Life-threatening immune checkpoint inhibitor-induced myocarditis and myasthenia gravis overlap syndrome treated with abatacept: a case report

Abstract

We present here the second documented case of severe immune checkpoint inhibitor-induced myocarditis successfully treated with abatacept. The patient was started on pembrolizumab for stage IIIA malignant melanoma, and after the first dose was admitted for worsening shortness of breath and weakness. Her symptoms were refractory to high-dose steroids and she decompensated rapidly necessitating cardiopulmonary resuscitation and subsequent intubation and mechanical ventilation. Intravenous immunoglobulin and plasmapheresis did not invoke significant improvement, so abatacept was then initiated. She began to show improvement and was eventually discharged to a skilled nursing facility. This case highlights a severe adverse reaction to an immunomodulator class steadily growing in its application. Providers of all specialties should be aware of the side effects and treatment options. Our case demonstrates that continued investigation into the utilisation of CTLA-4 agonists in the treatment of severe adverse reactions like myocarditis caused by pembrolizumab is required.

Background

Immune checkpoint inhibitors (ICIs) are a growing class of immunomodulators that have become standard of care in the treatment of various malignancies. While this non-chemotherapy option has broadened our therapeutic arsenal, they are not without adverse effects. Myocarditis is a rare but often fatal side effect of ICIs, with limited treatment options. Case reports have suggested the use of immunomodulators such as abatacept, a CTLA-4 agonist, as a treatment option for steroid-refractory severe ICI-induced myocarditis with relative success. We discuss here a case of severe ICI-induced myocarditis and myasthenia gravis (MG) overlap syndrome treated with abatacept, with significant patient improvement.

Case presentation

A 55-year-old female with a medical history of hypertension presented to her local provider with concern for a mole on her left thigh. Punch biopsy was revealing of malignant cells, so she underwent wide local excision of the lesion with sentinel lymph biopsy revealing 1 of 1 lymph node positivity. Repeat lymph node sampling revealed 6 of 6 lymph nodes being negative. She was subsequently diagnosed with stage IIIA (pT2apN1aM0) malignant melanoma, BRAF V600E mutated. After establishing with an oncologist, she initiated a course of adjuvant pembrolizumab. Three weeks following her first dose, our patient presented to her oncologist complaining of diplopia, left-sided facial droop and ipsilateral lower extremity weakness. The second dose of pembrolizumab was held, and an MRI of the brain was obtained with unremarkable results. One dose of 20 mg intravenous dexamethasone was administered followed by a course of 8 mg orally two times a day. Five days later, she developed worsening shortness of breath and severe muscle cramps, prompting direct admission to her local hospital. Steroids were discontinued. Admission work-up revealed elevated troponins at 10 000 ng/L with non-specific ST changes on 12-lead ECG. A CT pulmonary angiogram excluded pulmonary embolism. Cardiac catheterisation did not reveal atherosclerotic coronary artery disease as a culprit of myocardial infarction, and left ventricular ejection fraction was estimated to be 55%. High-dose steroids were restarted on day 5 of admission with minimal symptomatic improvement. Subsequent ECGs and telemetry revealed a transient third-degree heart block. On day 6 of admission, she developed pulseless ventricular tachycardia necessitating cardiopulmonary resuscitation and defibrillation. She was transferred to our tertiary care facility after return of spontaneous circulation.

On arrival at our facility, ECG showed persistent ST changes with eventual progression to third-degree heart block (figure 1). Repeat transthoracic echocardiogram (TTE) showed an ejection fraction of 33% (figure 2). She was intubated on day 8 for increasing muscle weakness and concern for impending respiratory decompensation. Additional work-up of her weakness at this time revealed a C reactive protein of 16.3 mg/L and elevated anti-acetylcholine antibodies (0.76 mmol/L) and anti-striated muscle antibodies (1:61440). Plasmapheresis was initiated on day 9 for a total of 2 consecutive days with minimal symptomatic response. Abatacept, a CTLA-4 agonist (500 mg intravenously), was added after the second treatment of plasmapheresis on day 10. Her muscle strength began to improve, troponins trended down and she was extubated to nasal canula on day 12. Abatacept was continued for a total of five doses every 2 weeks. On day 15, intravenous immunoglobulin (IVIG) was added and she received five doses for 5 consecutive days with continued improvement. She underwent tracheostomy placement due to continued weakness and biventricular implantable cardioverter defibrillator (ICD) placement for third-degree heart block and secondary prophylaxis for life-threatening arrhythmia. Repeat TTE on day 32 of admission showed complete recovery of left ventricular function with an ejection fraction of 55%. After 51 days of inpatient care, she was discharged to a skilled nursing facility on stable ventilator settings after significant improvement in her weakness and shortness of breath. She remained on a slow steroid taper/maintenance schedule and pyridostigmine.

