Introduction
Capecitabine is a fluoropyrimidine-based oral prodrug of 5-fluorouracil (5-FU) used as an anti-neoplastic agent. It is currently approved for treatment of metastatic breast cancer and, according to the National Comprehensive Cancer Network guidelines, it is recommended for adjuvant treatment as monotherapy or in combination with other agents in advanced colorectal cancer (1).
The side effect profile of capecitabine varies from that of 5-FU. Capecitabine-induced myocardial toxicity is rare but includes serious adverse events. The few retrospective analyses published in the literature so far describe a wide range of cardiac manifestation, ranging from arrhythmias and acute ischemic events to heart failure (HF) and cardiogenic shock (2).
In this report, we present the clinical case of a patient developing ST-segment elevation myocardial infarction (STEMI) and acute HF resulting in shock as a consequence of capecitabine treatment, with a complete recovery of the myocardial function within 9 days after appropriate treatment. Our case is an example of potential capecitabine therapy cardiotoxicity followed by serious cardiovascular complications. Although capecitabine chemotherapy rarely causes cardiotoxic events, cardiologists must be aware of the potential complications due to the possible life-threatening cardiac events. Oncologists and cardiologists must maintain a high index of suspicion for less common presentations and collaborate closely, because some of these events may lead to a fatal outcome.
Case report
A 46-year-old patient was admitted to our clinic with severe chest pain and ST-segment elevation in the anterior leads. Two months before admission, he was diagnosed with malignant neoplasm of the rectum and underwent laparoscopic resection treatment of the anterior rectum cum colo-rectal anastomosis. After the surgical treatment, the patient was referred to the oncology department for further tests and chemotherapy. Treatment with capecitabine was initiated in a specific regimen: 1500 mg in the morning and 2000 mg at night per os. 3 days after the introduction this regimen, he presented with the current symptoms. The patient’s vital signs on admission were as follows: blood pressure (BP) 150/85 mmHg; heart rate 100/min; sPO2 96%; respiratory rate 22/min; body temperature 36.3 C. The admission ECG showed right bundle branch block (RBBB) with ST-segment elevation in the anterior leads (Figure 1). Laboratory findings showed hs-Troponin 3990 ng/L (normal values 15.6ng/l) and NT-proBNP 5251 pg/mL, and emergency coronary angiography was thus indicated.
In the meantime, the patient’s status deteriorated, with a drop in BP and elevation in the lactate (3.8 mmol/L). Emergency coronary angiography was performed, with finding of a thrombus in the left anterior descendent coronary artery (LAD). Subsequent percutaneous coronary intervention of the LAD was performed with endovascular prothesis implantation, and the final TIMI 3 flow result was obtained. The patient was in cardiogenic shock during the procedure, and transferred to the Intensive Coronary Unit (ICU) for further treatment. He was treated with inotropic and vasopressor support (dobutamine with noradrenalin) anticoagulant, antithrombotic therapy and statins. The next day, the patient’s arterial blood gasses (ABG) showed worsening, with further elevation in lactate values (8.3 mmol/L). Echocardiography examination showed a severe reduction in the left ventricular ejection fraction (EF) of 21% (Figure 2), with moderate mitral and tricuspid regurgitation. LV global longitudinal strain was severely reduced with a value of -5% (Figure 3). On the day of admission, the treatment with capecitabine was discontinued in consultation with an oncologist due to the suspected cardiotoxicity. Because of the constant hemodynamic instability, cardiac magnetic resonance was not feasible as diagnostic option.
On the second day spent in the hospital, a new onset of atrial fibrillation was noted; shortly after, a clinical worsening presenting with ventricular fibrillation followed by asystole after the electrical shock progressed to a complete reanimation. The patient was defibrillated several times within the cardio-pulmonary resuscitation protocol, and he was intubated and placed on mechanical ventilation as well as norepinephrine and dobutamine due to the cardiogenic shock. Because of the hemodynamic instability, the patient was transferred to Cardiac Surgery Department for further treatment with veno-arterial extracorporeal membrane oxygenation (ECMO) due to cardiogenic shock refractory to optimal inotropic and vasopressor support as the next step in the treatment. The patient’s parameters were stable and satisfactory in this setting, and we continued the close follow-up and proceeded without the need of mechanical circulatory support. Two days later, the patient was extubated with improvement in the ABG and general clinical status. Control bedside transthoracic echocardiography showed slight recovery of LV function. The treatment for HF started with perindopril 2 mg once a day (OAD), spironolactone 25 mg OAD, and furosemide 40 mg intravenous infusion, with continued antithrombotic and anticoagulant therapy with aspirin 100 mg OAD, clopidogrel 75 OAD, and enoxaparin 6000IE BID, with weaning off inotropes/vasopressors. ECG showed resolution of ST-segment elevation, RBBB with ST-segment descendent depression and a negative T wave in V1-V3 (Figure 4). On the 8th day, the echocardiography findings showed complete resolution of the LV dysfunction with hypokinesia on the anterior LV wall and an EF of 58%, along with mild mitral and tricuspid regurgitation and significant improvement of the global longitudinal strain (Figure 5).
After nine days and many challenging decisions about the clinical and therapeutic approach, the patient was discharged in a clinically stable state, with continued heart failure therapy (ACE inhibitor, MRA antagonist, beta blocker, dual antithrombotic therapy, and high dose statin), and with a recommendation for further oncology consultations and close cardiology follow-up.
