This article is more than 5 years old. Information may no longer be current.
Immune checkpoint inhibitors may cause myocarditis
Patients with cancer who were treated with immune checkpoint inhibitors had an increased risk for myocarditis that started early in treatment, was detectable using an ECG and measurement of serum troponin and responded to higher steroid doses, according to a new study.
In addition, a study of WHO case safety reports found an increasing incidence of severe myocarditis among patients treated with immune checkpoint inhibitors.
Syed S. Mahmood, MD, MPH, a fellow in the cardiology division at New York-Presbyterian Hospital/Weill Cornell Medical Center, and colleagues analyzed data from patients treated with immune checkpoint inhibitors between November 2013 and July 2017. Patients with myocarditis (n = 35; mean age, 65 years; 29% women) were compared with a control group treated with immune checkpoint inhibitors but without myocarditis (n = 105; mean age, 65 years; 31% women).
The most common types of cancer in the cohort were melanoma (46%) and non-small cell lung cancer (11%).
The outcome of interest was major adverse cardiac events, defined as a composite of cardiac arrest, CV death, hemodynamically significant complete heart lock and cardiogenic shock. Patients were followed up for a median of 102 days.
During a median time of 34 days from the initiation of immune checkpoint inhibitors to onset, 1.14% of patients developed myocarditis. Fifty-four percent of patients experienced no other immune checkpoint inhibitor-related side effects.
Compared with the control group, patients with myocarditis were more likely to have diabetes (34% vs. 13%; P = .01) and were treated with combination immune checkpoint inhibitors (34% vs. 2%; P < .001).
During follow-up, major adverse CV events occurred in 46% of patients, and 38% of those patients had normal left ventricular ejection fraction. Patients with a final troponin T level of at least 1.5 ng/mL had a fourfold increased risk for major adverse CV events (HR = 4; 95% CI, 1.5-10.9).
Steroids were used to initially treat 89% of patients who developed myocarditis. Patients who were treated with higher steroid doses had lower rates of major adverse CV events and lower final or discharge serum troponin levels.
“We need to figure out if there’s a better treatment out there more targeted to the actual pathology, more targeted to what’s going on than steroids are,” Tomas Neilan, MD, MPH, director of the cardio-oncology program at Massachusetts General Hospital, told Cardiology Today.
Javid J. Moslehi, MD, assistant professor of medicine and director of the cardio-oncology program at Vanderbilt University Medical Center and a Cardiology Today Editorial Board Member, and colleagues used data from WHO’s VigiBase of individual safety case reports to identify 101 patients (median age, 69 years) with severe myocarditis following treatment with immune checkpoint inhibitors. Cancer types ranged widely, the most common of which included melanoma and lung cancer.
Of the 101 incidents of myocarditis, 57% of patients were treated with anti-programmed cell death protein-1 (PD-1) therapy and 27% were treated with combination therapy of anti-PD-1/PD-L1 and anti-cytotoxic T lymphocyte-associated antigen-4 (CTLA-4).
Forty-six percent of patients died.
Among patients taking combination anti-PD-1/PD-L1 and anti-CTLA-4, 67% died compared with 36% of patients taking anti-PD-1/PD-L1 alone (P = .008).
Three of five patients treated with ipilimumab who developed myocarditis died.
“We have been tracking these cases of severe myocarditis and deaths for the past year to gain a better understanding of the frequency of these events and the speed of myocarditis onset following the initial exposure to immune checkpoint inhibitors,” Moslehi said in a press release.
Fifty-nine patients had available dosing information. Among those, 64% received one or two doses before developing myocarditis.
Among the 33 patients with data on precise timing of myocarditis onset, the median onset was 27 days (range, 5-155). Most cases (76%) occurred during the first 6 weeks of treatment.
The researchers were unable to assess for comorbidities; however, few assessed patients reported receipt of concurrent CV or diabetes medications.
During the study period, the researchers observed a 76% increase in incidence reporting, possibly due to increased use of immune checkpoint inhibitor therapy.
“Myocarditis was observed across immune checkpoint inhibitor regimens, although it remains too early to determine whether the incidence differs between use of anti-PD-1 and anti-PD-L1 drugs,” the researchers wrote. “Furthermore, this condition occurs early on during therapy and across cancer types.” – by Darlene Dobkowski and Cassie Homer
References:
Mahmood SS, et al. J Am Coll Cardiol. 2018;doi:10.1016/j.jacc.2018.02.037.
Moslehi JJ, et al. Lancet. 2018;doi:10.1016/S0140-6736(18)30533-6.
For more information:
Tomas Neilan, MD, MPH, can be reached at Cardiac Unit Associates, 55 Fruit St., Boston, MA 02114; email: tneilan@partners.org.
Disclosures: Mahmood and Neilan report no relevant financial disclosures. Please see the study for all other authors’ relevant financial disclosures.
Perspective
Back to Top
Even though this study is limited in being a retrospective case control study, it raises several pertinent questions, particularly the increase in overall reported incidence of myocarditis at 1.14% vs. < 0.4% in earlier databases and pharma registries of immune checkpoint inhibitors since 2014. The question that emerges is: “Have we looked enough and have we looked long enough, or is this just the tip of the iceberg?” This question is not an ominous alarm bell to scare patients and practitioners from potentially lifesaving immunotherapy, but a lookout bell urging the need for early detection and understanding the immune basis of immune checkpoint inhibitor-induced myocarditis, both of which are the cornerstone of onco-cardiology.