Figure 1.

Twelve-lead ECG demonstrating third-degree heart block on day 8 of admission.

Figure 2.

A transthoracic echocardiogram (TTE) in a parasternal long-axis view of our patient’s heart on arrival at our facility showing a dilated left ventricle. Signs of myocarditis on TTE are non-specific, but when present are often seen as global left ventricular dysfunction including dilatation, hypokinesis or hypertrophy. Right ventricle hypokinesis or hyperkinesis can also be observed, along with pericardial effusions and ventricular thrombi.

Outcome and follow-up

Pembrolizumab was never resumed due to the grade 4 adverse reaction, and our patient remains on active surveillance for melanoma. Our patient continues to improve clinically 15 months after her adverse reaction to pembrolizumab with no evidence of recurrence of melanoma. Her tracheostomy was reversed and she is able to ambulate by herself with the support of a cane at home. For cardioprotective measures, she was discharged on metoprolol and maintains regular long-term follow-up appointments in our cardio-oncology clinic with surveillance cardiac imaging as warranted clinically.

Discussion

Endomyocardial biopsy or cardiac MRI was not obtained due to the severity of illness. Despite this, the patient’s clinical presentation, positive cardiac markers, ECG changes and TTE findings support a diagnosis of myocarditis. The most common causes of myocarditis include infectious, autoimmune, drug-induced hypersensitivity and systemic disease (sarcoid). A good patient history and initial basic work-up can stratify these aetiologies accordingly. ICI-induced immune-related adverse events (irAEs) are common, can affect any tissue in the body and are often under-reported.¹ Our patient lacked a history of autoimmune disease, heart disease, common medications implicated in autoimmunity, viral/bacterial prodrome and associated cultures. Her acute onset of illness following one cycle of pembrolizumab, MG-myocarditis overlap syndrome and rapid deterioration support pembrolizumab as the cause.

The cornerstone of treatment for severe irAEs (CTCAE V.5.0) as outlined by the American Society of Clinical Oncology (ASCO) are high-dose steroids,² but a steroid refractory syndrome is not uncommon, as demonstrated in our case. ASCO clinical practice guidelines recommend treating irAEs with 1 mg/kg daily of intravenous or oral steroids and then transitioning to a taper of 4–6 weeks.³ It is not uncommon for a viral myocarditis to receive steroid treatment and taper over 3 months to 1 year while monitoring troponin levels to suggest response to therapy. A formal consensus on steroid dosing and taper recommendations does not yet exist for ICI-induced myocarditis.³ Regarding irAEs, evidence-informed guidelines on the timing of plasmapheresis and IVIG administration are lacking. Presently, the treatment of severe irAEs relies heavily on provider experience. Suggestions of earlier plasmapheresis or IVIG in the treatment course may prevent the progression to severe disease, but prospective studies are required.³

Directed cardiac therapy must be applied in combination with immunosuppression in patients experiencing an ICI-induced myocarditis. If decompensated heart failure predominates, consideration of diuretics and other means of circulatory support should be made. Our patient presented with ventricular tachycardia cardiac arrest and defibrillation was required, with subsequent reduction in ejection fraction noted which improved in 4 weeks. A transvenous pacer was required throughout her stay with eventual placement of a biventricular ICD. Tachyarrhythmias and bradyarrhythmias should be managed according to American College of Cardiology or American Heart Association guidelines.⁴

Immunomodulators in the treatment of ICI-induced myocarditis are a recent addition to the limited arsenal of treatment for severe glucocorticoid-refractory irAEs based on the demonstrated success in a few published cases. Salem et al reported a similar case of glucocorticoid refractory MG-myocarditis overlap syndrome following a third dose of nivolumab for metastatic lung cancer.⁵ Additional case reports and small case series have demonstrated the success of alternative immunomodulatory agents to treat steroid refractory fulminant ICI-induced myocarditis including tocilizumab,⁶ alemtuzumab,⁷ anti-thymocyte globulin,⁸ infliximab⁹ and mycophenolate.¹⁰ While the prevalence of ICI-induced myocarditis is low, mortality is estimated to be 50%, justifying need for further efforts in further characterising the role of novel immunomodulators in the treatment of irAEs.⁵