Discussion
Capecitabine is a pyrimidine antimetabolite that inhibits thymidylate synthase (1-3). It is an oral prodrug which is enzymatically converted to active 5-FU in tissues, and it clinically resembles intravenous administration of 5-FU (4). The incidence of 5FU-induced cardiotoxicity in the literature ranges from 2.8% to 3.5%, although it is believed that the true number may be higher, given the fact that silent ST-segment deviations on electrocardiography are known to occur (5-8). Other manifestations of 5FU-induced cardiotoxicity include HF, hyper- or hypotension, cardiomyopathy, acute coronary syndrome (ACS), arrhythmias, conduction disturbances, and, less frequently, cardiogenic shock and cardiac arrest (8). Although cardiotoxicity is often reversible with discontinuation of the drug, fatal complications can still occur in a minority of patients, irrespectively of cumulative doses. Reported overall mortality rates range from 2.2% to as high as 13.3% (9). Most of the events are related to artery vasospasm and thrombus formation in the coronary arteries, and patients usually present with signs and symptoms of ACS (4,6). However, HF and cardiogenic shock requiring inotropic and vasopressor therapy have been described in the literature, raising concerns about a possible fatal outcome in some of the patients. McAndrew et al. reported a case series of patients with HF leading to cardiogenic shock shortly after exposure to capecitabine (9). Published case reports reveal that symptoms usually occur within two to three days after the initiation of capecitabine therapy (10). Risk markers for potential cardiotoxic effects have been examined in many studies. Kwakmall et al. performed a retrospective analysis and discovered that ischemic heart disease was a risk marker for cardiotoxicity in patients who received capecitabine (11).
The present case report highlights several key points that need to be taken into consideration in patients with suspected capecitabine cardiotoxicity. First, the initial presentation of the patient with chest pain, ECG changes, and significantly elevated cardiac markers are highly suggestive of ACS. Although vasospasm is the most common mechanism that causes this clinical scenario, few cases in the literature point out that, apart from vasospasm, coronary thrombosis could be observed after capecitabine treatment as a possible direct effect of this drug (12,13). Considering the fact that coronary vasospasm is not always the underlying mechanism, patients under capecitabine treatment presenting with ACS should be referred for primary percutaneous coronary intervention in order to prevent potential catastrophic events.
In our case, echocardiography demonstrated reduced LV ejection fraction and global hypokinesia, which did not correspond to the segmental distribution of the major coronary arteries. LAD stenting did not lead to improvement of the clinical condition of the patient, and once he was transferred to the ICU, he developed refractory shock despite the successful stenting. This raised the suspicion of another mechanism that was potentially responsible for the severe myocardial dysfunction. The treatment in this case was multidisciplinary, with the inclusion of cardiologists, oncologists, and cardiac surgeons. Prompt exclusion of capecitabine therapy and treatment of all cardiovascular complications lead to successful recovery of LV function and clinical stabilization of the patient.
Chemotherapy-induced Takotsubo cardiomyopathy is one of the potential causative mechanisms in patients with newly-developed cardiac dysfunction, but in our case the echocardiographic findings did not reveal the typical pattern-apical akinesia and ballooning with hyperdynamic base. Considering the full recovery of the myocardial function in nine days from the initial presentation, we believe that in our case the most probable mechanism of such severe myocardial depression is potential toxic myocarditis. Based animal models, a direct toxic effect on the coronary endothelium, or toxic myocarditis with vasospasm, have been proposed as possible mechanisms. A direct myocardial toxic effect attributed to the antimetabolite effects of the drug may provoke a toxic cardiomyopathic clinical picture (14).
Finally, individual sensitivity to cardiotoxicity could be a result of inherited variations in the enzyme pathways involved in the catabolism of fluoropyrimidines, leading to variable levels of cardiotoxic degradation products (15). In patients with dihydropyrimidine dehydrogenase (DPD) deficiency and colorectal cancer who have increased risk of cardiotoxicity, the onset of 5-FU toxicity usually happens rapidly, sometimes even within hours of the first dose. In such cases, an antidote, uridine triacetate, may be considered, and pretreatment DPD activity may be considered as well (15,16). Having in mind that capecitabine cardiotoxicity is common, patients should be closely followed up from the first cycle throughout the treatment. The follow-up should include electrocardiography and, according to the clinical symptoms, echocardiography as well, which allows the detection of subclinical cardiotoxicity. Patients with previous heart diseases should be monitored more closely.
One of the limitations in our case is the lack of CMR, which was not initially performed because of the rapid hemodynamic deterioration which necessitated the use of ECMO due to cardiogenic shock refractory to optimal inotropic and vasopressor support. Finally, the patient’s rapid recovery with complete return of the LV function in a few days and successful weaning of inotropic stimulation confirmed the initial suspicion of possible direct toxic myocardial injury. A myocardial biopsy may have also provided useful information, but performing such an invasive procedure in a hemodynamically unstable patient is extremely difficult and creates an additional risk.
The management of acute cardiogenic shock due to the cardiotoxic effect of capecitabine and 5-FU requires supportive inotropic and vasopressor therapy and immediate discontinuation of the drug. Reversal of cardiomyopathy over several days is usually expected, as has been described in this case. After such severe cardiac toxicity, rechallenge with 5-FU should not be attempted and is considered contraindicated.
Conclusion
This case highlights the importance of rapid diagnostics and considering potential capecitabine cardiotoxicity, with the inclusion of a multidisciplinary team in the management of serious cardiovascular complications. Acute HF leading to cardiogenic shock is a rare form of cardiotoxicity that should be considered as a possible complication. In such cases, capecitabine should be promptly discontinued. Patients who experience cardiotoxicity related to chemotherapeutic agents should be referred to cardio-oncology specialists in order to receive the best possible treatment without the risk of potential cardiotoxic complications, ultimately resulting in improved oncologic outcomes. Oncologists and cardiologists must maintain a high index of suspicion for less common presentations and should collaborate closely, as some of these events may cause a fatal outcome.