Cardivascular Toxicities of Immunotherapy

Cardiac Manifestations With Anti–CTLA-4 Treatment

Cases of fatal myocarditis, myocardial fibrosis, reversible left ventricular dysfunction, late onset pericardial effusion, cardiac tamponade, constrictive pericarditis, and Takotsubo cardiomyopathy with apical ballooning on echocardiography have all been observed with anti–CTLA-4 therapy.

Cardiac Manifestations With Anti–PD-1 Treatment

Clinically significant cardiotoxic events associated with anti–PD-1 therapy include pericarditis, hypertension, atrial and ventricular arrhythmia, and myocardial infarction. A case series of melanoma patients treated with anti–PD-1 therapy showed a 1% incidence of cardiac disorders including a case of fatal ventricular arrhythmia due to myocarditis, various other arrhythmias (atrial flutter, ventricular arrhythmia), asystole due to cardiomyopathy, hypertension, myocarditis, and left ventricular dysfunction. The onset of these toxicities ranged from 2 to 17 weeks after treatment. There does not appear to be any correlation between the type of anti–PD-1 therapy, tumor response, tumor type, or any particular clinical features that predispose these patients to adverse cardiac events.

Cardiac Manifestations With Combination Immune Checkpoint Inhibitors

• ■

  Toxicities associated with ICIs are enhanced with combination therapy compared with monotherapy alone. Combination ICI (ipilimumab-nivolumab) resulted in grades 3 and 4 adverse events in 55% of patients compared with 16% of patients only on nivolumab and 27% of patients only on ipilimumab.

• ■

  Combination ipilimumab and nivolumab in two melanoma patients led to fulminant myositis with rhabdomyolysis, early progressive and refractory cardiac electrical instability, and myocarditis. Despite aggressive interventions with high-dose glucocorticoids and in one case, infliximab, both of the patients died. Another case of myocarditis which was salvaged presented with symptoms of heart failure and left ventricular (LV) dysfunction with reduction in left ventricular ejection fraction (LVEF) from 50% to 15% after 3 combination infusions of ipilimumab and nivolumab. However, the LVEF improved to 40% after 2 months of high-dose glucocorticoids and treatment for heart failure.

• ■

  Johnson et al. reported the frequency of cardiovascular complications, namely, myocarditis and myositis, in a large population extracted from the of Bristol-Myers Squibb corporate safety database. Among a total of 20,594 patients studied, 0.09% drug-related severe adverse events of myocarditis were reported. Combination therapy was associated with a higher incidence of frequent and severe myocarditis than with nivolumab alone (0.27% or 5 fatal events vs. 0.06% or 1 fatal event, P < .001). Mortality was high, with death occurring secondary to refractory arrhythmias or cardiogenic shock. Median time to development of myocarditis was 17 days (range, 13–64 days). Severe myositis (grade 3 to 4) also appeared more frequently when the combination of drugs was used than when nivolumab was the only agent used (0.24% vs. 0.15%). In clinical trials involving nivolumab, ipilimumab, or both, there was no routine testing for myocarditis by means of either biochemical analysis or cardiac imaging. It is important to recognize, however, that the actual incidence of cardiac events post ICI may be more substantial than what is already known because cardiac monitoring was not a routine part of clinical trials. Also, it is important to remember that the data were collected retrospectively from a single manufacturer in the absence of prospective standardized screening of cardiac issues, and hence, it is very likely that this underrepresents the true incidence.

• ■

  A paper published in Circulation analyzed a total of 30 patients with ICI-related cardiotoxicity which included 12 newly diagnosed patients and 24 patients with previous data that had been reported in case series. Cardiotoxicity was diagnosed at a median of 65 days (range, 2–454 days) after the initiation of ICIs, and occurred after a median of 3 infusions (range, 1–33). It was observed in the study that cardiotoxicity was higher after the first and third infusions. The most frequent clinical manifestations observed in the patients were dyspnea, palpitations, and signs of congestive heart failure. The development of LV dysfunction was observed in 79% of patients, and 14% patients developed a Takotsubo-syndrome–like appearance. Atrial fibrillation, ventricular arrhythmia, and conduction disorders were observed in 30%, 27%, and 17% of patients, respectively, who were treated with ICIs; however, after excluding the finding of LV dysfunction, they were observed to occur in 3%, 7%, and 13% of patients, respectively. Myositis was noted to develop in 23% of patients. It was also observed that cardiovascular mortality was significantly associated with conduction abnormalities (80% vs. 16%, P = .003) and ipilimumab-nivolumab combination therapy (57% vs. 17%, P = .04).

• ■

  A review of complied case reports and case series by Jain et al. revealed that the onset of cardiovascular irAEs can be seen as early as 2 weeks and as late as 32 weeks after initiation of ICI, with a median onset at 10 weeks after initiation.

Mechanism of Immune Checkpoint Inhibitor–Mediated Cardiotoxicity

A plausible mechanism behind ICI-mediated myocarditis is that shared targeted antigens (epitopes)/high frequency T-cell receptors may exist among tumor cells and the cardiac myocytes that could become a target for activated T cells and thus lead to myocardial lymphocytic infiltration resulting in heart failure and several other conduction abnormalities. Among patients who died from myocarditis, autopsy showed abundant CD4+ and CD8+ T-cell infiltration of the tumor, cardiac muscle (cardiac sinus and the AV node), and skeletal muscle. These were indicative of lymphocytic myocarditis and myositis. Pathology review also showed myocardial fibrosis, and cardiomyopathy predisposing to heart failure, and conduction abnormalities, including heart block and cardiac arrest. Pericarditis and pericardial effusion have also been described. Although rare, there has also been a case report of irAE-associated acute coronary syndrome. PD-L1 expression has been noted on the membranous surface of the injured myocytes and on the infiltrating CD8+ T cells and histiocytes from the inflamed myocardium. Along with this, the over-expression of IFN-γ, granzyme B, and tumor necrosis factor-α (TNF-α), produced by the activated T cells, could also contribute to cardiac damage. On the other hand, the skeletal muscles and the tumor had negative/lower expression for PD-L1. PD-L1 upregulation in the myocardium could be a cytokine-induced cardioprotective mechanism that is abrogated by immune-checkpoint blockade. It remains undetermined as to what are the causative epitopes that are recognized by these T-cell receptors within the multitude of antigens. Mice studies have shown that genetic deletion of PD-1 in mice models results in cardiomyopathy caused by antibodies against cardiac troponin I; however, no such mechanism has been identified in humans. ^, Hence, Nishimura and colleagues concluded that PD-1 may be an important receptor contributing to autoimmune cardiac diseases. Several mouse models of T-cell–dependent myocarditis exist where genetic deletion of PD-L1/L2, as well as treatment with anti–PD-L1, transformed transient myocarditis into a lethal disease. Preclinical studies have also shown that CTLA-4−/− mice develop severe autoimmune myocarditis mediated by CD8+ T cells, which is rapidly fatal at birth.