Metastatic Melanoma Presenting as Polymorphic Ventricular Tachycardia

Abstract

A 41-year-old woman with metastatic melanoma was admitted to the hospital because of syncopal episodes, which had developed after the administration of an experimental chemotherapy agent that targeted Notch signaling, as part of a phase I clinical trial. Cardiac monitoring revealed recurrent episodes of polymorphic ventricular tachycardia correlating with the patient's syncope. Investigations into the cause of the arrhythmia led to the discovery of metastatic lesions within the left ventricular myocardium. In presenting this case of polymorphic ventricular tachycardia as the antemortem clinical manifestation of metastatic melanoma involving the heart, we discuss the importance of recognizing that cardiac metastases can manifest themselves as arrhythmias in patients with malignant melanoma who are undergoing active anticancer treatment.

A woman with metastatic melanoma was admitted to the hospital because of syncope and was diagnosed with episodes of polymorphic ventricular tachycardia (VT). The episodes developed after the administration of an experimental chemotherapy agent that targeted Notch signaling as part of a phase I clinical trial. The patient was found to have metastatic lesions in her myocardium. We present and discuss this case of polymorphic VT as a manifestation of cardiac metastasis in malignant melanoma in a patient who was undergoing active anticancer treatment.

Case Report

In April 2011, a 41-year-old woman with a history of metastatic malignant melanoma was admitted to the hospital because of multiple syncopal episodes. Despite aggressive treatment with surgery and several chemotherapy regimens, including interferon and phase I clinical trial agents targeting proto-oncogene B-Raf for 9 years, her cancer had progressed. Two months before this presentation, the patient had begun therapy with temozolomide (an alkylating agent) and a small-molecule anti-angiogenic agent. Six weeks later, this treatment plan was abandoned because of disease progression: a computed tomographic scan of her brain showed multiple lesions suggesting intracranial metastases, the largest of which was in the cerebellum. After this discovery, the patient was started on whole-brain radiation and a new phase I chemotherapy regimen.

Four days after the patient underwent her first session of whole-brain radiation and took her first dose of the phase I chemotherapy agent—a γ-secretase inhibitor that targeted Notch signaling—she experienced a syncopal episode. She had a total of 2 syncopal episodes in 3 days. Each was preceded by palpitations, visual scotomas, and a rising sensation of warmth that would eventually culminate in loss of consciousness. She reported no bowel or bladder incontinence, chest discomfort, or shortness of breath. Each loss of consciousness lasted 1 to 2 minutes, with eventual return to mentation.

The patient had taken a 5-mg tablet of the phase I chemotherapy agent daily for 3 days. She had taken the last dose 3 days before this presentation. She reported that the syncope had begun only after she had started taking this drug.

Upon the emergent presentation, the patient's vital signs were stable; physical examination and a thorough neurologic evaluation yielded no unusual results. Results of laboratory tests, including electrolyte levels, were all within normal limits, and a computed tomographic scan of her brain revealed no acute intracranial pathology and a stable appearance of the brain metastases. An electrocardiogram (ECG) showed sinus rhythm with a normal QT interval and no evidence of ischemia (Fig. 1). The cause of the syncopal episodes was unclear, so she was admitted to the telemetry floor for further monitoring.

Fig. 1 Electrocardiogram upon emergent presentation shows sinus rhythm and a normal QT interval.

Within hours, the patient reported the same prodromal symptoms that she had previously experienced. Telemetry revealed multiple episodes of polymorphic VT (Fig. 2). She was immediately treated with magnesium sulfate and a lidocaine infusion and was transferred to the intensive care unit. The VT subsided within minutes after the therapy.

Fig. 2 Telemetry strip shows polymorphic ventricular tachycardia.

A repeat ECG was unchanged from that at admission. Electrolyte levels, particularly potassium, magnesium, and calcium, along with cardiac biomarkers of injury, including troponin, creatine kinase, and creatine kinase-MB fraction, were all within normal limits. An echocardiogram showed normal left ventricular systolic function with no significant valvular disease, regional wall-motion abnormalities, or masses. Further investigation, including cardiac computed tomography, revealed no obstructive coronary artery disease; however, a focal low-attenuation nodule involved the anterior wall of the left ventricle and was strongly suspected to be metastatic. Follow-up cardiac magnetic resonance (CMR) confirmed a metastatic nodule in the anterior wall of the left ventricle, 8 to 10 mm in its widest diameter. The CMR also revealed a 2nd nodule of approximately 1-cm diameter in the inferior wall near the apex (Fig. 3).

Fig. 3 Magnetic resonance images show A) a metastasis (arrow) in the anterior wall of the left ventricle and B) a metastasis (arrow) in the inferior wall of the left ventricle near the apex.

