This cross-sectional study assesses whether treatment-induced cardiotoxicity is associated with long-term cardiopulmonary dysfunction in survivors of ERBB2-positive breast cancer.
Key Points
Question
What are the long-term cardiopulmonary consequences of treatment-induced cardiotoxicity in patients who survive ERBB2-positive breast cancer?
Findings
In this cross-sectional case-control study of 42 patients who survived breast cancer, those with prior cardiotoxicity had significantly lower left ventricular ejection fraction, global longitudinal strain, and peak oxygen consumption. Global longitudinal strain was the only echocardiographic parameter associated with peak oxygen consumption.
Meaning
These findings suggest that cardiotoxicity due to breast cancer therapies appears to be associated with significant long-term impairment of cardiopulmonary function, and interventions to prevent and mitigate decline in peak oxygen consumption among breast cancer survivors are warranted.
Abstract
Importance
Trastuzumab improves outcomes in patients with ERBB2-positive (formerly HER2) breast cancer but is associated with treatment-induced cardiotoxicity, most commonly manifest by an asymptomatic decline in left ventricular ejection fraction (LVEF). Little is known to date regarding the long-term effects of treatment-induced cardiotoxicity on cardiopulmonary function in patients who survive trastuzumab-treated breast cancer.
Objective
To determine whether treatment-induced cardiotoxicity recovers or is associated with long-term cardiopulmonary dysfunction in survivors of ERBB2-positive breast cancer.
Design, Setting, and Participants
This cross-sectional case-control study enrolled patients with nonmetastatic ERBB2-positive breast cancer after completion of trastuzumab-based therapy (median, 7.0 [interquartile range (IQR), 6.2-8.7] years after therapy) who met 1 of 2 criteria: (1) cardiotoxicity (TOX group) developed during trastuzumab treatment (ie, asymptomatic decrease of LVEF≥10% from baseline to <55% [n = 22]) or (2) no evidence of cardiotoxicity during trastuzumab treatment (NOTOX group [n = 20]). Patients were treated at the Memorial Sloan Kettering Cancer Center. Fifteen healthy control participants (HC group) were also enrolled for comparison purposes. All groups were frequency matched by age strata (<55, 55-64, and ≥65 years). Data were collected from September 9, 2016, to August 10, 2018, and analyzed from November 20, 2018, to August 12, 2019.
Main Outcomes and Measures
Speckle-tracking echocardiography and maximal cardiopulmonary exercise testing were performed to measure indices of left ventricular function (including LVEF and global longitudinal strain [GLS]) and peak oxygen consumption (peak VO2).
Results
A total of 57 participants (median age, 60.8 [IQR, 52.7-65.7] years) were included in the analysis. Overall, 38 of 42 patients with breast cancer (90%) were treated with anthracyclines before trastuzumab. Resting mean (SD) LVEF was significantly lower in the TOX group (56.9% [5.2%]) compared with the NOTOX (62.4% [4.0%]) and HC (65.3% [2.9%]) groups; similar results were found for GLS (TOX group, −17.8% [2.2%]; NOTOX group, −19.8% [2.2%]; HC group, −21.3% [1.8%]) (P < .001). Mean peak VO2 in the TOX group (22.9 [4.4] mL/kg/min) was 15% lower compared with the NOTOX group (27.0 [5.3] mL/kg/min; P = .03) and 25% lower compared with the HC group (30.5 [3.4] mL/ kg/min; P < .001). In patients with breast cancer, GLS was significantly associated with peak VO2 (β coefficient, −0.75; 95% CI, −1.32 to −0.18).
Conclusions and Relevance
Treatment-induced cardiotoxicity appears to be associated with long-term marked impairment of cardiopulmonary function and may contribute to increased risk of late-occurring cardiovascular disease in survivors of ERBB2-positive breast cancer.
Introduction
Trastuzumab-containing regimens for ERBB2-positive (OMIM 164870) (formerly HER2) breast cancer are associated with cardiotoxicity. Risk of cardiotoxicity is highest when trastuzumab is administered sequentially after anthracyclines, with development of a significant decline in left ventricular ejection fraction (LVEF) in as many as 25% of patients or symptomatic heart failure in 0.8% to 4.0% of patients.1,2,3,4 Although the prevailing perception is that cardiotoxicity related to trastuzumab is reversible, full recovery of left ventricular systolic function is not always achieved, despite interruption of therapy, and may be attributable to the synergistic cardiac impairment caused by sequential treatment with anthracyclines and trastuzumab.5 Reports from clinical practice suggest that cardiotoxicity can result in long-term impairment of left ventricular systolic function with increased long-term risk of heart failure in patients who survived breast cancer (hereinafter referred to as breast cancer survivors).6,7
Evidence of the long-term cardiac safety of trastuzumab among breast cancer survivors is based on studies that use resting LVEF assessments to identify patients with left ventricular systolic dysfunction.8 However, LVEF as a measure of left ventricular systolic function is limited by load dependency, which can lead to changes in left ventricular volumes that do not directly reflect a change in myocardial contractility and reliance on geometric assumptions for calculation of left ventricular volumes that is subject to technical error.9 Systemic chemotherapy for breast cancer treatment causes impairment of cardiopulmonary function that cannot be detected by resting LVEF assessment.10,11 Cardiorespiratory fitness (CRF), as measured by peak oxygen consumption (peak VO2), is an integrative assessment of global cardiovascular function that declines as much as 26% after exposure to various systemic cancer treatment regimens and may not recover despite treatment cessation.10,12 Emerging evidence indicates that poor peak VO2 is associated with a higher symptom burden (eg, exercise intolerance or lower health-related quality of life), increased risk of cardiovascular disease, and increased risk of death after a cancer diagnosis.13,14,15 Despite the significant prognostic value of CRF, little is known to date on the late effects of trastuzumab on CRF in breast cancer survivors.
The primary objectives of this study were to compare cardiac structure, function, and CRF in ERBB2-positive breast cancer survivors with a history of treatment-induced cardiotoxicity (TOX group), relative to trastuzumab-treated control participants without cardiotoxicity (NOTOX group) and age-matched healthy controls (HC group). The secondary objective was to explore the association between echocardiographic parameters of left ventricular function (at rest and after exercise) and CRF among ERBB2-positive breast cancer survivors.
