Cancer and myocardial dysfunction: is there a link?

KEYWORDS:

• Cancer
• myocardial dysfunction
• cardiovascular biomarkers
• myocardial strain
• prognosis

Cancer is the second most common cause of death after heart disease and its prevalence is steadily increasing. Over the next 20 years, it is estimated that the number of new cases will rise by 70% [1]. However, the increase in the incidence of cancer is being paralleled by significant improvements in survival rates. In the United States, over the next decade, the number of cancer patients surviving more than five years from diagnosis is expected to increase by about one-third [2]. Advances in the earlier detection of cancer and a widening array of antineoplastic therapies are primarily responsible for the improved survival.

It is well known that certain anti-cancer drugs, most notably anthracyclines, may have direct toxic effects on myocardial function. The introduction of molecular targeted therapies, although leading to improved efficacy in specific cancers, has further complicated the cardiovascular toxicity profile of anticancer treatment regimes. Given the resultant improved survival from cancer, the potential cardiotoxic effects may assume a more important role in determining long-term outcome, independent of the neoplastic disease process [3,4]. Consequently, cancer patients at a perceived increased risk of cardiovascular toxicity are increasingly being treated collaboratively by oncologists and cardiologists. Whilst it is well known that several cancer drug therapies are potentially cardiotoxic, recent studies have raised the question as to whether cancer per se may directly affect myocardial function. Although there are currently no clear data to affirm a direct link between cancer and myocardial dysfunction, in this article, we discuss the hypothesis generating studies that may support this relationship.

A recent study by Pavo et al. systematically measured a panel of serum biomarkers prior to commencement of anticancer therapy in 555 patients with a broad range of cancers [5]. These biomarkers included inflammatory markers, vaso-active peptides, cardiac hormones, and troponin T, a marker of myocardial injury. The levels of these biomarkers were related to advancing tumor stage and the ability of these biomarkers to predict all cause death was assessed over a median of two years follow-up. The cardiovascular functional peptides N-terminal pro-B-type natriuretic peptide (NT-pro-BP), mid-range-pro atrial natriuretic peptide (MR-pro-ANP), mid-range pro-adrenomedullin (MR pro-ADM), C-terminal pro-endothelin-1 (CT-pro-ET-1), and copeptin, along with high-sensitivity troponin-T (hs TnT), had independent predictive value for all cause death after correcting for age, gender, type of cancer, stage of disease, and cardiac comorbidities. In addition, NT pro-BNP, MR pro-ANP, and MR pro-ADM were positively correlated with the inflammatory cytokine interleukin-6 and the inflammatory marker C-reactive protein, as well as with hs-TnT. Elevation of natriuretic peptides is a characteristic feature of heart failure and raised troponin levels are indicative of myocyte injury. On a broad scale, therefore, these findings may suggest an interplay between a pro-inflammatory state, vaso-active processes, myocyte injury, and, ultimately, myocardial dysfunction. In addition, there was a progressive increase in these biomarkers with more advanced tumor disease, further raising the possibility that the malignant disease process per se may have a direct effect on myocardial function, prior to any influence from anticancer drug therapy. However, a significant limitation of this study was the lack of a systematic evaluation of myocardial function to determine whether the elevations in the biomarkers correlated with myocardial dysfunction. Echocardiography was performed only in those with a NT-pro BNP level >400 pg/ml. Although the investigators excluded what they refer to as non-stable cardiac disease, 25% of the study population had an abnormal cardiac status. It is possible therefore that some patients had significant preexisting cardiovascular disease which may have impacted on the results. Moreover, cause of death was not documented and so it is difficult to know to what extent the raised biomarkers were predicting cardiovascular mortality.

