clinical variables associated with increased risk is an essential piece of this puzzle, ultimately interventional studies will be needed to fully understand which tools can mitigate risk before cardiovascular events occur. These therapies may already exist as cardioprotective options such as aspirin or statins, or new options may be studied. Collectively, this work will allow these women, who have many decades of life to protect, to safely remain on anti-neoplastic therapy and thereby lower both the burdens of cancer and cardiovascular treatment-related toxicity.