Methods
Participants and Study Design
This single-center cross-sectional case-control study enrolled 42 women with a history of nonmetastatic ERBB2-positive breast cancer who completed trastuzumab-based therapy at Memorial Sloan Kettering Cancer Center, New York, New York, at least 2 years before study entry (eFigure in the Supplement). The breast cancer survivors met 1 of the following criteria: (1) history of treatment-induced cardiotoxicity, defined as an absolute decrease of at least 10% from baseline LVEF to less than 55% during the trastuzumab treatment period without symptoms of heart failure (TOX group [n = 22]), or (2) normal left ventricular systolic function throughout the breast cancer treatment period as defined by a normal LVEF (≥55%) and a maximum decline in LVEF of 5% or less during at least 3 LVEF assessments (ie, at baseline and a minimum of 2 follow-ups during trastuzumab treatment) (NOTOX group [n = 20]). Inclusion in the NOTOX group was restricted to patients with no more than a 5% decline in LVEF during trastuzumab-based therapy to select for patients with stable left ventricular systolic function during breast cancer treatment. Exclusion criteria consisted of (1) symptomatic heart failure during trastuzumab treatment; (2) diagnosis of recurrent or metastatic breast cancer; (3) treatment with additional cardiotoxic cancer therapy since completion of targeted therapy for ERBB2-positive breast cancer; and (4) contraindication to cardiopulmonary exercise testing (CPET). Eligible patients were identified through an institutional database and were frequency matched by age strata at the time of enrollment (<55, 55-64, and ≥65 years) such that the proportion of enrolled patients in each age stratum was balanced between the TOX and NOTOX groups.16 Fifteen women with no history of cancer or known cardiac disease and normal LVEF (≥55%) and frequency matched by age strata to the TOX and NOTOX groups were recruited for the HC group. Data were collected from September 9, 2016, to August 10, 2018. Written informed consent was obtained from all participants. The study was approved by the institutional review board at Memorial Sloan Kettering Cancer Center and followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
2-Dimensional, Doppler, and Strain Echocardiography
Resting 2-dimensional (2-D) and Doppler echocardiography were performed using an ultrasound scanner (Vivid E9; GE Medical Systems) in accordance with American Society of Echocardiography guidelines.17 Apical 2-, 3-, and 4-chamber views and short-axis views at the midpapillary level were acquired at a frame rate of 40 to 80 frames per second. Speckle-tracking strain analysis was performed offline (EchoPAC; GE Healthcare) to calculate peak systolic global longitudinal strain (GLS), global radial strain, and global circumferential strain as previously described.18 Interobserver and intraobserver variability for GLS, global radial strain, and global circumferential strain have previously been reported.19
Cardiopulmonary Function
Cardiopulmonary exercise testing was conducted using an electronic motorized treadmill with 12-lead electrocardiographic monitoring (Mac 5000; GE Healthcare) according to standard guidelines.20 Before initiating the CPET, a warm-up was performed wherein the starting treadmill speed was selected based on participant comfort (1.0-4.5 mph). Breath-by-breath (with a mean value calculated every 30 seconds) expired gases were analyzed continuously by a calibrated metabolic measurement system (TrueOne 2400; Parvo Medics). To stabilize gas measurement, 3 minutes of resting metabolic data were collected before exercise initiation. After stable resting metabolic values were achieved, participants began walking at a participant-specific designated speed at 0% grade for 2 minutes, and the speed and grade were increased every minute until exhaustion or a symptom-limited peak was achieved measured by peak VO2 measurement (in milliliters per kilogram per minute). All tests were completed within 8 to 12 minutes in line with CPET guidelines.20 During exercise, all participants were monitored continuously with 12-lead electrocardiography, oxyhemoglobin saturation was monitored continuously using a handheld pulse oximeter (BCI), and blood pressure was measured every 2 minutes. Rating of perceived exertion was evaluated at the end of each workload using the Borg scale. Acceptable test criteria for this assessment included any of the following: (1) a respiratory exchange ratio of at least 1.1, (2) attainment (±10 bpm) of age-predicted maximal heart rate, or (3) a rating of perceived exertion of at least 18 on the Borg scale. All VO2 data were calculated for the 30-second mean values, and the highest value achieved during the final 90 seconds of the test was considered peak. This protocol has previously been demonstrated to be safe and feasible in survivors of early-stage breast cancer.21
Immediately after exercise, participants were placed in the supine position and additional echocardiographic images were acquired in the apical 4-, 2-, and 3- chamber views. Cardiac index (at rest and after exercise) was calculated using 2-D pulsed-wave Doppler echocardiography as left ventricular stroke volume times heart rate (cardiac output), indexed to body surface area.22 Contractile reserve was calculated as the difference between the resting and postexercise cardiac index. Arterial-venous oxygen content difference (AVO2 difference), based on the Fick equation, was calculated as peak VO2 divided by cardiac output.
Statistical Analysis
Data were analyzed from November 20, 2018, to August 12, 2019. The sample size was calculated based on prior studies that have shown that GLS is a sensitive marker that can detect subclinical abnormalities in left ventricular systolic function23,24 and has significant prognostic value in populations with and without cancer.25,26 Assuming an SD of 2,21 enrollment of 20 participants in the TOX and NOTOX groups provided 88% power to detect a 2% absolute difference in GLS with a 2-sided type I error of 0.05.
Continuous data are summarized as mean and SD or median and interquartile range (IQR), as appropriate, and categorical measures as frequency and percentage. Differences in baseline characteristics were studied using Cochran-Mantel-Haenszel tests or linear models, both adjusted for the age strata used for matching. Differences in cardiopulmonary function between all groups were explored by analysis of variance. Linear regression was used to further investigate differences in resting (LVEF, GLS, global circumferential strain, and global radial strain) and postexercise (peak VO2, LVEF, contractile reserve, and AVO2 difference) parameters as the primary outcomes and study group (TOX vs NOTOX vs HC) as the primary variable associated with outcome, adjusting for covariates including age group, hypertension (yes or no), body mass index (calculated as weight in kilograms divided by the square of height in meter), and time since completion of targeted therapy for ERBB2-positive cancer as continuous variables. Univariate and multivariate linear regression were performed to identify covariates associated with peak VO2 using a stepwise approach with a threshold of 2-sided P < .05. All statistical analysis was performed in R, version 3.5.2 (R Foundation for Statistical Computing), and Stata, version 15.1 (StataCorp, LLC).
