Abstract
Background
Immune checkpoint inhibitors (ICIs) have revolutionized cancer treatment but carry a risk of immune‐related adverse events, reported in 15% to 90% of the patients. ICI‐associated pericardial and myocardial disease has been increasingly recognized, yet data on outcomes remain limited. We aimed to use a large real‐world database to investigate the short‐ and long‐term outcomes of ICI‐associated pericarditis, myocarditis, and myopericarditis versus the non‐ICI cases.
Methods
We conducted a retrospective analysis using the TriNetX database to identify patients >18 years old on ICI and diagnosed with pericarditis, myocarditis, or myopericarditis within 1 year of initiation of therapy. The primary outcome assessed was all‐cause mortality, tamponade, constrictive pericarditis, and need for pericardiocentesis at 90‐day, 1‐year, and 5‐year follow‐up.
Results
A total of 3661, 226, and 46 patients were identified in matched cohorts of ICI‐induced pericarditis, myocarditis, and myopericarditis, respectively. One‐ and 5‐year mortality was greater for pericarditis (44.3% versus 37.3%; 54.3% versus 46.9%), myocarditis (38.9% versus 20.8%; 47.8% versus 34.3%), and myopericarditis (56.5% versus 28.3%; 63.0% versus 28.3%; all P≤0.002). A multivariable Cox model showed increased ICI‐related mortality for pericarditis 1‐year hazard ratio [HR], 1.20 (95% CI, 1.12–1.29), 5‐year HR, 1.26 (95% CI, 1.18–1.35), myocarditis 90‐day HR, 1.82 (95% CI, 1.21–2.73), 1‐year HR, 2.04 (95% CI, 1.43–2.91), 5‐year HR, 1.64 (95% CI, 1.26–2.61), and myopericarditis 1‐year HR, 2.24 (95% CI, 1.15–4.37), and 5‐year HR, 2.78 (95% CI, 1.44–5.36), all with P<0.001.
Conclusion
Among 3661, 226, and 46 patients with ICI‐associated pericarditis, myocarditis, and myopericarditis, respectively, we demonstrated an increased risk of all‐cause mortality as compared with the respective non‐ICI population.
Keywords: immune checkpoint inhibitors, myocarditis, myopericarditis, pericarditis
Subject Categories: Pericardial Disease, Cardiotoxicity
Nonstandard Abbreviations and Acronyms
- ICI
immune checkpoint inhibitors
- irAE
immune‐related adverse event
Clinical Perspective.
What Is New?
To the best of our knowledge, this is the largest real‐world cohort study to date evaluating short‐ and long‐term outcomes of pericarditis, myocarditis, and myopericarditis in patients receiving immune checkpoint inhibitors (ICIs).
ICI‐associated pericardial and myocardial inflammatory diseases are associated with significantly increased 1‐year and 5‐year all‐cause mortality compared with patients not treated with ICI, despite comparable baseline characteristics after propensity matching.
Among the ICI classes, combination PD‐1/CTLA‐4 (programmed cell death protein 1/cytotoxic T‐lymphocyte‐associated protein 4) therapy was associated with the highest 5‐year mortality, with gastrointestinal and respiratory cancers showing especially poor prognoses.
What Are the Clinical Implications?
Clinicians should maintain a high index of suspicion for pericardial and myocardial toxicities in cancer patients treated with ICIs, as early recognition and treatment may be critical in reducing the high associated mortality.
Standard therapies (NSAIDs, colchicine, steroids) appear insufficient to mitigate the elevated risk, highlighting the need for tailored monitoring protocols and possibly earlier cardiology involvement.
Risk stratification tools are needed to identify high‐risk patients before ICI initiation to guide surveillance and preemptive management strategies.