Aforementioned, treatment recommendations for ICI-induced myocarditis are limited and often are extrapolated from treatment regimens directed towards viral myocarditis. The improvement in myocarditis is often monitored by the resolving myocardial injury and clinical status of the patient. Generally, in mild or grade 1 irAEs, ICI can be continued; however, it is recommended to discontinue ICI permanently even in mild cases of myocarditis associated with ICI therapy. This is supported by a case report describing an individual who developed myocarditis while receiving nivolumab. Following resolution of myocarditis, reinitiation with a different ICI, pembrolizumab, was trialled. Within 2 weeks of switch in ICI therapy, the patient’s heart failure significantly worsened requiring hospitalisation and permanent discontinuation of ICI therapy.¹¹

The mechanism of immunotherapy-induced myocarditis is poorly understood, but it is suspected that PD-L1 and CTLA-4 pathways play a role in cardioprotection through T-cell-mediated interactions.⁵ Spencer et al demonstrated this with a preclinical mouse model with monoallelic loss of CTLA-4 and complete absence of PD-1, which mimicks an ICI-induced myocarditis. Subsequent treatment of the mice with a CTLA-4 agonist improved the progression of disease.¹² The prevention of serious irAEs may lie in the early identification of populations at risk. Recent case reports have documented HLA subtypes that may be at higher risk of irAE development when treated with pembrolizumab. Botta et al described a patient treated with pembrolizumab for urothelial cancer who developed myositis/myasthenia-like syndrome during treatment. On further characterisation, the patient presented an autoimmunity-associated haplotype (HLA-A*02/HLA-B*08/HLA-C*07/HLA-DRB1*03).¹¹ In addition, Nardone et al explored the utility of inflammatory biomarkers in the predictive value of poor outcomes in individuals with metastatic non-small cell lung cancer receiving PD-1/PD-L1 inhibitory monoclonal antibodies.¹³ This is further shown in Correale et al’s retrospective analysis showing a strong correlation of the rate of immune-related pneumonitis development in patients receiving PD-1/PD-L1 blocking monoclonal antibodies and germinal expression of HLA-B*35 and DRB1*11 alleles.¹⁴ While further investigation is required, a link between irAEs from pembrolizumab and systemic inflammation/autoimmunity has been drawn and may help in risk-stratification of patients prior to initiating ICI therapy.

ICIs have revolutionised oncological treatments in the past decade, allowing individuals diagnosed with malignancies that once carried terminal diagnoses to now experience long-term remissions. Fatal toxicities associated with ICIs are estimated to occur at a rate of 0.3%–1.3%. The onset of irAE symptoms can be swift following ICI initiation, with an average of 14.5–40 days.⁵ Myocarditis often presents with an overlap syndrome such as MG and with vague, non-specific symptoms, making early diagnosis difficult. Rapid clinical deterioration is common. Regimented screening parameters have not been established, but in light of the increasing utilisation of ICIs and their expanding significance in the field of oncology, the treating provider’s timely recognition and appropriate management are of paramount importance.

Learning points.

• Immune checkpoint inhibitors (ICIs) have an increasing role in many disease states and are becoming more widely used. As such, it is important to be aware of some of their life-threatening toxicities such as myocarditis.

• High clinical suspicion is needed to make the diagnosis. The staple therapy for ICI-induced myocarditis is steroids, but progression can be rapid and early detection is imperative. Additional therapy includes intravenous immunoglobulin, plasmapheresis and immunomodulators such as abatacept, tocilizumab and alemtuzumab.

• ICI toxicity can affect any tissue in the body. The degree of the toxicity usually dictates whether the ICI therapy should be continued or not. Due to the high mortality rate of ICI-induced myocarditis, it is generally recommended to discontinue the ICI permanently.

• Patients with ICI-induced myocarditis clinically deteriorate rapidly. CTLA-4 agonist use has been demonstrated in case reports to improve the condition which, additionally, has been demonstrated in a mouse model.

• Active areas of research involve risk stratification techniques to identify patients in the population at higher risk of developing immune-related adverse events to PD-1/PD-L1 blockade. These may include the monitoring of inflammatory markers (erythrocyte sedimentation rate, C reactive protein) or further HLA characterisation.

Ethics statements

Patient consent for publication

Consent obtained directly from patient(s)