Results
Population
At the time of study enrollment, the median age among the 57 study participants was 60.8 (IQR, 52.7-65.7) years. Median time since completion of targeted therapy for ERBB2-positive cancer was 7.0 (IQR, 6.2-8.7) years. The prevalence of hypertension and hyperlipidemia was similar among the TOX, NOTOX, and HC groups. Median body mass index was higher in the TOX group (25.9 [IQR, 23.7-29.5]) compared with the NOTOX (24.6 [IQR, 22.1-29.3]) and HC groups (23.1 [IQR, 20.0-25.6]; P = .04). Baseline clinical characteristics of all participants are presented in Table 1.
Table 1. Participant Characteristicsa.
| Characteristic | TOX Group (n = 22) | NOTOX Group (n = 20) | HC Group (n = 15) | P Value |
|---|---|---|---|---|
| Age, median (IQR), yb | 60.5 (52.2-65.5) | 61.1 (50.8-65.8) | 60.9 (54.0-63.8) | NA |
| BMI, median (IQR)b | 25.9 (23.7-29.5) | 24.6 (22.1-29.3) | 23.3 (20.0-25.6) | .04 |
| Age at cancer treatment, median (IQR), y | 53.0 (44.0-57.0) | 52.0 (43.8-55.3) | NA | .10 |
| Race, No, (%) | ||||
| White | 16 (73) | 14 (70) | 9 (60) | .16 |
| Black | 2 (9) | 4 (20) | 0 | |
| Other/unknown | 4 (18) | 2 (10) | 6 (40) | |
| Invasive ductal carcinoma, No. (%) | 22 (100) | 20 (100) | NA | >.99 |
| Hormone receptor status, No. (%) | ||||
| Positive | 10 (45) | 13 (65) | NA | .29 |
| Negative | 12 (55) | 7 (35) | NA | |
| Cancer stage, No. (%) | ||||
| I | 10 (45) | 8 (40) | NA | .95 |
| II | 9 (41) | 9 (45) | NA | |
| III | 3 (14) | 3 (15) | NA | |
| Chemotherapy regimen | ||||
| Anthracycline-based | 20 (91) | 18 (90) | NA | .98 |
| Non–anthracycline-based | 2 (9) | 2 (10) | NA | |
| Cumulative doxorubicin-equivalent dose, median (IQR), mg/m2 | 240 (240-240) | 240 (240-240) | NA | .71 |
| Targeted ERBB2-regimen | ||||
| Trastuzumab | 20 (91) | 20 (100) | NA | .51 |
| Trastuzumab plus pertuzumab | 2 (9) | 0 | NA | |
| Cumulative trastuzumab dose, median (IQR), mg/kg | 102 (68-106) | 106 (100-108) | NA | .050 |
| Time since trastuzumab therapy completion, median (IQR), y | 6.8 (6.2-8.2) | 7.1 (6.3-9.0) | NA | .28 |
| Surgery | ||||
| Mastectomy | 10 (45) | 13 (65) | NA | .32 |
| Lumpectomy | 12 (55) | 7 (35) | NA | |
| Breast radiotherapy | ||||
| Right | 6 (27) | 6 (30) | NA | .98 |
| Left | 7 (32) | 6 (30) | NA | |
| None | 9 (41) | 8 (40) | NA | |
| Cardiovascular risk factorsb | ||||
| Hypertension | 4 (18) | 5 (25) | 1 (7) | .42 |
| Diabetes mellitus | 0 | 2 (10) | 0 | .20 |
| Hyperlipidemia | 6 (27) | 7 (35) | 3 (20) | .66 |
| Cardiac medicationsb | ||||
| β-Blocker | 4 (18) | 2 (10) | 0 | .20 |
| ACE inhibitor/ARB | 3 (14) | 2 (10) | 0 | .39 |
| Calcium-channel blocker | 0 | 3 (15) | 0 | .07 |
Abbreviations: ACE, angiotensin-converting enzyme; ARB, angiotensin receptor blocker; BMI, body mass index (calculated as weight in kilograms divided by square of height in meters); HC, healthy control; NA, not applicable.
Patients who developed cardiotoxicity during trastuzumab treatment constituted the TOX group; those with no evidence of cardiotoxicity during trastuzumab treatment, the NOTOX group.
Represents baseline characteristics at the time of study enrollment.
At the time of initial breast cancer diagnosis (before beginning chemotherapy), mean LVEF was similar between the TOX (64.5% [4.6%]) and NOTOX (65.4% [6.4%]; P = .58) groups. Hypertension and hyperlipidemia before breast cancer treatment were present in 1 (5%) and 3 (14%) patients in the TOX group, respectively, and 4 (20%) and 4 (20%) patients in the NOTOX group, respectively. Median body mass index was 25.0 (IQR, 23.7-29.4) in the TOX group and 25.1 (IQR, 21.8-27.5) in the NOTOX group. Overall, 38 of 42 participants in both breast cancer survivor groups (90%) were treated with an anthracycline-based regimen (median doxorubicin-equivalent dose, 240 [IQR, 240-240] mg/m2). During the trastuzumab treatment period, the nadir mean LVEF was 48.2% (4.1%) in the TOX group compared with 65.5% (5.9%) in the NOTOX group.
After diagnosis of cardiotoxicity, 14 patients had treatment interrupted or discontinued and 6 patients who had a nadir LVEF of greater than 50% had no treatment interruption. Two patients had received the final dose of trastuzumab at the time of LVEF decline. Five patients received a β-blocker, 7 received an angiotensin-converting enzyme inhibitor or angiotensin receptor blocker, and 13 received a cardiology consultation. The median cumulative trastuzumab dose received was 102 (IQR, 68-106) mg/kg in the TOX group and 106 (IQR, 100-108) mg/kg in the NOTOX group.