Immune checkpoint inhibitors (ICIs) have revolutionized cancer treatment by achieving unprecedented rates of durable antitumor response. 1 ICIs are monoclonal antibodies directed against immune checkpoint proteins, designed to “turn back on” the cellular immune response against tumor cells. Increasingly, ICIs are being used to increase antitumor activity. 2 Systemic augmentation of immune responses by ICIs has led to the management of immune‐related adverse events (irAEs) as an important component of oncological care. ICI‐related major adverse cardiac events, although rare (<1%), involve the cardiovascular system on multiple levels. 3 Myocarditis is the most common and severe ICI‐major adverse cardiac event, but other significant events include pericardial diseases, Takotsubo cardiomyopathy, arrhythmias, and vasculitis. 4 In recent years, the literature has underscored the importance of vigilant monitoring for cardiovascular irAEs in patients undergoing ICI therapy, given their potential severity and impact on patient outcomes. 4 , 5 , 6
Pericarditis is the inflammation of the double‐walled sac (pericardium) containing the heart and the roots of the great vessels. The pathogenesis is diverse, with idiopathic and viral causes being the most common in developed countries. Other causes include postpericardiectomy syndrome, autoimmune diseases, and cancer. ICI‐related pericardial disease can present as pericarditis, pericardial effusion, or cardiac tamponade. 7 , 8 Myocarditis is defined as an inflammatory disease of the myocardium, confirmed by histopathology or, when not feasible, by a combination of clinical, laboratory, and imaging criteria. 9 , 10 , 11 Additionally, myopericarditis is defined as a clinical syndrome characterized by concurrent inflammation of both the myocardium and pericardium, with predominant pericarditic features (such as chest pain, pericardial rub, or pericardial effusion) and evidence of mild myocardial involvement (such as elevated troponin or imaging findings), but without significant impairment of ventricular function. 12 Pericarditis, myocarditis, or myopericarditis due to ICI, although rare, requires special attention due to its severity and need for immunosuppressive therapy. 13 Pericarditis and pericardial effusions can result in cardiac tamponade, a life‐threatening complication. Currently, there are limited data regarding the outcomes of patients who develop pericarditis, myocarditis, or myopericarditis while receiving ICIs. This study aims to evaluate short‐ and long‐term outcomes in patients treated with ICIs who subsequently develop pericarditis, myocarditis, or myopericarditis.
METHODS
Study Population
We conducted a retrospective analysis using the TriNetX database, which is an international database platform that collects up‐to‐date information from electronic health record systems from a network of up to 97 health care organizations at the time of this study. This involves pooled data from millions of deidentified patients, allowing for real‐world analyses of large patient cohorts. Information provided on the TriNetx platform includes patient demographics, diagnoses by the International Classification of Diseases, Tenth Revision (ICD‐10) codes, Current Procedural Terminology codes, medications, laboratory values, microbiology data, and echocardiography results. This study was exempt from institutional review board approval given that all data are deidentified. The data related to our study are available from the corresponding author upon reasonable request.
Patients were eligible if they were aged 18 years or older at the time of the index event and had a documented malignancy diagnosis, identified using standardized international classification codes for malignant neoplasms across various anatomical sites. The study period extended from March 25, 2011—aligned with the US Food and Drug Administration approval of ipilimumab, the first anti‐CTLA‐4 (cytotoxic T‐lymphocyte‐associated protein 4) monoclonal antibody—to January 1, 2025. Data were sourced from electronic health records, using standardized terminologies such as RxNORM for medications, Healthcare Common Procedure Coding System for procedures, and international classification codes for diagnoses.
Cohort Definitions
The ICI‐associated pericarditis cohort included patients who received at least 1 ICI treatment, such as anti‐PD‐1 (programmed cell death protein) antibodies (nivolumab, cemiplimab, dostarlimab, pembrolizumab), anti‐PD‐L1 (programmed death ligand) antibodies (atezolizumab, avelumab, durvalumab), or an anti‐CTLA‐4 antibody (ipilimumab), identified through RxNORM and Healthcare Common Procedure Coding System codes (eg, nivolumab: RxNORM 1 597 876, Healthcare Common Procedure Coding System J9299). ICI administration was required to occur between March 25, 2011, and January 1, 2025, following a malignancy diagnosis. The index event was defined as the first instance of pericarditis occurring within 1 year after ICI initiation, with patients excluded if they had concurrent myocarditis diagnoses. A malignancy diagnosis before ICI initiation was mandatory.
