Article Text
Abstract
Immunotherapy revolutionised oncology by harnessing the native immune system to effectively treat a wide variety of malignancies even at advanced stages. Off-target immune activation leads to immune-related adverse events affecting multiple organ systems, including the cardiovascular system. In this review, we discuss the current literature describing the epidemiology, mechanisms and proposed management of cardiotoxicities related to immune checkpoint inhibitors (ICIs), chimeric antigen receptor (CAR) T-cell therapies and bispecific T-cell engagers. ICIs are monoclonal antibody antagonists that block a co-inhibitory pathway used by tumour cells to evade a T cell-mediated immune response. ICI-associated cardiotoxicities include myocarditis, pericarditis, atherosclerosis, arrhythmias and vasculitis. ICI-associated myocarditis is the most recognised and potentially fatal cardiotoxicity with mortality approaching 50%. Recently, ICI-associated dysregulation of the atherosclerotic plaque immune response with prolonged use has been linked to early progression of atherosclerosis and myocardial infarction. Treatment strategies include immunosuppression with corticosteroids and supportive care. In CAR T-cell therapy, autologous T cells are genetically engineered to express receptors targeted to cancer cells. While stimulating an effective tumour response, they also elicit a profound immune reaction called cytokine release syndrome (CRS). High-grade CRS causes significant systemic abnormalities, including cardiovascular effects such as arrhythmias, haemodynamic compromise and cardiomyopathy. Treatment with interleukin-6 inhibitors and corticosteroids is associated with improved outcomes. The evidence shows that, although uncommon, immunotherapy-related cardiovascular toxicities confer significant risk of morbidity and mortality and benefit from rapid immunosuppressive treatment. As new immunotherapies are developed and adopted, it will be imperative to closely monitor for cardiotoxicity.
- cardiomyopathies
- myocarditis
- drug monitoring
- diagnostic imaging
Statistics from Altmetric.com
Introduction
Immunotherapies, including cytokines, targeted therapies, cell-based therapies and vaccines, harness the immune system to combat tumour cells. Increased understanding of antitumour mechanisms has led to a proliferation of new treatments; in 2017, approximately 940 clinical and 1064 preclinical immunotherapies were under investigation (figure 1).1 Such treatments target common antitumour mechanisms to allow generalisability to multiple malignancies, maximise efficacy and minimise side effects. However, immune-related adverse events (IRAEs) can affect every major organ system, including the cardiovascular system.2 With 1.8 million new cancer cases per year and increased indications for immunotherapy, uncommon cardiotoxicities are on the rise.3 The short-term and long-term cardiovascular effects of these novel treatments require vigilant monitoring. This review describes the current understanding of the mechanisms, epidemiology and management of immunotherapy-related cardiotoxicity, focusing on immune checkpoint inhibitors (ICIs) and chimeric antigen receptor (CAR) T-cell therapies.
Immune checkpoint inhibitors
Mechanism
A delicate balance of co-stimulatory and co-inhibitory pathways regulate the adaptive immune system. Antigen-presenting cells express membrane-bound major histocompatibility complex (MHC) together with the second immunological signal required to activate T cells, the co-stimulatory molecules B7-1 (CD80) and B7.2 (CD86). T cells recognise the antigen-MHC complex with the T cell receptor (TCR), and the co-stimulatory molecules with the activating receptor CD28. To prevent excessive immune activation, co-stimulation is antagonised by the dominant expression of the cytotoxic T lymphocyte antigen 4 (CTLA-4), which effectively competes with CD28 and limits the activation step of an immune response (figure 2). Antitumour T cells that pass this first immune checkpoint can be inactivated at the tumour site through the reactive expression of the programmed death ligand 1 (PD-L1). This allows cancer cells to turn off antitumour T cells by triggering the programmed death 1 (PD-1) receptor that leads to T cell exhaustion. This represents a second immune checkpoint at the effector step of an antitumour immune response. ICIs are blocking antibodies to either PD-1 or PD-L1, and to CTLA-4, which block these two immune checkpoints and reactivate the antitumour immune response.4–6 There has been a rapid increase in monotherapy and dual-therapy Food Drug and Administration (FDA)-approved indications for the seven current FDA-approved ICIs (table 1, figure 1).