Being cognizant of the emerging cardiotoxity of immune checkpoint inhibitors, a simple ECG and troponin I assay as a screening tool during first 3 months of therapy — wherein most of the cases seem to have been reported — appears to be the most valid suggestion of this study. Institution-wide algorithms incorporating this, akin to those for cardiotoxic breast cancer treatments, may result in improved understanding and outcomes of immune checkpoint inhibitor-induced myocarditis. The treatment of this entity, however, remains an inadequately answered question, with no established steroid dose having been studied and the additional wariness that using high-dose steroids in an already immunocompromised population confers.
The focus of future research starts at the pathology. We need more biopsies and a pathological database in addition to a clinical one, with routine assessment of lymphocytic markers and relative expression of checkpoint receptors such as PD-I, PDL–I and antigens like CTLA 4 on the cardiac myocytes. These will help us not only understand the immunopathology of the myocarditis, but portend a vulnerable population depending on their expression of checkpoint receptors. Mice studies have already demonstrated the important role for PD-1 in protecting the heart from T cell-mediated damage by demonstrating that. PD-1–deficient T lymphocytes caused enhanced disease with increased cytotoxic activity and inflammatory infiltrate (Tarrio ML, et al. J Immunol. 2012;doi:10.4049/jimmunol.1200389).
No. 2, since both CTLA-4 and PD-1 have an important role in maintaining immunologic homeostasis, there also remains a theoretical concern that therapeutic blockade of these receptors could lead to exacerbations of underlying autoimmune conditions or immune-related toxicity. Though the presence of autoimmune diseases was not addressed in this study, other retrospective studies have shown exacerbation of underlying autoimmune conditions in up to 38% of patients receiving PD-1 inhibitors (Johnson DB, et al. JAMA Oncol. 2016;doi:10.1001/jamaoncol.2015.4368). Therefore, the utility of routinely screening patients for subclinical autoimmune disease to predict cardiotoxicity remains an interesting topic of research.
No. 3, randomized controlled trials comparing the relative cardiotoxicity of these agents are needed to prove the suggestion of this study that some immune checkpoint inhibtors may be worse than the others, especially in combination. This could be similar to the anthracycline story, with their differential spectrum of cardiotoxicity.
And finally, there is paramount need for randomized controlled trials studying treatment with intermediate vs. high-dose steroids with or without other immunosuppressants to reduce major adverse cardiac events and affect early resolution of LV systolic function. These should also look at whether immunosuppression affects the oncological response to checkpoint inhibition in the targeted malignancy.
Perspective
Back to TopEver since the first reports of myocarditis associated with immune checkpoint inhibitors — including a very detailed description of two deaths that two members of this research group published in 2016 in The New England Journal of Medicine — clinicians have had a responsibility to discuss with their patients the rare but severe side effects that are possible following treatment with immune checkpoint inhibitors and emphasize the need to be vigilant about these effects.
Both patients and physicians have welcomed this new class of anticancer therapeutic agents, given the substantial and long-lasting beneficial effects that have been observed in the treatment of patients with a variety of tumor types. Physicians should be cautious, however, not to downplay this type of therapy as producing few side effects or as being easier to tolerate than chemotherapy. The side effects are just different and, in many cases, unpredictable.
Myocarditis appears to present early in the administration, and it is associated with myositis and generalized muscle weakness. T-cell infiltrates appear to be the culprit for myositis and myocarditis, which also can affect the heart electrical system, producing heart block and other dysrhythmias. Of note, mouse models of PD-1 deficiency predicted these types of side effects many years ago, as PD-L1 was shown to regulate T-cell-mediated injury in the heart.
In this study, Moslehi and colleagues discussed the observation of increased reporting of fatal immune checkpoint inhibitor-related myocarditis. Using WHO’s database of individual case safety reports, the authors identified 101 cases of severe myocarditis following immune checkpoint inhibitor treatment. More than half (57%) of these patients were receiving anti-PD-1 monotherapy.
It is extremely concerning that myocarditis developed after only one or two doses on average, and that the death rate for severe myocarditis was high (46%). The reporting rate appeared increased in 2017 compared with prior years, possibly due to the increasing use of this agent, but also possibly pointing toward increased recognition of the casual relationship for these events. Unfortunately, at this point, there is not a clear algorithm to recognize ahead of time the patients most predisposed to develop myocarditis or to identify it early enough to make a clinical difference.
It has been suggested that patients with previous myocardial infarction may have increased PD-1 expression, as reperfusion healing may increase lymphocytes infiltration. Obtaining a baseline ECG — with follow-up ECG for atypical symptoms — and possibly following with troponin for a couple weeks after the first administration may prove valuable in the early identification of this toxicity.
References:
Grabie N, et al. Circulation. 2007;116:2062-2071.
Johnson DB, et al. N Engl J Med. 2016;375:1749-1755.
Moslehi JJ, et al. Lancet. 2018;doi:10.1016/S0140-6736(18)30533-6.
Nishimura H, et al. Science. 2001;291:319-322.
Okazaki T, et al. Nat Med. 2003;9:1477-1483.
Tarrio ML, et al. J Immunol. 2012;doi:10.4049/jimmunol.1200389.