The cohort with non‐ICI‐associated pericarditis comprised patients with a documented visit and a pericarditis diagnosis between March 25, 2011, and January 1, 2025, alongside a prior malignancy diagnosis but no record of ICI administration. Patients with any history of ICI‐related medications or procedures (eg, ipilimumab: RxNORM 1 094 833, Healthcare Common Procedure Coding System J9228) were excluded. The index event was defined as the first pericarditis diagnosis following a malignancy diagnosis, excluding those with concurrent myocarditis.
Similar separate analyses were performed to compare ICI‐associated myocarditis with non‐ICI‐associated myocarditis and ICI‐associated myopericarditis with non‐ICI‐associated myopericarditis.
Outcome Definitions
Primary outcomes assessed were all‐cause mortality, cardiac tamponade, constrictive pericarditis, need for pericardiocentesis, and pericardial window procedures, evaluated at 90‐day, 1‐year, and 5‐year follow‐up periods, with an additional 5‐year mortality analysis for subgroups. All‐cause mortality was identified through demographics (deceased status) or classification codes for ill‐defined or unknown causes of mortality. Cardiac tamponade was defined by specific diagnostic criteria for the condition. Constrictive pericarditis was identified by chronic constrictive pericarditis diagnoses. Need for pericardiocentesis was determined using procedure codes such as percutaneous insertion of a pericardial drainage tube or related interventions. Pericardial window outcomes were identified through codes for pericardium release or drainage procedures. Risk analysis and Kaplan–Meier survival analysis were employed, including patients with outcomes before the time window, except for the pericardial window, which excluded such patients.
Propensity Score Matching
Propensity score matching was conducted on 50 characteristics spanning demographics, diagnoses, and laboratory measures to reduce confounding. Demographic variables included age at index, sex, race, and ethnicity. Diagnoses covered cardiovascular conditions (eg, atrial fibrillation, heart failure), malignancies across multiple sites, and comorbidities (eg, diabetes, chronic kidney disease). Laboratory measures included CRP (C‐reactive protein), erythrocyte sedimentation rate (ESR), hemoglobin, left ventricular ejection fraction, troponin I, NT‐proBNP (N‐terminal‐pro‐B‐type natriuretic peptide), hemoglobin A1c, blood pressure, albumin, and glomerular filtration rate. Matching ensured balanced baseline characteristics, with pre‐ and postmatching characteristics summarized.
Statistical Analysis
Baseline and procedural characteristics between study groups were compared using either a 2‐sided Student’s t test or the Mann–Whitney U test for continuous variables, depending on the distribution of the data. Categorical variables were compared using analysis of variance. Continuous variables are presented as mean±SD, and categorical variables are reported as percentages. Multivariable Cox proportional hazards models and Kaplan–Meier survival curves were created for all‐cause mortality. The multivariable Cox proportional hazard model was adjusted to numerous demographics, clinical (including types of malignancy), and laboratory variables. A comprehensive list of the variables adjusted for is mentioned in the supplemental material. Subgroup analyses investigated 5‐year mortality by ICI class (anti‐PD‐1, anti‐PD‐L1, anti‐CTLA‐4), pericarditis treatment (colchicine, NSAIDs, steroids, dual/triple therapy), and malignancy type. A post hoc power calculation was performed using the observed effect size, sample size, and an alpha level of 0.05 to determine the power of the study. Analyses were conducted within the study period, with the current date of March 23, 2025, considered for data relevance.
RESULTS
Clinical Characteristics
Table 1 describes all the clinical characteristics of the patients with pericarditis in detail. A total of 4815 patients were identified to have pericarditis from ICI, and 85 095 without ICI in the unmatched cohort. The matched groups had 3661 in both cohorts. In the matched group, there was no significant difference in the mean age between the cohorts with pericarditis with ICI and without ICI (64±12 years versus 64±13 years, P=0.60). There were no significant differences in races and ethnicities. There was no significant difference in the mean ESR (38.21±31.17 versus 36.67±32.30, P=0.17) and CRP (P=0.51).
Table 1.