Left Ventricular Structure and Function
Resting echocardiographic measurements are shown in Table 2. Five patients in the TOX group but none in the NOTOX or HC groups had grade I diastolic dysfunction (P = .06). Echocardiographic parameters of left ventricular systolic function were lower in the breast cancer survivor groups (TOX and NOTOX) compared with the HC group in adjusted analyses. The mean LVEF was lower in the TOX group (56.9% [5.2%]) compared with the NOTOX group (62.4% [4.0%]; P < .001) and the HC group (65.3% [2.9%]; P < .001). Three participants in the TOX group but none in the NOTOX or HC groups had a LVEF of less than 53%. The mean GLS was worse in the TOX group (−17.8% [2.2%]) compared with the NOTOX group (−19.8% [2.2%]; P = .005) and the HC group (−21.3% [1.8%]; P < .001). Six participants in the TOX group but none in the NOTOX or HC groups had a GLS of less than −16%. Mean global circumferential strain and global radial strain were also reduced in the TOX group (−15.1% [3.9%] and 37.7% [16.9%], respectively) compared with the HC group (−19.6% [3.0%] and 57.3% [14.9%], respectively), but not the NOTOX group (−17.1% [2.2%] and 46.5% [21.1%], respectively).
Table 2. Resting Echocardiographic Measuresa.
| Characteristic | TOX Group (n = 22) | NOTOX Group (n = 20) | HC Group (n = 15) | P Valueb |
|---|---|---|---|---|
| IVSd, cm | 0.9 (0.2) | 0.9 (0.2) | 0.9 (0.1) | .26 |
| LVPWd, cm | 0.8 (0.1) | 0.8 (0.1) | 0.7 (0.1) | .22 |
| LVIDd, cm | 4.4 (0.4) | 4.3 (0.4) | 4.3 (0.3) | .84 |
| LVMI, kg/m2 | 69.8 (14.3) | 64.9 (11.8) | 65.7 (8.4) | .39 |
| RWT | 0.4 (0.1) | 0.4 (0.1) | 0.3 (0.0) | .57 |
| Mean E/e′ | 8.2 (1.9) | 8.7 (3.7) | 8.0 (2.0) | .70 |
| LAVI, mL/m2 | 25.4 (6.3) | 28.6 (6.3) | 28.8 (6.0) | .17 |
| Diastolic dysfunction grade, No. (%)c | ||||
| 0 | 14 (64) | 17 (85) | 14 (93) | .06 |
| 1 | 5 (23) | 0 | 0 | |
| Indeterminate | 3 (14) | 3 (15) | 1 (7) | |
| LVEF, % | 56.9 (5.2) | 62.4 (4.0) | 65.3 (2.9) | <.001 |
| GLS, % | −17.8 (2.2) | −19.8 (2.2) | −21.3 (1.8) | <.001 |
| GCS, % | −15.1 (3.9) | −17.1 (2.2) | −19.6 (3.0) | <.001 |
| GRS, % | 37.7 (16.9) | 46.5 (21.1) | 57.3 (14.9) | .009 |
Abbreviations: E/e′, ratio of the peak early mitral inflow velocity to the early diastolic mitral annular velocity; GCS, global circumferential strain; GLS, global longitudinal strain; GRS, global radial strain; HC, healthy controls; IVSd, interventricular septum thickness at end-diastole; LAVI, left atrial volume index; LVEF, left ventricular ejection fraction; LVIDd, left ventricular internal dimension at end-diastole; LVMI, left ventricular mass index; LVPWd, left ventricular posterior wall thickness at end-diastole; RWT, relative wall thickness.
Patients who developed cardiotoxicity during trastuzumab treatment constituted the TOX group; those with no evidence of cardiotoxicity during trastuzumab treatment, the NOTOX group. Unless otherwise indicated, data are expressed as mean (SD).
Represents a test of equality across groups and was calculated by analysis of variance for continuous variables and Fisher exact test for categorical variables.
Percentages have been rounded and may not total 100.
CRF and Postexercise Cardiac Function
Cardiopulmonary exercise performance data are presented in Table 3. Before exercise, resting heart rate, systolic blood pressure, and diastolic blood pressure were highest in the TOX group compared with the NOTOX and HC groups. The mean (SD) peak heart rate was 166 (11) bpm in the TOX group, 163 (14) bpm in the NOTOX group, and 167 (10) bpm in the HC group. Three participants in the TOX group did not achieve a peak VO2 and were excluded from the analysis; 2 developed an exercise-induced arrhythmia (1 with rapid atrial fibrillation and 1 with ventricular bigeminy), and 1 was unable to comply with the CPET testing equipment, leading to early termination. The remaining 54 of 57 participants (95%) met the acceptable criteria for a peak VO2 during CPET testing. A peak test based on heart rate occurred in 50 participants (88%); based on respiratory exchange ratio, in 33 (58%); and based on rating of perceived exertion, in 35 (61%). Mean peak VO2 in the TOX group (22.9 [4.4] mL/kg/min) was 15% lower compared with the NOTOX group (27.0 [5.3] mL/kg/min; P = .03) and 25% lower compared with the HC group (30.5 [3.4] mL/kg/min; P < .001) (Figure). In a sensitivity analysis, differences in peak VO2 between groups remained significant after exclusion of patients with a LVEF of less than 53% (TOX vs NOTOX groups, 22.8 [4.6] vs 27.0 [5.3] mL/kg/min [P = .006]; vs HC group, 30.5 [3.4] mL/kg/min [P < .001]).
