This preplanned analysis included first 174 patients who reached the end of treatment assessment.
Changes from baseline to end of treatment in LVEF (A and B) and GLS (C and D) as expressed in percentage points with 95% CIs. BMI indicates body mass index (calculated as weight in kilograms divided by height in meters squared).
eTable 1. Allowed chemotherapy regimens (both in the adjuvant and neoadjuvant treatment setting).
eTable 2. Patient characteristics.
eTable 3. Echocardiography imaging measures and study drug hemodynamic effect stratified by treatment arm at each time-point (ITT analysis).
eTable 4. Echocardiography imaging measures and study drug hemodynamic effect comparison between baseline and end of treatment (ITT analysis).
eTable 5. eTable 5. Echocardiography imaging measures and study drug hemodynamic effect comparison between baseline, end of treatment, end of study observation (ITT analysis).
eTable 6. Study drug dose titration and tolerability.
eMethods 1. Main inclusion and exclusion criteria.
eMethods 2. Cardiac assessment.
eMethods 3. Statistical Methods.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Identify all potential conflicts of interest that might be relevant to your comment.
Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.
Err on the side of full disclosure.
If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.
Not all submitted comments are published. Please see our commenting policy for details.
Livi L, Barletta G, Martella F, et al. Cardioprotective Strategy for Patients With Nonmetastatic Breast Cancer Who Are Receiving an Anthracycline-Based Chemotherapy: A Randomized Clinical Trial. JAMA Oncol. 2021;7(10):1544–1549. doi:10.1001/jamaoncol.2021.3395
Can pharmacological cardioprevention avoid subclinical damage in patients with breast cancer who are undergoing cardiotoxic therapy?
In this randomized clinical trial of 262 patients with breast cancer, 174 of whom were analyzed, at 12 months, left ventricular ejection fraction as evaluated by 3-dimensional echocardiography worsened by 4.4% in the placebo arm and 3%, 1.9%, 1.3% in the ramipril, bisoprolol, and ramipril plus bisoprolol arms, respectively. Global longitudinal strain worsened by 6.0% in the placebo arm and 1.5% and 0.6% in the ramipril and bisoprolol arms, respectively, whereas it was unchanged (0.1% improvement) in the ramipril plus bisoprolol arm.
The results of this trial suggest that during cardiotoxic therapy for breast cancer, the use of pharmacological cardioprevention may be warranted.
Several studies have evaluated cardioprotective strategies to prevent myocardial dysfunction in patients who are receiving cardiotoxic therapies. However, the optimal approach still represents a controversial issue.
To determine whether pharmacological cardioprevention could reduce subclinical heart damage in patients with breast cancer who are being treated with anthracycline-based chemotherapy.
Design, Setting, and Participants
The SAFE trial was a 4-arm, randomized, phase 3, double-blind, placebo-controlled, national multicentric study conducted at 8 oncology departments in Italy. It was a prespecified interim analysis on the first 174 patients who had completed cardiac assessment at 12 months. The study recruitment was conducted between July 2015 and June 2020. The interim analysis was performed in 2020. Patients were eligible for trial inclusion if they had indication to receive primary or postoperative systemic therapy using an anthracycline-based regimen. Patients with a prior diagnosis of cardiovascular disease were excluded.
Cardioprotective therapy (bisoprolol, ramipril, or both drugs compared with placebo) was administered for 1 year from the initiation of chemotherapy or until the end of trastuzumab therapy in case of ERBB2-positive patients. Doses for all groups were systematically up-titrated up to the daily target dose of bisoprolol (5 mg, once daily), ramipril (5 mg, once daily), and placebo, if tolerated.
Main Outcomes and Measures
The primary end point was defined as detection of any subclinical impairment (worsening ≥10%) in myocardial function and deformation measured with standard and 3-dimensional (3D) echocardiography, left ventricular ejection fraction (LVEF), and global longitudinal strain (GLS).