The 1-year absolute risk of cardiac events, including non-immune-related events, ranged from 6.6% to 9.7% for patients with melanoma and lung cancer receiving ICIs (table 2).7 Most patients undergoing ICI therapy experience at least one IRAE.8 Cardiovascular IRAEs occur with an incidence of 1.14%–5%,4 7 9 10 but have the highest mortality rate among IRAEs up to 50% (table 2).9–13 Immune checkpoints suppress T cell-mediated autoimmune activation in myocardium. Possible cardiovascular IRAE aetiologies include activation of local T cells, cross-reaction of antitumour T cells with myocardial antigens or systemic immune activation.6 14 15 Cardiovascular IRAEs include myocarditis (0.3%–1.4% of patients on ICI therapy), arrhythmias (3.6%–4.8%), pericardial disease (1.74%), vasculitis (0.27%), acute coronary syndrome (ACS) (0.95%–7%) and heart failure (HF, 1.6%) (figure 2, table 2).7 12 13 16
Risk factors
Cardiovascular IRAE risk factors are poorly understood. Combination therapy is associated with increased incidence, severity and mortality of ICI-associated myocarditis compared with monotherapy.14 Concomitant non-cardiac IRAEs confer increased risk of ICI-associated myocarditis, particularly myositis (reported OR (ROR) 25, 95% CI 18.7 to 34.9), encephalitis (ROR 2.9, 95% CI 1.4 to 6.3) and hepatitis (ROR 2.9, 95% CI 1.9 to 4.5).17 In a single-centre study, myocarditis was associated with pre-existing cardiovascular disease, including HF and ACS, and increased age (>80 years).13 Investigations are needed to identify high-risk patients, including concurrent cardiotoxic chemotherapies or targeted therapies, tumour types and underlying cardiovascular disease.
ICI-associated myocarditis
ICI-associated myocarditis presentations range from asymptomatic biomarker elevation to cardiogenic shock. Symptomatic ICI-associated myocarditis commonly presents with shortness of breath, chest pain or palpitations.12 Over 40% of patients have concurrent severe IRAEs, particularly myositis (25%) and myasthenia gravis (10%).10 12 Onset is common after two to three infusions with up to 80% occurring within 3 months of initiation,7 9 10 12 13 but range from days to years (table 2). Given the non-specific presentation and timing, a high index of suspicion for ICI-associated myocarditis is required.
Monitoring and diagnosis
Baseline ECG and biomarkers, including troponin and brain natriuretic peptide (BNP), followed by ongoing monitoring for new cardiovascular symptoms are recommended with ICI therapy (figure 3).2 While evidence is limited for risk stratification, routine biomarker monitoring may be pursued in select individuals. ICI therapy is not contraindicated with underlying cardiovascular disease, but patients should be medically optimised prior to therapy. Cardiac biomarkers are sensitive; troponin and BNP are elevated in 94% and 66% of cases of ICI-associated myocarditis, respectively. Additionally, higher admission and peak troponin level were associated with increased major adverse cardiac events (MACE).9 Although ECG is abnormal in 89% of cases, diagnostic utility is limited due to increased ICI-associated arrhythmias distinct from myocarditis.9
Echocardiography is the first-line imaging modality for suspected ICI-associated myocarditis, but should be interpreted with caution as left ventricular ejection fraction (LVEF) is normal in up to 50% of cases.9 Beyond LVEF, reduction in global longitudinal strain is associated with the presence (14.1%±2.8% myocarditis vs 20.5%±1.9% control, p<0.001) and severity of myocarditis (1.5-fold and 4.4-fold increased risk of MACE in reduced and preserved LVEF, respectively).18
Cardiac magnetic resonance imaging (CMR) further characterises myocardium using late gadolinium enhancement (LGE) and T2-weighted short-T1 inversion recovery imaging of oedema, but has limited sensitivity in ICI-associated myocarditis. In an international registry of ICI-associated myocarditis (n=103), only 48% of all patients had LGE compared with 90% in other causes of acute myocarditis.19 LGE was distributed in subendocardial, transmural, subepicardial, mid-myocardial and diffuse patterns, and independent of LVEF, with 43% of LGE in preserved LVEF. LGE was not associated with symptoms, outcomes, ECG or echocardiographic findings. Importantly, timing alters CMR sensitivity in ICI-associated myocarditis. CMR administered >4 days after admission showed greater rates of LGE compared with earlier (72.0% vs 21.6%, OR 9.35, p<0.001).20
Endomyocardial biopsy is the diagnostic gold standard in ICI-associated myocarditis, but the invasive nature and procedural risks limit widespread use. Accordingly, diagnosis is based on clinical and biomarker assessment, aided with echocardiography and CMR.