Table 3. Cardiopulmonary Exercise Performance.
| Characteristic | Mean (SD) Valuea | P Valueb | ||
|---|---|---|---|---|
| TOX Group (n = 22) | NOTOX Group (n = 20) | HC Group (n = 15) | ||
| Resting vital signs | ||||
| Heart rate, bpm | 81 (14) | 72 (11) | 68 (7) | .002 |
| Blood pressure, mm Hg | ||||
| Systolic | 128 (15) | 121 (12) | 113 (11) | .006 |
| Diastolic | 78 (8) | 75 (8) | 69 (8) | .01 |
| Oxygen saturation, % | 97 (1) | 98 (1) | 98 (1) | .62 |
| Postexercise parameters | ||||
| Heart rate, bpm | 166 (11) | 163 (14) | 167 (10) | .63 |
| Heart rate, % of MPHR | 104 (7) | 101 (6) | 104 (7) | .14 |
| Blood pressure, mm Hg | ||||
| Systolic | 172 (13) | 182 (13) | 169 (14) | .02 |
| Diastolic | 71 (9) | 71 (9) | 64 (8) | .03 |
| Oxygen saturation, % | 96 (2) | 95 (2) | 95 (2) | .67 |
| Peak VO2, mL/kg/min | 22.9 (4.4) | 27.0 (5.3) | 30.5 (3.4) | <.001 |
| Peak VO2, % estimated | 78 (33) | 99 (16) | 109 (16) | .003 |
| Peak VO2, L/min | 1.6 (0.2) | 1.8 (0.4) | 1.8 (0.2) | .01 |
| LVEF, % | 65.6 (7.7) | 74.5 (5.4) | 75.6 (5.1) | <.001 |
| CO, L/min | 10.2 (1.9) | 11.6 2.4) | 11.0 (2.3) | .18 |
| CI, L/min/m2 | 5.9 (1.3) | 6.6 (1.4) | 6.8 (1.3) | .08 |
| Contractile reserve, L/min/m2 | 3.3 (0.9) | 4.0 (1.5) | 4.4 (1.2) | .03 |
| AVO2 difference, % | 15.9 (3.9) | 16.3 (4.3) | 16.6 (3.3) | .86 |
Abbreviations: AVO2, arterial-venous oxygen content; CI, cardiac index; CO, cardiac output; HC, healthy controls; LVEF, left ventricular ejection fraction; MPHR, maximum predicted heart rate; peak VO2, peak oxygen consumption.
Patients who developed cardiotoxicity during trastuzumab treatment constituted the TOX group; those with no evidence of cardiotoxicity during trastuzumab treatment, the NOTOX group.
P value represents a test of equality across groups and was calculated using analysis of variance.
Figure. Cardiopulmonary Function as Assessed by Resting Left Ventricular Ejection Fraction (LVEF), Resting Global Longitudinal Strain (GLS), and Peak Oxygen Consumption (Peak VO2).
Box-plot analysis of resting LVEF, resting GLS, and peak VO2 among the patients who developed cardiotoxicity during trastuzumab treatment (TOX group), those with no evidence of cardiotoxicity during trastuzumab treatment (NOTOX group), and age-matched healthy controls (HC group). The horizontal line within each box indicates the median, while the top and bottom borders of the box mark the 75th and 25th percentiles, respectively. The whiskers above and below the box mark the 90th and 10th percentiles. The blue dot represents outliers.
The TOX group had a significantly lower mean postexercise LVEF compared with the HC group (65.6% [7.7%] vs 75.6% [5.1%]; P < .001). Similarly, the mean contractile reserve was 25% lower in the TOX group compared with the HC group (3.3 [0.9] vs 4.4 [1.2] L/min/m2; P = .02). There was no significant difference in AVO2 difference between the TOX group compared with the NOTOX or the HC group.
Echocardiographic Parameters and CRF
Among the breast cancer survivors, regression analysis was performed to identify factors associated with peak VO2. After multivariate analysis, age (β coefficient for 55-64 years, −3.85 [95% CI, −7.06 to −0.64; β coefficient for ≥65 years, −4.75 [95% CI, −8.19 to −1.31), BMI (β coefficient, −0.59 [95% CI, −0.85 to −0.32]), and resting GLS (β coefficient, −0.75 [95% CI, −1.32 to −0.18]) were associated with peak VO2 (Table 4). Because AVO2 difference was derived from peak VO2 rather than directly measured, it was not included in the regression analysis.
Table 4. Variables Associated With Peak VO2 in Survivors of ERBB2-Positive Breast Cancer .
| Variable | Univariate Analysis | Multivariate Analysisa | ||
|---|---|---|---|---|
| β Coefficient (95% CI) | P Value | β Coefficient (95% CI) | P Value | |
| Age group, y | ||||
| <55 | 1 [Reference] | .30 | 1 [Reference] | .02 |
| 55-64 | −2.50 (−6.65 to 1.64) | −3.85 (−7.06 to −0.64) | ||
| ≥65 | −3.18 (−7.59 to 1.24) | −4.75 (−8.19 to −1.31) | ||
| BMI | −0.55 (−0.84 to −0.25) | <.001 | −0.59 (−0.85 to −0.32) | <.001 |
| Hypertension | −2.5 (−6.69 to 1.68) | .23 | NA | NA |
| Treatment-induced cardiotoxicity | −4.11 (−7.27 to −0.94) | .01 | NA | NA |
| Resting | ||||
| LVEF | 0.08 (−0.24 to 0.40) | .63 | NA | NA |
| GLS | −0.63 (−1.35 to 0.08) | .08 | −0.75 (−1.32 to −0.18) | .01 |
| After exercise | ||||
| LVEF | 0.14 (−0.08 to 0.36) | .21 | NA | NA |
| Cardiac index | 1.41 (0.14 to 2.68) | .03 | NA | NA |
| Contractile reserve | 1.35 (−0.03 to 2.72) | .06 | NA | NA |
Abbreviations: BMI, body mass index; GLS, global longitudinal strain; LVEF, left ventricular ejection fraction; NA, not applicable; peak VO2, peak oxygen consumption.
Variables in the multivariate model were chosen using a stepwise regression analysis where all variables listed in the univariate analysis were included.
Discussion
Several novel and important findings emerge from this study, which characterized the long-term cardiopulmonary effects of treatment-induced cardiotoxicity. First, patients in the TOX group who had an asymptomatic decline of LVEF during trastuzumab treatment had a persistent reduction in left ventricular systolic function as measured by LVEF and GLS compared with the NOTOX and HC groups at a median follow-up of 7 years after completing trastuzumab therapy. We note that 90% of patients in the TOX and NOTOX groups received an anthracycline-based trastuzumab regimen. Second, exercise capacity among survivors of treatment-induced cardiotoxicity was severely impaired, with a peak VO2 that was 25% lower than that in the HC group and 15% lower than in the NOTOX group. Finally, resting GLS was significantly associated with peak VO2. These findings suggest that cardiac factors, specifically impairment of left ventricular systolic function, play an important role in limitation of exercise capacity among breast cancer survivors. To date, there has been uncertainty regarding the clinical significance of an asymptomatic LVEF decline during trastuzumab therapy, which is the most common manifestation of cardiotoxicity during targeted therapy for ERBB2-positive breast cancer. Considering the adverse prognostic significance associated with low peak VO2, our findings suggest that treatment-induced cardiotoxicity may affect long-term cardiovascular outcomes and provide a strong rationale to develop effective strategies for prevention and management of treatment-induced cardiotoxicity.