The analysis was performed on 174 women (median age, 48 years; range, 24-75 years) who had completed a cardiological assessment at 12 months and reached the end of treatment. At 12 months, 3D-LVEF worsened by 4.4% in placebo arm and 3.0%, 1.9%, 1.3% in the ramipril, bisoprolol, ramipril plus bisoprolol arms, respectively (P = .01). Global longitudinal strain worsened by 6.0% in placebo arm and 1.5% and 0.6% in the ramipril and bisoprolol arms, respectively, whereas it was unchanged (0.1% improvement) in the ramipril plus bisoprolol arm (P < .001). The number of patients showing a reduction of 10% or greater in 3D-LVEF was 8 (19%) in the placebo arm, 5 (11.5%) in the ramipril arm, 5 (11.4%) in the bisoprolol, arm and 3 (6.8%) in the ramipril plus bisoprolol arm; 15 patients (35.7%) who received placebo showed a 10% or greater worsening of GLS compared with 7 (15.9; ramipril), 6 (13.6%; bisoprolol), and 6 (13.6%; ramipril plus bisoprolol) (P = .03).
Conclusions and Relevance
The interim analysis of this randomized clinical trials suggested that cardioprotective pharmacological strategies in patients who were affected by breast cancer and were receiving an anthracycline-based chemotherapy are well tolerated and seem to protect against cancer therapy–related LVEF decline and heart remodeling.
ClinicalTrials.gov identifier: NCT2236806
For decades, anthracycline-based chemotherapy regimens have been a mainstay in treating breast cancer (BC).1 However, anthracyclines may result in mild or severe short- and long-term toxic effects, with cardiotoxic effects being a well-known complication.2,3 Reduced left ventricular ejection fraction (LVEF) leads to interruption or discontinuation of therapy and is associated with an increased death risk.4
Echocardiography is considered the criterion standard in the cardiac imaging evaluation of patients during and after cancer therapy.5 Remodeling that precedes ventricular dysfunction and a reduction in global linear strain (GLS) seems to be the most sensitive parameter to predict early cardiotoxic effects.6,7
Angiotensin-converting enzyme inhibitors (ACEis)/angiotensin-II receptor blockers and β-blockers (BB) have been shown to prevent cardiac remodeling and reduce mortality in patients with cardiac dysfunction8 and have been proposed for cardioprotection in oncology.9 In this article, we examine the results of the preplanned interim analysis of the SAFE phase 3 randomized clinical trial that assessed a cardioprotective strategy in preventing subclinical cardiac damage caused by anthracycline-based chemotherapy.
SAFE (NCT2236806) was a 4-arm, randomized, phase 3, double-blind, placebo-controlled study that aimed to evaluate the effect of bisoprolol, ramipril, or both drugs compared with placebo on subclinical cardiac dysfunction that was evaluated by 3-dimensional (3D) and speckle tracking cardiac echocardiography in patients with nonmetastatic BC (Supplement 1). The trial was approved by the local and national ethical committees and participants provided written informed consent. Cardioprotective therapy was administered for 1 year from the initiation of chemotherapy or until the end of trastuzumab therapy. The primary end point was defined as detection of any subclinical impairment (worsening ≥10%) in myocardial function and deformation measured with standard and 3D echocardiography, LVEF, and GLS. Doses for all groups were up titrated at 1-week intervals up to the target dose of bisoprolol (5 mg), ramipril (5 mg), and placebo, if tolerated. All patients underwent cardiac surveillance at baseline (T0), 3 months (T1), 6 months (T2), 12 months (T3, end of treatment [EOT]), and 24 months (T4, end of study) from enrollment. This was a prespecified interim analysis on the first 174 patients who had completed cardiac assessment at T3 and reached the EOT. Eighty-eight patients were not included in this analysis because they had not reached yet the 24-month follow-up (Figure 1). Patient selection (eMethods 1 in Supplement 2), chemotherapy regimens (eTable 1 in Supplement 2), cardiac assessment (eMethods 2 in Supplement 2), and statistics (eMethods 3 in Supplement 2) are reported. Statistical analyses were conducted using SPSS, version 26. 0 (IBM); statistical significance was set at 2-side α = .05.
Between July 2015 and June 2020, 262 patients were enrolled in the trial and randomized (Figure 1). The median age was 48 years (range, 24-75 years), and most patients were stage I to II (143 [82.2%]) and hormonal receptor–positive (132 [75.9%]). Adjuvant trastuzumab was administered for 64 patients (36.8%) and radiotherapy for 101 patients (58%). No significant differences were found in group distribution, although there were more hormonal receptor–positive patients in the placebo (36 [86%]) rather than ramipril plus bisoprolol (30 [70%]) arm (P = .74). Patient baseline characteristics are summarized in eTable 2 in Supplement 2.