Treatment
ICI-associated myocarditis is a potentially fatal complication of ICI therapy with mortality approaching 50% (table 2).9–11 13 Importantly, among survivors with reduced LVEF approximately half completely recover LVEF.10 Limited by the lack of randomised controlled trials, current treatment strategies are based on expert consensus. The 2018 American Society for Clinical Oncology (ASCO) guidelines define severity as grades 1–4 (figure 3).2
ICI-associated myocarditis management includes stopping ICIs, immunosuppression and supportive therapy. The risk of restarting ICIs after myocarditis recovery is unknown and permanent discontinuation is recommended for severity greater than grade 1.2 21 ICIs have a long half-life (ipilimumab 14.5 days, pembrolizumab 25.0 days, nivolumab 26.7 days and atezolizumab 27.0 days); in severe cases, plasmapheresis can clear antibodies from the bloodstream.15
Corticosteroids are the backbone of immunosuppression in ICI-associated myocarditis. ASCO guidelines recommend oral or intravenous high-dose corticosteroids (prednisone 1–2 mg/kg) for 4–6 weeks depending on severity of symptoms and clinical response (figure 3, table 3).2 An international multicentre registry of 126 ICI-associated myocarditis cases found that early, high-dose corticosteroids were associated with improved outcomes. Early corticosteroid administration (<24 hours since admission) was associated with lower rates of MACE compared with 24–72 or >72 hours (7.0%, 34.3% and 85.1%, respectively, p<0.001). High-dose corticosteroids (greater than methylprednisolone 500 mg/day) was associated with a reduced risk of MACE independent of timing (HR 0.27, 95% CI 0.09 to 0.84, p=0.024).22 Randomised controlled trials are needed to define the optimal timing and dose of corticosteroid therapy in ICI-associated myocarditis.
Adjuvant immunosuppressive agents are used in fulminant, grade 4 ICI-associated myocarditis. ICI-associated myocarditis and cardiac allograft cellular rejection have similar histological appearances and may have similar mechanisms. Therefore, antirejection drugs and protocols have been adapted to treat ICI-associated myocarditis (figure 3), including mycophenolate, antithymocyte globulin, abatacept and infliximab (table 3).2 6 9 However, infliximab should be avoided due to reports of increased cardiovascular mortality, particularly with moderate-to-severe HF.23
ICI-associated myocarditis can progress to cardiogenic shock, life-threatening arrhythmias and death. Inpatient monitoring of grade 3 or 4 ICI-associated myocarditis patients is needed to monitor for decompensation and arrhythmias requiring temporary or permanent pacemaker placement. Advanced mechanical circulatory support, including intra-aortic balloon pump, Impella (Abiomed, Danvers, Massachusetts, USA) or extracorporeal membrane oxygenation, may be considered in cardiogenic shock. Decisions for invasive management of arrhythmias or haemodynamic complications are complex, incorporating likelihood of cardiac recovery, comorbidities, underlying malignancy and associated prognosis, and require a multidisciplinary approach.
