Author + information
- Heart Failure Center and Cardio-Oncology Center, Harrington Heart & Vascular Institute, University Hospitals Health System, Case Western Reserve University School of Medicine, Cleveland, Ohio
- ↵∗Address for correspondence:
Dr. Guilherme H. Oliveira, Heart Failure Center and Cardio-Oncology Center, Harrington Heart & Vascular Institute, University Hospitals Health System, Case Western Reserve University School of Medicine, 11100 Euclid Avenue, LKS 3012, Cleveland, Ohio 44106.
Effective prevention of left ventricular systolic dysfunction and heart failure (HF) from cancer therapies remains the Holy Grail of cardio-oncology. For almost half a century, chemotherapy-induced cardiomyopathy has been the downfall of anthracycline-based therapies. Early experience with doxorubicin firmly established a dose-dependent cardiotoxic effect that could lead to early discontinuation of therapy or end-stage HF in cancer survivors (1). Despite this, anthracyclines have remained a mainstay of cancer therapy for patients with hematologic malignancies, sarcomas, and breast cancer, among others. More recently, tyrosine kinase inhibitors and monoclonal antibodies—in particular, human epidermal growth factor receptor 2 (HER2) inhibitors—have also been associated with clinically important cardiotoxicity (2).
Over the last 2 decades, cardiologists and oncologists partnered to establish the field of cardio-oncology with the overarching goal of helping prolong the lives of cancer patients and survivors. This concerted effort led to growing recognition of the cardiovascular consequences of cancer treatment, a rapidly accumulating body of scientific evidence, and the explosive proliferation of cardio-oncology programs around the world. Although cardio-oncology has since expanded its mission and reach to include management of all cardiovascular aspects of cancer patients, cardiotoxicity has endured as its centerpiece. As a result, much has been learned about anthracycline and trastuzumab cardiotoxicity; increasingly referred to as cancer therapeutics-related cardiac dysfunction (CTRCD). For example, the pathophysiology of anthracycline-induced cardiac damage has been found to be predominantly mediated by topoisomerase (Top) 2ß (3). Anthracycline antibiotics indiscriminately inhibit both Top 2α in rapidly replicating neoplasia, and Top 2ß in quiescent cardiomyocytes, causing double-stranded DNA breaks and killing both. In addition, Top 2ß is also implicated in reactive oxygen species production, activation of the p53 survival pathway and, once deleted from mouse hearts, affords protection against anthracycline cardiotoxicity (4). Similarly, human epidermal growth factor (HER2/ERbB2) inhibition impairs cardiomyocyte resistance to stress, rendering them more susceptible to apoptosis (5). Concomitant or sequential use of these agents have additive cardiotoxicity that may be mechanistically linked through Top 2ß as well.
Despite better understanding of the basic mechanisms of cardiotoxicity, translation into development of agents to prevent CTRCD has remained elusive. In view of this, cardio-oncologists have sought chemoprevention among the “miracle drugs” that recover failing hearts and prolong life of patients with HF: ß-adrenergic blockers (BBs), angiotensin converting-enzyme inhibitors (ACEIs), angiotensin receptor blockers (ARBs), and mineralocorticoid receptor antagonists (MRAs): collectively known as neurohormonal antagonists. The trouble with this strategy is that, mechanistically, it requires a leap of faith. Whereas cardiotoxicity involves cardiomyocyte dysfunction and death mediated by DNA breaks, inhibition of cellular survival pathways, and activation of apoptosis, neurohormonal therapies appear to lack the mechanistic capabilities to counteract these events at the cellular level. Although carvedilol has been shown to reduce doxorubicin-induced cardiomyocyte apoptosis in vitro (6), similar data are lacking for other BBs and ACEIs/ARBs.
Regardless of absent robust biological plausibility, multiple small and medium-sized studies have been performed to test the hypothesis that neurohormonal modulation with BBs and/or ACEIs/ARBs can prevent or attenuate CTRCD. Even more surprising, numerous position papers, society guidelines, and expert consensus have been published attempting to standardize and guide the approach to prevention of cardiotoxicity in the clinical setting. In this context, further evidence-based knowledge in cardio-oncology is very much welcome.
In this issue of JACC CardioOncology, Vaduganathan et al. (7) present a meticulous and contemporary meta-analysis of 17 randomized controlled trials in an earnest attempt to settle the question of neurohormonal chemoprevention in cardiotoxicity once and for all. Unfortunately, through no fault of the authors, the strength of the analyzed evidence is insufficient to draw a definitive conclusion. Amidst high heterogeneity, with inconsistency indices upwards of 90%, substantial publication bias, and only modest numbers of randomized patients in each trial, the authors found a small but statistically significant benefit favoring neurohormonal chemoprevention. Even though statistically significant, the clinical relevance of their findings is less certain and more difficult to interpret.
After pooled analysis, patients treated with neurohormonal therapies had a left ventricular ejection fraction (EF) at follow-up 3.96% higher than the control group, with negligible changes in left ventricular dimensions. Global longitudinal strain was only measured in 3 studies and therefore could not be adequately interpreted. Four different types of BBs were studied: carvedilol, metoprolol, nebivolol, and bisoprolol. Of these, carvedilol was the most frequently studied in 8 of 12 trials involving BBs. Similarly, 5 ACEIs/ARBs were tested, of which enalapril was studied 4 times; candesartan twice; lisinopril, perindopril, and telmisartan once. One trial tested spironolactone against placebo. The results of both BB and ACEI/ARB trials were conflicting: some showing benefit, others not. At the end, using rigorous statistics, there appeared to be a modest benefit toward using neurohormonal therapies to prevent cardiotoxicity.
Interestingly, the incidence of significant cardiotoxicity reflected by EF decrements at follow-up was small. Only 2 trials reported mean EF of <50% at follow-up among the control groups, and most had no or very minimal EF decrements from baseline. The reasons for this finding may reflect a true low incidence of cardiotoxicity, very low doses of anthracyclines, or intrinsic patient referral bias in which predominantly healthy and low-risk patients were enrolled in these trials. Because doses and complete cardiovascular risk profiling were not available in most studies, this question remains unanswered for now.
Although not able to provide definitive recommendations regarding the role of neurohormonal agents to prevent cardiotoxicity, this meta-analysis indeed educates the cardio-oncology community. It highlights the amount of work still to be done. It sheds light on the gaps of knowledge that afflict the field, even over a decade since its inception. It alerts to the fact that, in cardio-oncology, practice patterns have outpaced scientific evidence and guidelines are ahead of their time; in fact, the cart may be ahead of the horses. This work offers a sobering appraisal of where we stand and where we need to go. From the trials analyzed here, one can safely conclude that small single-center trials will not provide definitive answers and that the question is more complicated than initially believed.
In conclusion, today there is no definitive evidence to support using or withholding neurohormonal antagonists as cardioprotective agents. For the time being, cardio-oncologists will continue to try them until large-scale, multicenter, randomized control trials unveil the truth or myth of cardioprotection with neurohormonal therapies.
↵∗ Editorials published in JACC: CardioOncology reflect the views of the author and do not necessarily represent the views of JACC: CardioOncology or the American College of Cardiology.
The author has reported that he has no relationships relevant to the contents of this paper to disclose.
- 2019 The Author
- Hudis C.,
- Seidman A.,
- Paton V.,
- et al.
- Vejpongsa P.,
- Yeh E.T.
- Vaduganathan M.,
- Hirji S.A.,
- Qamar A.,
- et al.