Baseline Clinical Characteristics of Patients With Pericarditis With and Without Immune Checkpoint Inhibitors
| Characteristic | Before propensity matching | SMD | After propensity matching | SMD | ||||
|---|---|---|---|---|---|---|---|---|
| ICI pericarditis (N=4815) | Non‐ICI pericarditis (N=85 095) | P value | ICI pericarditis (N=3661) | Non‐ICI pericarditis (N=3661) | P value | |||
| Demographics | ||||||||
| Age, y, mean±SD | 64±12 | 66±13 | <0.001 | 0.13 | 64±12 | 64±13 | 0.60 | 0.01 |
| Female sex, n (%) | 2218 (46.06) | 43 381 (50.98) | <0.001 | 0.04 | 1669 (45.59) | 1704 (46.55) | 0.17 | 0.00 |
| White race, n (%) | 3377 (70.14) | 58 232 (68.43) | 0.03 | 0.10 | 2578 (70.42) | 2572 (70.25) | 0.83 | 0.02 |
| Black race, n (%) | 549 (11.40) | 11 092 (13.04) | <0.001 | 0.05 | 409 (11.17) | 428 (11.69) | 0.80 | 0.02 |
| Hispanic or Latino ethnicity, n (%) | 268 (5.57) | 4704 (5.53) | 0.99 | 0.00 | 205 (5.60) | 208 (5.68) | 0.45 | 0.00 |
| Comorbidities | ||||||||
| Essential (primary) hypertension, n (%) | 3094 (64.26) | 52 688 (61.92) | <0.001 | 0.05 | 2343 (64.00) | 2316 (63.26) | 0.47 | 0.02 |
| Hyperlipidemia, n (%) | 2466 (51.22) | 44 906 (52.77) | 0.10 | 0.03 | 1864 (50.92) | 1837 (50.18) | 0.87 | 0.01 |
| Chronic lower respiratory diseases, n (%) | 2254 (46.81) | 27 669 (32.52) | <0.001 | 0.30 | 1638 (44.74) | 1623 (44.33) | 0.68 | 0.01 |
| Chronic ischemic heart disease, n (%) | 1512 (31.40) | 25 546 (30.02) | <0.001 | 0.03 | 1126 (30.76) | 1131 (30.89) | 0.48 | 0.00 |
| Diabetes, n (%) | 1298 (26.96) | 25 089 (29.48) | <0.001 | 0.06 | 993 (27.12) | 965 (26.36) | 0.55 | 0.02 |
| Chronic kidney disease, n (%) | 990 (20.56) | 20 898 (24.56) | <0.001 | 0.10 | 744 (20.32) | 734 (20.05) | 0.37 | 0.01 |
| Cerebrovascular diseases, n (%) | 995 (20.67) | 15 794 (18.56) | <0.001 | 0.05 | 743 (20.30) | 743 (20.30) | 0.33 | 0.00 |
| Peripheral vascular diseases, n (%) | 442 (9.18) | 8861 (10.41) | 0.01 | 0.04 | 328 (8.96) | 351 (9.59) | 0.70 | 0.02 |
| Acute myocardial infarction, n (%) | 412 (8.56) | 8474 (9.96) | <0.001 | 0.05 | 293 (8.00) | 302 (8.25) | 0.69 | 0.01 |
| Cardiomyopathy, unspecified, n (%) | 248 (5.15) | 5889 (6.92) | <0.001 | 0.07 | 163 (4.45) | 158 (4.32) | 0.55 | 0.01 |
| Heart failure, n (%) | 824 (17.11) | 20 388 (23.96) | <0.001 | 0.17 | 568 (15.52) | 573 (15.65) | 0.25 | 0.00 |
| Atrial fibrillation and flutter, n (%) | 918 (19.07) | 19 100 (22.45) | <0.001 | 0.08 | 643 (17.56) | 638 (17.43) | 0.75 | 0.00 |
| Pericardial conditions | ||||||||
| Pericardial effusion (noninflammatory), n (%) | 1234 (25.63) | 128 (0.15) | <0.001 | 0.82 | 105 (2.87) | 87 (2.38) | 0.52 | 0.03 |