Survivorship studies in several different cancer populations have shown that cardiotoxic cancer therapy is associated with reductions in CRF.10,11,21 The present study expands on these previous methods to demonstrate that treatment-induced cardiotoxicity plays an important role in the chronic reduction of CRF. We believe these findings are of particular clinical significance, given the association between low CRF and all-cause mortality, cardiovascular mortality, and cancer-related mortality.27 Low peak VO2 may also be linked to the high prevalence of fatigue and poor exercise tolerance that can negatively affect quality of life and are commonly reported by breast cancer survivors.28,29 In acknowledgment of the independent and additive prognostic information that is gained from a CRF assessment, the American Heart Association suggests that CRF be routinely measured in clinical practice among the general population.30 Among cancer survivors, CRF assessment may unmask the cardiotoxic effects of cancer therapy that cannot be detected with traditional cardiac imaging in this patient population such as 2-D echocardiography.
To further elucidate the underlying mechanism of impaired peak VO2 associated with treatment-induced cardiotoxicity, we evaluated for differences in resting and postexercise echocardiographic parameters. Based on the Fick equation, determinants of peak VO2 include cardiac output and AVO2 difference; therefore, reductions in either of these parameters will lead to a reduction in peak VO2.31 At rest, left ventricular dimensions and continuous diastolic function parameters were similar between all 3 groups, whereas resting LVEF and GLS were lower in the TOX group. Chronotropic incompetence during exercise was not a contributing factor to low peak VO2, with the TOX and NOTOX groups achieving a maximum heart rate that was similar to the HC group and 49 (86%) of participants achieving age-predicted maximal HR. Postexercise assessments of cardiac function, including LVEF and contractile reserve, were also lower in the TOX group compared with the HC group, whereas AVO2 difference was similar among all 3 groups. These findings point toward left ventricular systolic dysfunction as an important contributor to low peak VO2 in the TOX group.
In a multivariate analysis among the breast cancer survivor groups, GLS was the only echocardiographic parameter that was independently associated with peak VO2. These findings confirm and extend previous observations in patients with heart failure with preserved ejection fraction and breast cancer survivors. In heart failure with preserved ejection fraction, there is a strong association between GLS and peak VO2,32,33 suggesting that GLS assessment may help to identify patients with subclinical left ventricular systolic dysfunction and impaired exercise capacity. A study of breast cancer survivors21 previously reported an association between inotropic reserve and peak VO2, but not GLS, although that study evaluated patients with ERBB2-negative cancer who received anthracyclines without targeted ERBB2 agents.
Our findings suggest that with continued improvements in breast cancer outcomes, a growing population of survivors previously treated with cardiotoxic cancer therapy will be at risk for long-term impairment in CRF, and by association, shortened survival. Preventive and/or therapeutic strategies are needed to counteract the cardiotoxic effects of cancer therapy and prevent reduction of peak VO2. Aerobic exercise training represents a potential intervention that can target the cardiac-specific limitation associated with cardiotoxicity and improve peak VO2.12,34,35,36 Data from Imboden et al37 suggest that every 1 metabolic equivalent increase in peak VO2 is associated with a 12% reduction in all-cause mortality and 16% reduction in cardiovascular mortality in asymptomatic men and women after approximately 18 years of follow-up.
Limitations
This study has several limitations. The cross-sectional design of this study introduces potential selection bias among the recruited participants. However, the cancer treatment exposures and cardiovascular comorbidities of the TOX group closely matched those of the NOTOX group. Temporality between cardiotoxicity and low peak VO2 cannot be inferred owing to the cross-sectional design but will be addressed in a future longitudinal study. Although the number of patients included in this study was modest, our study is the first, to our knowledge, to comprehensively evaluate the long-term sequelae of treatment-induced cardiotoxicity among ERBB2-positive breast cancer survivors and contributes to the growing knowledge base of the adverse effects of cancer treatment on CRF. Grade I diastolic dysfunction was more prevalent among the TOX group; however, assessment of an association between peak VO2 and categorical classifications of diastolic dysfunction was limited owing to the small number of patients meeting these criteria. Cardiopulmonary exercise testing was performed during upright exercise, whereas postpeak images were acquired in the supine position; thus, it is unknown whether the observed differences in contractile reserve extend to the upright position. Nevertheless, differences in contractile reserve likely persist given that peak heart rate was similar between groups.38 Finally, AVO2 difference was derived from peak VO2 rather than directly measured; therefore, the relative contribution of AVO2 difference on impaired CRF could not be assessed. However, the absence of any significant difference in AVO2 difference among the TOX, NOTOX, and HC groups in this study closely match findings by Jones et al39 in a similar population of breast cancer survivors.
Conclusions
In summary, ERBB2-positive breast cancer survivors with prior treatment-induced cardiotoxicity have a significant impairment of CRF many years after the completion of therapy. Impairment of cardiac function appears to play an important role in the mechanism of low peak VO2. Assessment of GLS may be helpful for identifying breast cancer survivors with impaired exercise capacity and poor prognosis. We think future interventional studies are needed to prevent and mitigate the decline in peak VO2 associated with treatment-induced cardiotoxicity, thereby reducing cardiovascular disease risk and all-cause mortality associated with low peak VO2. Aerobic exercise training represents a potential intervention to improve CRF among breast cancer survivors that warrants further investigation.