Cardiac features stratified by treatment-arm at each point are reported in eTable 3 in Supplement 2. The study drug hemodynamic effect as stratified by treatment arm and estimated changes from baseline to EOT using analysis of covariance are summarized in the Table. At 12 months, 3D LVEF worsened by 4.4% in the placebo arm and 3.0%, 1.9%, and 1.3% in the ramipril, bisoprolol, and ramipril plus bisoprolol arms, respectively (P = .01). Global longitudinal strain worsened by 6.0% in placebo arm and 1.5% and 0.6% in the ramipril and bisoprolol arms, respectively, whereas was unchanged (0.1% improvement) in the ramipril plus bisoprolol arm (P < .001). The number of patients whose 3D LVEF worsened by 10% or greater was 8 (19%) in the placebo arm, 5 (11.5%) in the ramipril arm, 5 (11.4%) in the bisoprolol arm, and 3 (6.8%) in the ramipril plus bisoprolol arm; 15 patients (35.7%) who received placebo showed a 10% or greater worsening of GLS compared with 7 (15.9%; ramipril), 6 (13.6%; bisoprolol), and 6 (13.6%; ramipril plus bisoprolol) (P = .03).
Patients in a bisoprolol-containing arm exhibited a more favorable cardiac profile than those in the ramipril-containing arms, as showed by changes in 3D LVEF and GLS from baseline to 12 months (eTable 4 in Supplement 2) that were independent of patient- and treatment-related characteristics (Figure 2). End-diastolic volume (EDVI) and end-systolic volume (ESVI) values significantly favored patients in bisoprolol- and ramipril-containing arms compared with the placebo arm, especially in bisoprolol-containing arms.
A total of 125 patients reached the end of the study at T4 assessment (eTable 5 in Supplement 2). Compliance with the study drugs was good overall, although patients who received ramipril plus bisoprolol experienced higher dose titration (8 [18.6%]) and a higher rate of hypotension (6 [14%]) and cough (2 [4.7%]) compared with other arms (P < .05). Details are summarized in eTable 6 in Supplement 2.
To our knowledge, this study is one of the largest published series of homogenously treated patients who received a cardioprotective strategy who were assessed at 12- and 24-month follow up using 3D LVEF and GLS as subclinical heart functional assessment. This study reinforced the importance of a cardioprotective strategy for early patients with BC who received an anthracycline-based chemotherapy. Ramipril and bisoprolol seemed to attenuate anthracycline-related decline in LVEF with a good safety profile, and the benefit was maintained at 12 and 24 months for most patients independently of patient-, cancer-, and treatment-related features. Women in the placebo arm showed adverse cardiac remodeling and decline in cardiac performance at 12 and 24 months. Conversely, patients who received a cardioprevention strategy using ramipril and/or bisoprolol showed a significant protective effect in terms of GLS and 3D LVEF values, notably favoring the bisoprolol-containing arms. It has been demonstrated that EDVI and ESVI might act as a surrogate for evaluating cardiac remodeling. In patients who receive early diagnosis/intervention, cardiac function is more likely to recover with receipt of heart failure treatment.10
In this study, the protective effect of a cardioprevention was more evident at 12 months than at the end of active treatment. However, the benefit was maintained at 24 months for EDVI and ESVI, supporting the previously mentioned concept of a favorable effect on cardiac remodeling. In the MANTICORE trial,11 ACEi and BB attenuated trastuzumab-mediated decline in LVEF without preventing heart remodeling.
In regard to 3D LVEF, all patient subgroups significantly benefitted from a bisoprolol-based therapy, whereas differences were not significant for those in ramipril-containing arms (Figure 2, A and B). In regard to GLS (Figure 2, C and D), all patients benefitted from bisoprolol-containing treatments.