ICI-associated atherosclerosis
Recent investigations have focused on ICI-associated immune dysregulation and atherosclerotic cardiovascular disease (ASCVD). The proposed connection was first noted as a 1% incidence of ACS in a meta-analysis of 22 ICI trials in lung cancer.24 Recently, a single-centre study of 3326 patients on ICIs showed a 7% incidence of each ACS and stroke.13 In fact, in a matched cohort study, ICI therapy conferred a threefold increased risk of ASCVD (HR 3.3, 95% CI 2.0 to 5.5, p<0.001).16
Initially, the increased risk of ACS was attributed to other ICI-associated complications, such as vasculitis or coronary vasospasm, rather than atherosclerotic plaque rupture. However, recent advances demonstrate an important role for activated T cells in vulnerable plaques that may be modulated by ICI therapy. Large vessel atheroma in humans and mice with melanoma undergoing ICI treatment demonstrated no change in 18FDG-PET-positive vascular inflammation. However, the intraplaque immune response shifted from macrophage to T cell-predominant with increased necrotic core and endothelial activation indicating plaque vulnerability to rupture.25
Beyond the acute risk of ACS, ICIs lead to atherosclerosis progression and ASCVD risk.16 PD-1 expression modulates the arterial wall immune response; disruption via PD-1 inhibition leads to atherosclerosis progression. Single-cell analysis of human atheroma with recent ACS showed a distinct pattern of T cell activation. In particular, increased PD-1 expression may accelerate atherosclerosis and ACS in patients with pre-existing ASCVD.26 Importantly, concurrent statin or corticosteroid use was associated with reduced aortic atherosclerotic plaque progression.16 Therefore, cardiac care for the oncology patient must become proactive with aggressive risk factor modification at initiation of therapy and close cardiovascular monitoring.27
Other ICI-associated cardiotoxicities
ICI-associated pericardial disease, including pericarditis, myopericarditis and pericardial effusion, have been most commonly reported in lung cancer.11 In a single-centre retrospective study of patients with non-small cell lung cancer, pericardial effusions were increased with ICIs compared with chemotherapy (3.3% vs 1.6%, respectively).28 Clinical features include new onset chest pain, pericarditis-related ECG changes and pericardial effusion on imaging. Notably, mortality in ICI-associated pericarditis is 21%—significantly higher than non-ICI-associated pericarditis.11 Therapeutic options, including non-steroidal anti-inflammatories, steroids or colchicine, have not been studied in this population and merit consideration.
ICI-associated arrhythmias are the most common cardiotoxicity and portend significant mortality, particularly sudden cardiac death.7 11 12 ICI-associated arrhythmias include conduction delays, ventricular and supraventricular arrhythmias.12 29 Only 17% of supraventricular arrhythmias are isolated, with the overwhelmingly majority associated with concurrent IRAEs, including colitis (26.1%), thyroiditis (13.1%), pneumonitis (13.1%) and myocarditis (12.2%).11 In contrast, local ventricular or conduction system inflammation likely causes conduction delays and ventricular arrhythmias. In patients with palpitations or syncope, evaluation includes ECG, ambulatory ECG monitoring and evaluation for myocarditis or other IRAEs.
ICI-associated vasculitis, including polymyalgia rheumatica and temporal arteritis, presents relatively delayed with median onset 55 days (table 2). Although mortality is comparatively low, impaired vision occurred in 27.8% of cases.11 The ASCO guidelines recommend corticosteroid treatment to prevent vision loss.4
Chimeric antigen receptor T-cell therapy
CAR T-cell therapy is a cell-based immunotherapy whereby patient’s cells are genetically engineered to target tumour-specific proteins to induce tumour-cell apoptosis (figure 2).30 Currently, three CD19-directed CAR T-cell therapies (axicabtagene ciloleucel, brexucabtagene autoleucel, tisagenlecleucel) are FDA-approved for diffuse large B-cell lymphoma (DLBCL), acute lymphoblastic leukaemia (ALL) and mantle cell lymphoma. CAR T-cell therapy in solid malignancies is under investigation.31 Despite significant clinical efficacy,32 life-threatening toxicities, including haemodynamic instability and cardiotoxicity, complicate CAR T-cell therapy.