| History of cardiac tamponade, n (%) | 271 (5.63) | 535 (0.63) | <0.001 | 0.29 | … | … | … | … |
| Malignant pericardial effusion, n (%) | 122 (2.53) | 10 (0.01) | <0.001 | 0.23 | 10 (0.27) | 10 (0.27) | 1.00 | 0.00 |
| Malignant neoplasms | ||||||||
| Melanoma and other malignant neoplasms of skin, n (%) | 796 (16.53) | 10 376 (12.19) | <0.001 | 0.12 | 634 (17.32) | 649 (17.73) | 0.38 | 0.01 |
| Malignant neoplasms of urinary tract, n (%) | 743 (15.43) | 5047 (5.93) | <0.001 | 0.31 | 576 (15.73) | 549 (15.00) | 0.84 | 0.02 |
| Malignant neoplasms of respiratory and intrathoracic organs, n (%) | 2839 (58.96) | 10 373 (12.19) | <0.001 | 1.12 | 1995 (54.49) | 2019 (55.15) | 0.26 | 0.01 |
| Malignant neoplasms of digestive organs, n (%) | 817 (16.97) | 10 430 (12.26) | <0.001 | 0.13 | 663 (18.11) | 733 (20.02) | 0.27 | 0.05 |
| Malignant neoplasms of ill‐defined, other secondary, and unspecified sites, n (%) | 4020 (83.49) | 19 739 (23.20) | <0.001 | 1.52 | 2972 (81.18) | 3051 (83.34) | 0.05 | 0.06 |
| Malignant neoplasms of lymphoid, hematopoietic and related tissue, n (%) | 503 (10.45) | 13 796 (16.21) | <0.001 | 0.17 | 391 (10.68) | 378 (10.33) | 0.64 | 0.01 |
| Malignant neoplasms of breast, n (%) | 395 (8.20) | 9677 (11.37) | <0.001 | 0.11 | 316 (8.63) | 325 (8.88) | 0.67 | 0.01 |
| Malignant neoplasms of mesothelial and soft tissue, n (%) | 361 (7.50) | 2620 (3.08) | <0.001 | 0.20 | 273 (7.46) | 309 (8.44) | 0.36 | 0.04 |
| Malignant neoplasms of female genital organs, n (%) | 317 (6.58) | 4157 (4.89) | <0.001 | 0.07 | 237 (6.47) | 265 (7.24) | 0.40 | 0.03 |
| Malignant neoplasms of lip, oral cavity, and pharynx, n (%) | 272 (5.65) | 1943 (2.28) | <0.001 | 0.17 | 222 (6.06) | 235 (6.42) | 0.96 | 0.01 |
| Malignant neoplasms of male genital organs, n (%) | 239 (4.96) | 6113 (7.18) | <0.001 | 0.09 | 192 (5.24) | 187 (5.11) | 0.83 | 0.01 |
| Malignant neoplasms of eye, brain, and other parts of the central nervous system, n (%) | 208 (4.32) | 1181 (1.39) | <0.001 | 0.18 | 167 (4.56) | 161 (4.40) | 0.95 | 0.01 |
| Malignant neuroendocrine tumors, n (%) | 167 (3.47) | 1194 (1.40) | <0.001 | 0.13 | 122 (3.33) | 127 (3.47) | 0.69 | 0.01 |
| Malignant neoplasms of bone and articular cartilage, n (%) | 150 (3.12) | 988 (1.16) | <0.001 | 0.14 | 119 (3.25) | 121 (3.31) | 0.27 | 0.00 |
| Malignant neoplasms of thyroid and other endocrine glands, n (%) | 124 (2.58) | 1924 (2.26) | 0.33 | 0.02 | 95 (2.60) | 102 (2.79) | 0.40 | 0.01 |
| Secondary neuroendocrine tumors, n (%) | 94 (1.95) | 480 (0.56) | <0.001 | 0.12 | 69 (1.89) | 69 (1.89) | 0.72 | 0.00 |