To our knowledge, few studies have previously evaluated BB and ACEi alone or in combination to prevent myocardial dysfunction in patients who are receiving cardiotoxic therapies, although none of them evaluated GLS.12,13 Our data support the high sensitivity of GLS to identify subclinical damage and that cardiotoxic effects after anthracyclines occur within the first year after administration.14 The benefit related to a cardioprotective strategy was most evident in physically healthy and younger patients and patients without comorbidities, thus confirming that a cardioprotective strategy should be highly considered for all patients and not only for those classically considered at high risk.9 Conversely, the PRADA study (median follow-up, 23 months) showed that candesartan treatment was associated with a significant reduction in EDVI and only an attenuated decline in GLS and modest effect on LVEF at 2 years, suggesting that a cardioprotective approach may not be required in patients without preexisting cardiovascular disease.15
This study had limitations. This was a preplanned interim analysis and the results, although significantly favoring a cardioprotective strategy, have to be confirmed at the final study analysis. Moreover, the slow rate of enrollment, worsened by the outbreak of COVID-19 that has severely affected the ability to safely conduct clinical trials, led to the prematurely termination of recruitment.
This preplanned interim analysis of the SAFE randomized clinical trial suggests that cardioprotective pharmacological strategies in patients who are affected by BC and receiving an anthracycline-based chemotherapy are overall well tolerated and seem to protect against LVEF decline and heart remodeling. Nevertheless, these encouraging preliminary results will have to be confirmed at the final results analysis.
Accepted for Publication: June 7, 2021.
Published Online: August 26, 2021. doi:10.1001/jamaoncol.2021.3395
Corresponding Author: Icro Meattini, MD, Department of Experimental and Clinical Biomedical Sciences, University of Florence, Viale Morgagni 50, 50134 – Florence, Italy (email@example.com).
Author Contributions: Drs Barletta and Livi had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Livi, Barletta, Desideri, Airoldi, Visani, Salvestrini, Mariotti, Pedani, Meattini.
Acquisition, analysis, or interpretation of data: Barletta, Martella, Saieva, Bacci, Del Bene, Amoroso, Coltelli, Scotti, Becherini, Visani, Salvestrini, Bernini, Sanchez, Orzalesi, Nori, Bianchi, Olivotto, Meattini.
Drafting of the manuscript: Livi, Barletta, Saieva, Del Bene, Airoldi, Amoroso, Coltelli, Becherini, Visani, Salvestrini, Mariotti, Pedani, Nori, Meattini.
Critical revision of the manuscript for important intellectual content: Livi, Barletta, Martella, Desideri, Bacci, Scotti, Becherini, Visani, Salvestrini, Mariotti, Bernini, Sanchez, Orzalesi, Bianchi, Olivotto, Meattini.
Statistical analysis: Barletta, Saieva.
Obtained funding: Livi, Desideri, Meattini.
Administrative, technical, or material support: Desideri, Bacci, Becherini, Visani, Bernini, Sanchez, Meattini.
Supervision: Livi, Barletta, Del Bene, Airoldi, Amoroso, Scotti, Becherini, Visani, Salvestrini, Mariotti, Pedani, Bernini, Orzalesi, Nori, Bianchi, Meattini.
Conflict of Interest Disclosures: Dr Olivotto reported grants from Myokardia, BMS, Shire, Takeda, Sanofi Genzyme, Amicus, and Bayer outside the submitted work. Dr Meattini reported personal fees from Novartis, Eli Lilly, Roche, and Pfizer outside the submitted work. No other disclosures were reported.
Additional Contributions: We thank the women enrolled in this trial, the study coinvestigators, and the staff of all the involved centers. We also thank Francesca Simontacchi (Fondazione Radioterapia Oncologica), Francesco Venditti, Giuseppe Pilato, Giorgio Galanti, Francesca Terziani, Noela Gori, Nicola Flego (Azienda Ospedaliero Universitaria Careggi), Roberto Tarquini (Empoli Hospital), Marco Di Lieto and Ilaria Pazzagli (Pistoia Hospita), Carmelo Bengala (Grosseto Hospital), Sara Donati (Versilia Hospital), Giacomo Allegrini (Livorno Hospital), Hector Jose Soto Parra (Azienda Ospedaliero Universitaria Policlinico), Giuseppe Curigliano (Istituto Europeo di Oncologia), and Valentina Guarneri (Istituto Oncologico Veneto) for their valuable contributions, for which they were not compensated.
Data Sharing Statement: See Supplement 3.