Cardiotoxicity pathophysiology
On engagement with target antigen domains, CAR T-cells elicit a strong immune response critical to their therapeutic effect. Activation of CAR T-cells leads to massive systemic release of cytokines, including IL-6, TNF-α and IFN-γ, causing prostaglandin activation and triggering cytokine release syndrome (CRS).33 CRS manifests as fever, tachycardia, hypotension, hypoxia and multiorgan dysfunction34 35 and can be associated with cardiotoxicity.36
IL-6 is a primary driver of inflammation in CRS, leading to increased B and T cell activity and release of acute phase reactive proteins.37 A marked rise in IL-6 accompanies CAR T-cell-associated cardiotoxicity in CRS.15 Elevated IL-6, previously studied in diabetes, is associated with cardiovascular complications including myocardial ischaemia, atherosclerosis, HF and hypertension.38 IL-6 may similarly lead to CAR T-cell-associated cardiotoxicity.39
Cardiotoxicity presentation
Cardiovascular toxicity following CAR T-cell therapy is often accompanied by CRS, neurotoxicity and graft versus host disease.40 Among CAR T-cell-associated cardiovascular events reported to the FDA Adverse Events Reporting System from 2017 to 2019 (n=196), CRS was present in two-thirds of patients.41 The most common study-defined MACE were arrhythmias (77.6%), HF (14.3%) and myocardial infarction (0.5%). The most frequently reported system-specific adverse event, immune effector cell-associated neurotoxicity syndrome (ICANS)—previously named CAR T-cell-related encephalopathy syndrome—was associated with a combined measure of CRS and cardiovascular events.41
Retrospective studies of CAR T-cell therapy have characterised cardiotoxicity (table 4). Cardiovascular events include decompensated HF, arrhythmia, ACS, stroke, prolonged QTc42 and cardiovascular death. In an adult study of CAR T-cell therapy (n=137), all MACE (n=17) occurred with grade ≥2 CRS and 95% occurred with elevated troponin.39 In another retrospective study (n=145), high-grade CRS was strongly associated with increased risk of MACE. MACE (n=31) occurred at a median of 11 days with a wide range (IQR=6–151), underscoring a need for longitudinal cardiovascular monitoring.43 In a retrospective cohort of 116 patients treated with CD19-directed CAR T-cells undergoing serial echocardiograms, 10% of patients developed new or worsening cardiomyopathy with a median LVEF decline from 58% to 37% and a median of 12.5 days following therapy.44
Paediatric CAR T-cell therapy populations are also susceptible to cardiotoxicity. Among paediatric and young adult patients (n=52) with haematological malignancies treated with CD19- 28ζ CAR T-cells, patients developed CRS (n=37) left ventricular dysfunction (n=6) and hypotension requiring vasoactive agents (n=9). Notably, 31% of patients with troponin measurements and CRS had elevated troponin and reduced LVEF.45
Monitoring and treatment
While CAR T-cell therapy has no absolute contraindications, cardiovascular toxicity risk is not uniform. Increased age, hyperlipidaemia, coronary artery disease and outpatient medications including beta-blockers, ACE inhibitors and angiotensin receptor blockers were associated with cardiomyopathy, but not cancer type, prior anthracycline use or mediastinal radiation.44 In paediatric patients with ALL, hypotension requiring vasopressors was more likely in those with higher tumour burden, lower baseline LVEF or diastolic dysfunction.46 Because of the common association of CRS, co-occurring inflammatory toxicities, such as ICANS, may increase cardiotoxicity risk.41
Monitoring, diagnosis and treatment of cardiotoxicity during CAR T-cell therapy remains challenging (figure 4). In a case report, a patient with cardiomyopathy and refractory DLBCL with planned CAR T-cell therapy underwent CardioMEMS (Abbott Medical, Abbott Park, Illinois, USA) implantation prior to lymphodepletion. CardioMEMS, conventionally used for outpatient HF management, was repurposed for invasive haemodynamic monitoring throughout hospitalisation.47 This novel application demonstrated a proof of concept for haemodynamic guidance in high-risk patients during CRS.