| Secondary malignant neoplasm of other and unspecified sites, n (%) | 2628 (54.58) | 9252 (10.87) | <0.001 | 1.05 | 1907 (52.09) | 1936 (52.88) | 0.83 | 0.02 |
| Secondary and unspecified malignant neoplasm of lymph nodes, n (%) | 2376 (49.35) | 7214 (8.48) | <0.001 | 1.01 | 1755 (47.94) | 1734 (47.36) | 0.12 | 0.01 |
| Secondary malignant neoplasm of respiratory and digestive organs, n (%) | 2126 (44.15) | 7380 (8.67) | <0.001 | 0.88 | 1560 (42.61) | 1602 (43.76) | 0.65 | 0.02 |
| Baseline laboratory values | ||||||||
| Albumin, g/dL, mean±SD | 3.49±0.63 | 3.65±0.69 | <0.001 | 0.24 | 3.49±0.63 | 3.49±0.72 | 0.73 | 0.01 |
| Hemoglobin, g/dL, mean±SD | 11.09±2.12 | 11.56±2.39 | <0.001 | 0.21 | 11.09±2.15 | 11.10±2.22 | 0.78 | 0.00 |
| 10–13.5 g/dL, n (%) | 4343 (90.20) | 58 455 (68.69) | <0.001 | 0.55 | 3281 (89.62) | 3284 (89.70) | 0.70 | 0.00 |
| 8–10 g/dL, n (%) | 3050 (63.34) | 34 939 (41.06) | <0.001 | 0.46 | 2309 (63.07) | 2305 (62.96) | 0.94 | 0.00 |
| At most 8 g/dL, n (%) | 1742 (36.18) | 21 702 (25.50) | <0.001 | 0.23 | 1319 (36.03) | 1316 (35.95) | 1.00 | 0.00 |
| Glomerular filtration rate, mL/min per 1.73 m2, mean±SD | 84.29±37.52 | 71.83±35.13 | <0.001 | 0.34 | 84.27±37.81 | 82.08±39.42 | 0.02 | 0.06 |
| Blood pressure, systolic, mm Hg, mean±SD | 119.67±19.97 | 124.31±23.59 | <0.001 | 0.22 | 119.77±19.86 | 121.02±21.44 | 0.03 | 0.06 |
| Blood pressure, diastolic, mm Hg, mean±SD | 70.79±11.95 | 70.82±13.34 | 0.72 | 0.00 | 70.88±11.94 | 70.13±12.56 | 0.03 | 0.06 |
| Hemoglobin A1c, %, mean±SD | 6.14±1.50 | 6.23±1.56 | <0.001 | 0.06 | 6.17±1.53 | 6.20±1.51 | 0.51 | 0.02 |
| C‐reactive protein, mg/L, mean±SD | 47.63±68.45 | 39.19±62.53 | <0.001 | 0.13 | 46.75±67.64 | 44.99±62.54 | 0.51 | 0.03 |
| Erythrocyte sedimentation rate, mm/h, mean±SD | 37.58±31.31 | 33.19±29.65 | <0.001 | 0.14 | 38.21±31.17 | 36.67±32.30 | 0.17 | 0.05 |
| Troponin I, ng/mL, mean±SD | 0.34±2.59 | 0.69±9.24 | 0.29 | 0.05 | 0.41±3.02 | 0.26±1.98 | 0.14 | 0.06 |
| N‐terminal‐pro‐B‐type natriuretic peptide, pg/mL, mean±SD | 1571.83±4100.17 | 3532.03±7976.43 | <0.001 | 0.31 | 1637.20±4398.31 | 2079.86±4985.20 | 0.13 | 0.09 |
| At most 400 pg/mL, n (%) | 465 (9.66) | 5782 (6.80) | <0.001 | 0.10 | 331 (9.04) | 312 (8.52) | 0.48 | 0.02 |
| 400–2000 pg/mL, n (%) | 321 (6.67) | 5620 (6.60) | 0.84 | 0.00 | 224 (6.12) | 213 (5.82) | 0.56 | 0.01 |
| At least 2000 pg/mL, n (%) | 174 (3.61) | 4846 (5.70) | <0.001 | 0.10 | 133 (3.63) | 144 (3.93) | 0.61 | 0.02 |
| Left ventricular ejection fraction, %, mean±SD | 58.77±11.55 | 57.04±13.36 | <0.001 | 0.14 | 59.08±11.51 | 58.89±11.90 | 0.81 | 0.02 |
