By Michael H. Crawford, MD, Editor
SYNOPSIS: Using left ventricular peak systolic global longitudinal strain vs. left ventricular ejection fraction to survey patients undergoing potentially cardiotoxic chemotherapy with at least one other risk factor for heart failure showed no difference in the primary endpoint of the difference in ejection fraction between the two groups at one year.
SOURCE: Thavendiranathan P, Negishi T, Somerset E, et al. Strain-guided management of potentially cardiotoxic cancer therapy. J Am Coll Cardiol 2021;77:392-401.
Traditional surveillance of cardiac function during cancer chemotherapy has relied on left ventricular (LV) ejection fraction (EF) measured by echocardiography. Echocardiographic peak systolic LV global longitudinal strain (GLS) has shown promise for detecting subclinical cardiac dysfunction before LVEF is abnormal. Whether GLS would be useful for implementing cardioprotective therapy (CPT) or ceasing chemotherapy is unclear.
Thus, this analysis of the one-year follow-up results of the multicenter, multinational Strain Surveillance of Chemotherapy for Improving Cardiovascular Outcomes (SUCCOUR) study is of interest. The authors randomized 331 patients treated with anthracycline-based chemotherapy and another risk factor for heart failure at 28 centers on four continents between 2014 and 2019 to an EF-guided arm and a GLS-guided arm measured by 3D echocardiography every three months. Other risk factors included trastuzumab therapy in HER2+ breast cancer; use of tyrosine kinase inhibitors; doxorubicin doses of > 450 mg/m2; or any traditional heart failure risk factor, such as older than age 65 years, diabetes, hypertension, or prior myocardial infarction. Exclusion criteria included a history of heart failure or an EF < 50%, moderate or more valvular heart disease, systolic blood pressure lower than 110 mmHg, and heart rate < 60 bpm.
In the EF-guided group, a symptomatic drop in EF of > 5% or an asymptomatic drop of > 10% compared to baseline was considered cancer chemotherapy-related cardiac dysfunction (CRD). In the GLS-guided arm, a decrease of > 12% in LV function from baseline was considered significant. Those with CRD started on an ACE inhibitor or an ARB, followed by a beta-blocker titrated to maximally tolerated doses. The primary outcome was the difference between baseline and one-year LVEF between the two arms.
During follow-up, 24 patients in each group did not complete the one-year follow-up, mainly because they dropped out or there were missing data. Only two patients died during the year. The remaining 307 patients were mean age 54 years, and 91% were women with breast cancer. The median dose of doxorubicin was 218 mg/m2, with no significant difference between the two arms. About half the patients received left chest radiation therapy, too. At one year, 19 patients in the EF group and 29 in the GLS group could not be included because of poor images, mainly caused by breast implants or chest wall sensitivity from treatment. At baseline, there was no significant difference in EF or GLS between the two arms.
At one year, EF was lower in the EF arm compared to the GLS arm (55% vs. 57%; P = 0.05). After an adjustment for baseline differences in comorbid conditions and their related therapy, this difference was not significant. However, 14% of the EF arm and 6% of the GLS arm met criteria for CRD at one year (relative risk reduction, 57%; 95% CI, 10-80%), with a number needed to treat of 13 to prevent one CRD event using the GLS-guided approach to employing CPT). Also, patients who received CPT in the EF arm recorded a greater reduction in EF than those in the GLS arm who received CPT (-9% vs. -3%; P = 0.03). In addition, in those meeting CRD criteria during follow-up, 45% of the EF arm and 14% of the GLS arm met CRD criteria on the final visit (P = 0.01) for a relative risk reduction of 70% (95% CI, 26-88%) with GLS-guided CPT. The authors concluded the results endorse the use of GLS rather than EF to determine which patients with at least one risk factor for heart failure should receive CPT during potentially cardiotoxic chemotherapy.
According to the authors, this was the first randomized, controlled trial of GLS compared to EF for guiding the decision to employ CRT in patients undergoing potentially cardiotoxic chemotherapy. However, based on retrospective, observational data, the American and European Societies of Echocardiography had endorsed GLS for this purpose in 2014.1 Indeed, our echo laboratory routinely measures GLS in chemotherapy patients, but we have not moved to using it as the sole criterion for initiating CRT. Should we? Based on the SUCCOUR study, the question remains open. This was a well-conducted study that was multicentered and international. It focused on high-risk patients for cardiotoxicity, and the authors enrolled a reasonable number of such patients. The centers used the same echocardiographic equipment, and the results were corroborated by a core echo lab. Also, in 60% of patients, the authors used 3D echo to calculate EF. However, clinical decisions were made locally by the treating doctors, which made this a real-world study. In addition, the authors randomized the screening strategy, not the CRT. Thus, it is unknown whether those given CRT would have developed CRD if it had not been administered.
A potential weakness is the study was not blinded, so one cannot exclude some biases. Additionally, the cutpoints the authors used for deciding what was a significant change in GLS warranting CRT were not well validated. The trial did not meet the primary endpoint after adjustment for other factors that could influence the results. The final EF and the change in EF from baseline to one year was the same for both groups. In this sense, it was a negative trial.
Despite this, the authors, based on a variety of secondary endpoints, concluded GLS and not EF should be used in the surveillance for CRD. Their explanation for the negative primary endpoint was that those without CRD recorded an increase in EF at one year, which skewed the results toward null. The authors did not explain why this increase in EF occurred in some patients. Other factors may have included the fact the patients in both groups were well managed. Only one patient in each group was hospitalized for heart failure, and few cut back or ended their chemotherapy. Yet more than twice as many patients in the GLS group initiated CRT vs. the EF group. One must wonder if subjecting all these patients to the risk and expense of CRT was worth it. Unfortunately, these authors did not provide sufficient data to support the use of GLS exclusively for surveillance and, if used at all, at what value CRT is indicated.
- 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;27:911-939.