eFigure. Consort Diagram
References
- 1.Romond EH, Jeong JH, Rastogi P, et al. . Seven-year follow-up assessment of cardiac function in NSABP B-31, a randomized trial comparing doxorubicin and cyclophosphamide followed by paclitaxel (ACP) with ACP plus trastuzumab as adjuvant therapy for patients with node-positive, human epidermal growth factor receptor 2-positive breast cancer. J Clin Oncol. 2012;30(31):3792-3799. doi: 10.1200/JCO.2011.40.0010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Goldhirsch A, Gelber RD, Piccart-Gebhart MJ, et al. ; Herceptin Adjuvant (HERA) Trial Study Team . 2 years versus 1 year of adjuvant trastuzumab for HER2-positive breast cancer (HERA): an open-label, randomised controlled trial. Lancet. 2013;382(9897):1021-1028. doi: 10.1016/S0140-6736(13)61094-6 [DOI] [PubMed] [Google Scholar]
- 3.Tarantini L, Cioffi G, Gori S, et al. ; Italian Cardio-Oncologic Network . Trastuzumab adjuvant chemotherapy and cardiotoxicity in real-world women with breast cancer. J Card Fail. 2012;18(2):113-119. doi: 10.1016/j.cardfail.2011.10.015 [DOI] [PubMed] [Google Scholar]
- 4.Slamon D, Eiermann W, Robert N, et al. ; Breast Cancer International Research Group . Adjuvant trastuzumab in HER2-positive breast cancer. N Engl J Med. 2011;365(14):1273-1283. doi: 10.1056/NEJMoa0910383 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Guarneri V, Lenihan DJ, Valero V, et al. . Long-term cardiac tolerability of trastuzumab in metastatic breast cancer: the MD Anderson Cancer Center experience. J Clin Oncol. 2006;24(25):4107-4115. doi: 10.1200/JCO.2005.04.9551 [DOI] [PubMed] [Google Scholar]
- 6.Chen J, Long JB, Hurria A, Owusu C, Steingart RM, Gross CP. Incidence of heart failure or cardiomyopathy after adjuvant trastuzumab therapy for breast cancer. J Am Coll Cardiol. 2012;60(24):2504-2512. doi: 10.1016/j.jacc.2012.07.068 [DOI] [PubMed] [Google Scholar]
- 7.Banke A, Fosbøl EL, Ewertz M, et al. . Long-term risk of heart failure in breast cancer patients after adjuvant chemotherapy with or without trastuzumab. JACC Heart Fail. 2019;7(3):217-224. doi: 10.1016/j.jchf.2018.09.001 [DOI] [PubMed] [Google Scholar]
- 8.Ganz PA, Romond EH, Cecchini RS, et al. . Long-term follow-up of cardiac function and quality of life for patients in NSABP Protocol B-31/NRG Oncology: a randomized trial comparing the safety and efficacy of doxorubicin and cyclophosphamide (AC) followed by paclitaxel with AC followed by paclitaxel and trastuzumab in patients with node-positive breast cancer with tumors overexpressing human epidermal growth factor receptor 2. J Clin Oncol. 2017;35(35):3942-3948. doi: 10.1200/JCO.2017.74.1165 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Cikes M, Solomon SD. Beyond ejection fraction: an integrative approach for assessment of cardiac structure and function in heart failure. Eur Heart J. 2016;37(21):1642-1650. doi: 10.1093/eurheartj/ehv510 [DOI] [PubMed] [Google Scholar]
- 10.Jones LW, Courneya KS, Mackey JR, et al. . Cardiopulmonary function and age-related decline across the breast cancer survivorship continuum. J Clin Oncol. 2012;30(20):2530-2537. doi: 10.1200/JCO.2011.39.9014 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Klassen O, Schmidt ME, Scharhag-Rosenberger F, et al. . Cardiorespiratory fitness in breast cancer patients undergoing adjuvant therapy. Acta Oncol. 2014;53(10):1356-1365. doi: 10.3109/0284186X.2014.899435 [DOI] [PubMed] [Google Scholar]
- 12.Scott JM, Zabor EC, Schwitzer E, et al. . Efficacy of exercise therapy on cardiorespiratory fitness in patients with cancer: a systematic review and meta-analysis. J Clin Oncol. 2018;36(22):2297-2305. doi: 10.1200/JCO.2017.77.5809 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Jones LW, Hornsby WE, Goetzinger A, et al. . Prognostic significance of functional capacity and exercise behavior in patients with metastatic non–small cell lung cancer. Lung Cancer. 2012;76(2):248-252. doi: 10.1016/j.lungcan.2011.10.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Sloan RA, Sawada SS, Martin CK, Church T, Blair SN. Associations between cardiorespiratory fitness and health-related quality of life. Health Qual Life Outcomes. 2009;7:47. doi: 10.1186/1477-7525-7-47 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Herrero F, Balmer J, San Juan AF, et al. . Is cardiorespiratory fitness related to quality of life in survivors of breast cancer? J Strength Cond Res. 2006;20(3):535-540. [DOI] [PubMed] [Google Scholar]
- 16.Schlesselman JJ, Stolley PD. Case-Control Studies: Design, Conduct, Analysis. New York, NY: Oxford University Press; 1982. [Google Scholar]
- 17.Lang RM, Bierig M, Devereux RB, et al. ; Chamber Quantification Writing Group; American Society of Echocardiography’s Guidelines and Standards Committee; European Association of Echocardiography . Recommendations for chamber quantification: a report from the American Society of Echocardiography’s Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr. 2005;18(12):1440-1463. doi: 10.1016/j.echo.2005.10.005 [DOI] [PubMed] [Google Scholar]
- 18.Yu AF, Raikhelkar J, Zabor EC, et al. . Two-dimensional speckle tracking echocardiography detects subclinical left ventricular systolic dysfunction among adult survivors of childhood, adolescent, and young adult cancer. Biomed Res Int. 2016;2016:9363951. doi: 10.1155/2016/9363951 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Yu AF, Ho AY, Braunstein LZ, et al. . Assessment of early radiation-induced changes in left ventricular function by myocardial strain imaging after breast radiation therapy. J Am Soc Echocardiogr. 2019;32(4):521-528. doi: 10.1016/j.echo.2018.12.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.American Thoracic Society; American College of Chest Physicians . ATS/ACCP statement on cardiopulmonary exercise testing. Am J Respir Crit Care Med. 2003;167(2):211-277. doi: 10.1164/rccm.167.2.211 [DOI] [PubMed] [Google Scholar]
