Myocardial Dysfunction in Breast Cancer Survivors: ‘You Can Observe a Lot by Just Watching’

This article refers to ‘Myocardial dysfunction in long-term breast cancer survivors treated at ages 40–50 years’ by J.N. Jacobse et al., published in this issue on pages xxx. Anthracyclines have entered the cancer therapeutic armamentarium over half a century ago, added to surgical and radiotherapy techniques. They are highly effective chemotherapeutic agents that possess marked antitumour activity, but also are hampered by significant off-target toxicities.1 Despite recent advances in sophisticated molecularly targeted and immune-mediated oncologic therapies, anthracyclines continue to play a prominent role in the management of less differentiated subtypes of breast cancer (BC), such as oestrogen receptor-, progesterone receptor-, and her2-negative tumors.2 Off-target cardiotoxicity leading to progressive cardiomyopathy and ultimately heart failure has plagued the utilization of anthracyclines since the inception of their clinical application.3, 4 Since the degree of cardiotoxicity is directly related to the cumulative anthracycline dose,5 standard protocols limit the exposure. However, even relatively low anthracycline doses, like 240 mg/m2 used in the current standard of care for BC treatment, have a risk for inducing cardiomyopathy (defined as an absolute left ventricular ejection fraction (LVEF) reduction of 10 points to below the normal limit of 53%, or a relative drop in the left ventricular global longitudinal strain (GLS) >15% compared to baseline) as high as 43% in high-risk individuals.6, 7 With continually improving survival after cancer diagnosis and treatment, the long-term focus has been shifting at the intersection between oncology and cardiology toward detection, monitoring, and management of cardiac dysfunction in cancer survivors exposed to anthracycline therapy. Growth in this field is well-illustrated by a large number of ongoing prospective studies (Table 1). In this issue of the Journal, Jacobse and colleagues have conducted a cross-sectional cohort study aimed at clarifying the prevalence and risk factors for cardiac dysfunction in 569 women diagnosed and treated for either invasive BC (TNM I–III) or ductal carcinoma in situ (DCIS) at age 40–50 years old at either the Netherlands Cancer Institute – Antoni van Leeuwenhoek or University Medical Center Groningen between 2002 and 2012.8 Women were included in two groups based on time from treatment: 5–7 or 10–12 years. Potential participants were excluded if they had additionally received chemotherapy or radiation for a reason unrelated to BC/DCIS, recurrent BC/DCIS, or a second malignancy; or with a history of cardiovascular disease before the BC/DCIS diagnosis. Eligible women were invited to participate and completed a detailed questionnaire on their medical and psychosocial histories. In addition to a comprehensive history and physical examination, detailed data on their BC/DCIS tumour and treatment characteristics were abstracted from their medical records, hospital registries, or directly obtained from the participants' medical providers. At the study visit, all participants had their amino-terminal pro-B-type natriuretic peptide (NT-proBNP) levels measured in addition to a centrally analysed echocardiographic evaluation that included standard techniques for determining LVEF and GLS. This analysis replicates several previously reported observations regarding the risk for cardiac dysfunction in a large cohort of young cancer survivors. Unique advantages of this cohort are the selection of low cardiovascular baseline risk study participants and sensitive imaging and blood biomarkers to specifically address the long-term effect of anthracycline chemotherapy in the current practice. First, increasing cumulative anthracycline dose was linearly associated with lower LVEF, more impairment in GLS, and higher NT-proBNP levels. Importantly, these observations were independent of age, diastolic blood pressure, body size, and trastuzumab – a common therapy for BC/DCIS known to cause cardiac dysfunction.9 Second, left breast radiation was associated with additional impairment in GLS. This observation highlights the importance and utility of comprehensive strain-based echocardiography for monitoring cardiac dysfunction in women that received radiation as part of their BC/DCIS treatment. Third, in an otherwise low-risk population, the prevalence of elevated NT-proBNP (43.4%) was higher than an impairment in GLS (26.6%), or an LVEF <54% (7.6%) – findings which support the utility of NT-proBNP and of strain echocardiography to detect subclinical cardiac dysfunction in comparison to traditional LVEF measurement.10-13 Observations from epidemiological studies have highlighted the link between anthracycline exposure and a spectrum of cardiotoxicities from impairment in GLS to severe cardiomyopathy over a timeframe ranging from the acute setting to many years post-treatment.14-17 The report by Jacobse et al.8 solidifies these observations, and provides complementary data. The cardio-oncologic protocolised study design is its primary strength by attempting to capture thorough oncologic histories and creating an echocardiography and biomarker repository, not available in this combination in other ongoing larger scale cardiac cohort studies. The strong, dose-dependent association between cumulative anthracycline dose and directionally abnormal measures of LVEF, GLS, and NT-proBNP clearly affirms the causal nature of anthracycline use and cardiac dysfunction. Deliberate design with a selection of long-term survivors of BC/DCIS is a clear limitation of this analysis. Yet, the same selection or survival bias also highlights the practical clinical applicability of these observations for individuals that may otherwise be at natural risk for cardiovascular events. With continued improvements in treatment strategies for BC/DCIS, there has been and will be a continuing parallel increase in BC/DCIS survivors. As anthracyclines continue to be used in oncologic treatment strategies – sometimes in complex combination protocols with immuno-oncologic approaches, a better understanding of the diagnosis, monitoring, and management of anthracycline-induced cardiac dysfunction is essential. The present report marks a solid starting point toward this better understanding, although it suffers at this stage from a lack of necessary longitudinal imaging- and biomarker-based assessments. Because this study proposes to clarify the trajectories of these measures over time, it may then have the capacity to answer several important questions. What are the optimal modalities for screening for cardiac dysfunction? Is imaging sufficient? Are biomarkers sufficient? How frequently should these individuals be screened? What is the long-term risk of incident cardiovascular disease in long-term survivors of BC/DCIS? What interventions might prevent these untoward cardiovascular consequences? Fortunately, studies like the presently enrolling studies in Table 1 will complement observations from the study by Jacobse et al. with detailed longitudinal molecular, imaging, and functional status phenotyping coupled with preventive interventions in order to answer some of these crucial questions. Similar to the then-groundbreaking BC treatment approaches employed Bernard Peyrilhe in the 18th century, observations from this study and others currently enrolling point the way forward. They also underscore the need for well-designed studies to meet these ends and remind us of the old adage spoken by the New York Yankees legendary catcher, Lawrence Peter ‘Yogi’ Berra (1925–2015), ‘you can observe a lot by just watching.’ J.L.G. has grant support from the Texas Health Resources Clinical Scholarship. G.Z. acknowledges research funding from the Cancer Prevention Research Institute of Texas. Conflict of interest: J.L.G. has consulting income from Pfizer. The other authors have nothing to declare.