A potential contributor to cardiac impairment in cancer could be cachexia which often accompanies advanced cancer states. It is postulated that cardiac dysfunction and atrophy parallels the weight loss and skeletal muscle atrophy seen in cancer cachexia [6]. A growing body of evidence from animal models of cancer cachexia have shown a direct deleterious effect of tumor growth on myocardial function. In cancer-bearing mice, the tumor growth directly alters myocardial function at the cellular level, and alterations in cardiomyocyte function underlie changes in myocardial function detected in vivo by echocardiography [7,8]. In an animal model of hepatoma, elevated levels of BNP were attributed to cardiac cachexia associated with cardiac remodeling, deterioration of cardiac function, and cardiac death [9]. Treatment with spironolactone and beta-blockers, targeting established neurohumoural pathways of heart failure, succeeded in reducing mortality. A recent large cross-sectional study of patients who had head and neck cancer surgery showed that after controlling for all other variables, patients with weight loss had an approximate 30% increased risk of acute cardiac events compared to patients without weight loss [10]. In the study by Pavo et al., over half of the patients had advanced cancer (tumor stage 3 or 4) and half of the patients had BMI <25 implying significant cachexia. Moreover, there was a relatively high 2-year mortality of 34% further indicating more advanced disease. It may be postulated that a significant proportion of the patients in this study had a degree of cachexia-related myocardial dysfunction and an increased risk of cardiovascular death. Aside from myocardial dysfunction, autonomic imbalance affecting the cardiovascular system may also play a role in clinical outcome. In a study of patients with colorectal cancer and cancer cachexia, heart rate variability, a marker of sudden cardiac death, was significantly reduced compared to controls [11]. In this study, although ejection fraction was also lower than the control group, the absolute measure was within the normal range and not of clear clinical importance.

Newer imaging modalities now make it possible to assess myocardial mechanics in greater detail. The measurement of ejection fraction, although traditionally used to detect chemotherapy-induced cardiotoxicity, has been shown to be a relatively insensitive marker of early myocardial dysfunction. The measurement of myocardial strain by 2D and 3D speckle tracking echocardiography enables the detection of earlier, more subtle abnormalities in myocardial function [12]. Systolic strain measures the percentage change in length of myocardial fibers in longitudinal, radial, and circumferential directions during left ventricular contraction. Global longitudinal strain which measures myocardial shortening of left ventricular segments from base to apex appears to be the most reliable and clinically applicable parameter for determining early reductions in myocardial function, although reductions in global radial and circumferential strain have also been implicated [13]. Numerous studies in cancer patients on potentially cardiotoxic chemotherapy have shown that reductions in myocardial strain occur prior to any discernible change in ejection fraction and may predict future left ventricular dysfunction [13]. However, there are currently no published data on myocardial strain in as yet untreated cancer patients. Preliminary data from our institution in treatment-naïve cancer patients with normal ejection fraction have shown reductions in myocardial strain compared to controls. It would be of great interest to correlate elevations in cardiovascular biomarkers with these more subtle markers of myocardial dysfunction. Detection of subtle myocardial fibrotic changes with cardiac magnetic resonance imaging using late gadolinium enhancement and T1 mapping techniques may provide additional insight [13].

Despite the known diagnostic and prognostic value of the natriuretic peptides with left ventricular dysfunction and overt heart failure, it is possible that within the complex milieu of neoplastic disease, the processes leading to elevations of vaso-active peptides and cardiovascular biomarkers may not relate directly to myocardial dysfunction. Recent small studies have shown that some of these markers may be expressed from tumor cells and play a role in the preservation and proliferation of the malignant disease process. Adrenomedullin, endothelin-1 and vasopressin are thought to promote tumor progression through regulation of angiogenesis [14–16]. Indeed, adrenomedullin may be a future therapeutic target and marker of tumor activity. The natriuretic peptides may play an important role in regulating tumor cell growth and inflammation [17]. Elevations in troponin T have also been noted and associated with a worse prognosis [18]. In all of these observations, although the elevations of these biomarkers have not been associated with clear cardiac involvement, a detailed evaluation of myocardial function was not undertaken.

One may postulate from the available data that the increasing levels of cardiovascular biomarkers with more advanced disease, observed in the Pavo study, reflect a composite of cancer disease progression and associated myocardial dysfunction, with one or other feature being more dominant in a given patient. The measurement of NT-pro-BNP and hs-TnT is a recognized method for detecting cardiotoxicity during anticancer drug treatment. A major clinical implication of the current data is the need to measure these biomarkers at baseline, prior to commencement of anticancer medication. Subsequent elevations in natriuretic peptides during anticancer drug therapy should be interpreted with caution as on the one hand they might indicate myocardial dysfunction and on the other hand tumor progression. Detailed clinical evaluation and a full assessment of ventricular function not only with conventional echocardiography but also strain imaging may help to determine the extent to which the myocardium is involved.

Future research is needed in treatment-naïve cancer patients to establish the relationship of cardiovascular biomarkers with early markers of myocardial dysfunction, autonomic function, exercise performance, measures of cachexia, cancer type, and tumor burden. The relative predictive value of these variables on cardiovascular morbidity and mortality would be of great interest. In addition, serial measurements of cardiovascular biomarkers and myocardial strain may provide additional insight into the impact of anticancer drug therapies on myocardial function and tumor burden.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Additional information

Funding

This paper was not funded.