Further investigation revealed no other cause of the arrhythmia, and it was concluded that the polymorphic VT was secondary to the cardiac metastases. Given the temporal correlation of the patient's use of the phase I chemotherapy agent to the onset of the symptoms, it was thought that the phase I drug had exacerbated the polymorphic VT. As a result, that protocol was discontinued; however, the patient continued to have intermittent episodes of polymorphic VT for 9 days after her last dose.

The patient was subsequently treated with β-blockers, amiodarone, and a wearable external defibrillator as secondary prevention against sudden cardiac death. Two months after discharge from the hospital, the arrhythmia had not recurred despite the progression of her cancer. She died 6 months later.

Discussion

Malignant melanoma is aggressive and has the highest rate of cardiac involvement of all neoplasms.¹ Autopsy studies have shown that from 64% to 75% of patients with malignant melanoma have cardiac metastases at the time of death.^2,3 Most commonly, it metastasizes to the right-heart chambers and can involve any of the layers of the heart. However, in cases in which cardiac involvement is documented, the myocardium is involved 98% of the time, with epicardial and endocardial involvement 78% and 73% of the time, respectively.^2–4 The clinical manifestations of cardiac metastases can be nonspecific; many patients remain asymptomatic, so antemortem diagnosis is often difficult.

Polymorphic VT is a distinctive ventricular tachyarrhythmia that is characterized by constant changes in QRS morphology.⁵ In its most common clinical form, torsades de pointes, the arrhythmia is associated with a prolonged QT interval. Congenital and acquired forms of QT prolongation have been described.^6,7 The acquired form—the one seen more frequently in adults— is often associated with agents that cause bradycardia or electrolyte abnormalities (particularly hypokalemia and hypomagnesemia) or that prolong myocardial repolarization.^6,8

Polymorphic VT can also occur in the presence of a normal QT interval. Most such cases are associated with coronary artery disease.^9–11 Polymorphic VT can be a manifestation of acute ischemia or stable coronary artery disease with prior myocardial damage and scar formation. The mainstay of treatment is myocardial revascularization.⁵

To our knowledge, this is the first report of a patient who presented with polymorphic VT as a manifestation of cardiac metastases in malignant melanoma. A review of the medical literature reveals that complex ventricular arrhythmias are rarely seen in patients with metastatic melanoma.^12,13 The melanoma lesion itself could have been a nidus of the arrhythmia in our patient. Whereas structural heart disease is well known to cause monomorphic VT, the mechanisms underlying the initiation and maintenance of polymorphic VT are less clear.

Theoretical models suggest that polymorphic VT is the result of a single-reentrant focus rather than an automatic focus. The malignant reentrant foci may cause heterogeneities in the electrophysiologic properties of myocardial tissue, thus promoting the initiation of VT and maintaining reentry. Once the arrhythmia is initiated, the constant change in QRS morphology can be explained by the concept of spiral wave drift. In this phenomenon, a nonstationary core of electrical activity results in changes in VT cycle length, giving rise to ECG patterns of polymorphic VT.¹⁴

Alternatively, the malignant foci might act similarly to an ischemic scar with border-zone Purkinje fiber involvement. These fibers exhibit increased automaticity and triggered activity. These findings, when coupled with prolonged action-potential duration in the region, may result in the necessary milieu for polymorphic VT.^15,16

Another possibility is that the arrhythmia in our patient resulted from tumor cell cytotoxicity due to biologically active therapy, which produces a local inflammatory response that could increase arrhythmogenic potential, as seen in patients with myocarditis.¹⁷ Melanoma patients treated with interleukin-2 and interferon-γ have experienced local inflammation,¹⁸ which can lead to a reversible cardiomyopathy and predisposition to the development of organized atrial or ventricular arrhythmias, seen in up to 6% of reported cases.^19–21

A 3rd possibility is that the toxicity in our patient relates in part to a direct, on-target effect of the phase I chemotherapy agent, a γ-secretase inhibitor. On a cellular level, γ-secretase inhibitors block activation of the protein Notch. Notch proteins have been implicated in tumor genesis and appear to function as either oncogenes or tumor-suppressor proteins, depending on cellular context. Because γ-secretase inhibitors are still in early development without widespread clinical use, the exact side-effect profile of this class of agents is unknown. Results of preclinical studies have indicated a crucial role of the Notch signaling pathway in the development of the cardiac conduction system.^22,23 Accordingly, it is mechanistically plausible that a proarrhythmic phenotype in subsets of patients treated with γ-secretase inhibitors signifies a previously undescribed role of the Notch signaling pathway in the maintenance of normal cardiac conduction in adults.

Our patient underwent careful investigation into the most common causes of polymorphic VT. It is likely that a combination of factors was responsible for her arrhythmia. Clinicians should be aware that the presence of cardiac metastases could increase the risk of arrhythmia in a patient with malignant melanoma.