- 21.Khouri MG, Hornsby WE, Risum N, et al. . Utility of 3-dimensional echocardiography, global longitudinal strain, and exercise stress echocardiography to detect cardiac dysfunction in breast cancer patients treated with doxorubicin-containing adjuvant therapy. Breast Cancer Res Treat. 2014;143(3):531-539. doi: 10.1007/s10549-013-2818-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Lewis JF, Kuo LC, Nelson JG, Limacher MC, Quinones MA. Pulsed Doppler echocardiographic determination of stroke volume and cardiac output: clinical validation of two new methods using the apical window. Circulation. 1984;70(3):425-431. doi: 10.1161/01.CIR.70.3.425 [DOI] [PubMed] [Google Scholar]
- 23.Plana JC, Galderisi M, Barac A, et al. . Expert consensus for multimodality imaging evaluation of adult patients during and after cancer therapy: a report from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. J Am Soc Echocardiogr. 2014;27(9):911-939. doi: 10.1016/j.echo.2014.07.012 [DOI] [PubMed] [Google Scholar]
- 24.Negishi K, Negishi T, Hare JL, Haluska BA, Plana JC, Marwick TH. Independent and incremental value of deformation indices for prediction of trastuzumab-induced cardiotoxicity. J Am Soc Echocardiogr. 2013;26(5):493-498. doi: 10.1016/j.echo.2013.02.008 [DOI] [PubMed] [Google Scholar]
- 25.Kuznetsova T, Cauwenberghs N, Knez J, et al. . Additive prognostic value of left ventricular systolic dysfunction in a population-based cohort. Circ Cardiovasc Imaging. 2016;9(7):e004661. doi: 10.1161/CIRCIMAGING.116.004661 [DOI] [PubMed] [Google Scholar]
- 26.Oikonomou EK, Kokkinidis DG, Kampaktsis PN, et al. . Assessment of prognostic value of left ventricular global longitudinal strain for early prediction of chemotherapy-induced cardiotoxicity: a systematic review and meta-analysis. JAMA Cardiol. Published online August 21, 2019. doi: 10.1001/jamacardio.2019.2952 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Imboden MT, Harber MP, Whaley MH, Finch WH, Bishop DL, Kaminsky LA. Cardiorespiratory fitness and mortality in healthy men and women. J Am Coll Cardiol. 2018;72(19):2283-2292. doi: 10.1016/j.jacc.2018.08.2166 [DOI] [PubMed] [Google Scholar]
- 28.Bower JE. Cancer-related fatigue: mechanisms, risk factors, and treatments. Nat Rev Clin Oncol. 2014;11(10):597-609. doi: 10.1038/nrclinonc.2014.127 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Bower JE, Ganz PA, Desmond KA, et al. . Fatigue in long-term breast carcinoma survivors: a longitudinal investigation. Cancer. 2006;106(4):751-758. doi: 10.1002/cncr.21671 [DOI] [PubMed] [Google Scholar]
- 30.Ross R, Blair SN, Arena R, et al. ; American Heart Association Physical Activity Committee of the Council on Lifestyle and Cardiometabolic Health; Council on Clinical Cardiology; Council on Epidemiology and Prevention; Council on Cardiovascular and Stroke Nursing; Council on Functional Genomics and Translational Biology; Stroke Council . Importance of assessing cardiorespiratory fitness in clinical practice: a case for fitness as a clinical vital sign: a scientific statement from the American Heart Association. Circulation. 2016;134(24):e653-e699. doi: 10.1161/CIR.0000000000000461 [DOI] [PubMed] [Google Scholar]
- 31.Koelwyn GJ, Jones LW, Moslehi J. Unravelling the causes of reduced peak oxygen consumption in patients with cancer: complex, timely, and necessary. J Am Coll Cardiol. 2014;64(13):1320-1322. doi: 10.1016/j.jacc.2014.07.949 [DOI] [PubMed] [Google Scholar]
- 32.Hasselberg NE, Haugaa KH, Sarvari SI, et al. . Left ventricular global longitudinal strain is associated with exercise capacity in failing hearts with preserved and reduced ejection fraction. Eur Heart J Cardiovasc Imaging. 2015;16(2):217-224. doi: 10.1093/ehjci/jeu277 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Petersen JW, Nazir TF, Lee L, Garvan CS, Karimi A. Speckle tracking echocardiography-determined measures of global and regional left ventricular function correlate with functional capacity in patients with and without preserved ejection fraction. Cardiovasc Ultrasound. 2013;11:20. doi: 10.1186/1476-7120-11-20 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Lin X, Zhang X, Guo J, et al. . Effects of exercise training on cardiorespiratory fitness and biomarkers of cardiometabolic health: a systematic review and meta-analysis of randomized controlled trials. J Am Heart Assoc. 2015;4(7):e002014. doi: 10.1161/JAHA.115.002014 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Schmitz KH, Holtzman J, Courneya KS, Mâsse LC, Duval S, Kane R. Controlled physical activity trials in cancer survivors: a systematic review and meta-analysis. Cancer Epidemiol Biomarkers Prev. 2005;14(7):1588-1595. doi: 10.1158/1055-9965.EPI-04-0703 [DOI] [PubMed] [Google Scholar]
- 36.McNeely ML, Campbell KL, Rowe BH, Klassen TP, Mackey JR, Courneya KS. Effects of exercise on breast cancer patients and survivors: a systematic review and meta-analysis. CMAJ. 2006;175(1):34-41. doi: 10.1503/cmaj.051073 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Imboden MT, Harber MP, Whaley MH, et al. . The Association between the change in directly measured cardiorespiratory fitness across time and mortality risk. Prog Cardiovasc Dis. 2019;62(2):157-162. doi: 10.1016/j.pcad.2018.12.003 [DOI] [PubMed] [Google Scholar]
- 38.Beaudry RI, Howden EJ, Foulkes S, et al. . Determinants of exercise intolerance in breast cancer patients prior to anthracycline chemotherapy. Physiol Rep. 2019;7(1):e13971. doi: 10.14814/phy2.13971 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Jones LW, Haykowsky M, Pituskin EN, et al. . Cardiovascular reserve and risk profile of postmenopausal women after chemoendocrine therapy for hormone receptor–positive operable breast cancer. Oncologist. 2007;12(10):1156-1164. doi: 10.1634/theoncologist.12-10-1156 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eFigure. Consort Diagram