No randomised controlled trials have evaluated interventions to reduce cardiovascular morbidity in CAR T-cell therapy, but given their strong association, early intervention of CRS may prevent cardiotoxicity. Tocilizumab, an FDA-approved IL-6 receptor antagonist, prevents symptomatic CRS progression.30 48 Each 12-hour delay from CRS onset to tocilizumab administration was associated with a 1.7-fold increased risk of MACE.39 Both in vitro and clinical data suggest that tocilizumab does not impair CAR T-cell efficacy.49 50 Siltuximab, a monoclonal IL-6 targeted antibody, has similarly been used off-label to prevent CRS progression in CD19-directed CAR T-cell therapy.w1 Corticosteroids are used in severe, refractory CRS. Corticosteroids abrogate the inflammatory cascade, interfering with T-cell function and inducing T-cell apoptosis,w2 but do not alter clinical response to therapy.w3 Further research is needed to improve cardiovascular outcomes and monitoring protocols during and following CAR T-cell therapy, particularly in high-risk patients.
Bispecific T cell engager therapy
Bispecific T cell engager (BiTE) molecules link two single-chain variable fragments targeting T cell and tumour cell antigens to bring them into close physical proximity (figure 2).w4 Binding and colocation cause T cell activation and cytokine release. Blinatumomab, targeting CD3 and CD19, was the first FDA-approved BiTE agent in 2017 for B cell precursor ALL.w5 Although cardiovascular effects of BiTE therapy have not been specifically studied, blinatumomab causes CRS with likely similar cardiovascular effects as in CAR T-cell therapy. In the TOWER trial investigating blinatumomab, CRS and ACS occurred in 2.6% and 2.2% of patients, respectively.w6 However, a real-world study of blinatumomab showed grade 3–4 CRS in 19% of patients.w7 Ongoing surveillance for adverse cardiovascular effects is warranted for this evolving immunotherapy.
Conclusion
Immunotherapy has emerged as an important and diverse class of treatments for advanced malignancies; with improved survival rates, both use and duration of treatment has increased significantly over the past decade. For FDA-approved indications alone, the percentage of patients with cancer eligible for ICIs increased from 1.5% in 2011 to 42.6% in 2018.3 Although uncommon, myocarditis is a significant and potentially severe complication of immunotherapy. Like recent findings linking these agents to accelerated atherosclerosis, such therapies have the potential for more insidious, long-term effects that are only starting to be recognised. With CAR T-cell and BiTE therapy, the overlapping and detrimental effects of CRS on the cardiovascular system require further mechanistic study to develop effective diagnostic and treatment strategies. International multicentre collaborations are needed to identify high-risk individuals, develop monitoring guidelines and optimise treatment strategies to improve the care of patients with immunotherapy-associated cardiotoxicities for current and future therapies.
Supplemental material
Ethics statements
Patient consent for publication
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Twitter @A_SteinMerlob, @mvrothberg, @datsunian
Contributors All authors significantly contributed to manuscript design, planning, writing, visual design and editing and revision.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests EY reports receiving research funding from CSL Behring (non-relevant). AFS-M is supported by the National Institutes of Health Cardiovascular Scientist Training Program (T32 HL007895). MVR has no conflicts of interest to report. AR has received honoraria from consulting with Amgen, Bristol-Myers Squibb, Chugai, Genentech, Merck, Novartis, Roche, Sanofi and Vedanta, is or has been a member of the scientific advisory board and holds stock in Advaxis, Apricity, Arcus, Compugen, CytomX, Five Prime, Highlight, ImaginAb, Isoplexis, Kalthera, Kite-Gilead, Merus, PACT Pharma, RAPT, Rgenix and Tango, has received research funding from Agilent and from Bristol-Myers Squibb through Stand Up to Cancer (SU2C).
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Commissioned; externally peer reviewed.
Author note Additional references w1–8 can be found in online supplemental file 1.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.