| At most 35%, n (%) | 34 (0.71) | 978 (1.15) | <0.001 | 0.05 | 21 (0.57) | 23 (0.63) | 0.41 | 0.01 |
| 35%–50%, n (%) | 86 (1.79) | 1692 (1.99) | 0.32 | 0.01 | 51 (1.39) | 56 (1.53) | 0.56 | 0.01 |
| At least 50%, n (%) | 550 (11.42) | 8632 (10.14) | 0.01 | 0.04 | 365 (9.97) | 353 (9.64) | 0.48 | 0.01 |
| ICI therapies | ||||||||
| Pembrolizumab, n (%) | 2594 (53.87) | 0 (0.00) | <0.001 | 1.53 | … | … | … | … |
| Nivolumab, n (%) | 1283 (26.65) | 0 (0.00) | <0.001 | 0.85 | … | … | … | … |
| Ipilimumab, n (%) | 475 (9.87) | 0 (0.00) | <0.001 | 0.47 | … | … | … | … |
| Durvalumab, n (%) | 418 (8.68) | 0 (0.00) | <0.001 | 0.44 | … | … | … | … |
| Atezolizumab, n (%) | 436 (9.06) | 0 (0.00) | <0.001 | 0.45 | … | … | … | … |
| Cemiplimab, n (%) | 38 (0.79) | 0 (0.00) | <0.001 | 0.13 | … | … | … | … |
| Avelumab, n (%) | 23 (0.48) | 0 (0.00) | <0.001 | 0.10 | … | … | … | … |
| Dostarlimab, n (%) | 10 (0.21) | 0 (0.00) | <0.001 | 0.06 | … | … | … | … |
| Injection, pembrolizumab, 1 mg, n (%) | 1419 (29.47) | 0 (0.00) | <0.001 | 0.91 | … | … | … | … |
| Injection, nivolumab, 1 mg, n (%) | 692 (14.37) | 0 (0.00) | <0.001 | 0.58 | … | … | … | … |
| Injection, ipilimumab, 1 mg, n (%) | 246 (5.11) | 0 (0.00) | <0.001 | 0.33 | … | … | … | … |
| Injection, durvalumab, 10 mg, n (%) | 245 (5.09) | 0 (0.00) | <0.001 | 0.33 | … | … | … | … |
| Injection, atezolizumab, 10 mg, n (%) | 226 (4.69) | 0 (0.00) | <0.001 | 0.31 | … | … | … | … |
| Injection, cemiplimab‐rwlc, 1 mg, n (%) | 16 (0.33) | 0 (0.00) | <0.001 | 0.08 | … | … | … | … |
| Injection, avelumab, 10 mg, n (%) | 14 (0.29) | 0 (0.00) | <0.001 | 0.08 | … | … | … | … |
| Injection, dostarlimab‐gxly, 10 mg, n (%) | 10 (0.21) | 0 (0.00) | <0.001 | 0.06 | … | … | … | … |
ICI indicates immune checkpoint inhibitors; and SMD, standardized mean difference.
A total of 323 patients had myocarditis with ICI use and 2623 without ICI in the unmatched cohort. The matched cohorts had 226 patients in each group. After propensity matching, there was no significant difference in the mean age for the ICI and non‐ICI myocarditis group (67±14 years versus 66±14 years, P=0.05). There were no significant differences between races and ethnicities. Table 2 describes the clinical characteristics of patients with myocarditis in detail.
Table 2.
Baseline Clinical Characteristics of Patients With Myocarditis With and Without Immune Checkpoint Inhibitors
| Characteristic | Before propensity matching | SMD | After propensity matching | SMD | ||||
|---|---|---|---|---|---|---|---|---|
| ICI myocarditis (N=323) | Non‐ICI myocarditis (N=2623) | P value | ICI myocarditis (N=226) | Non‐ICI myocarditis (N=226) | P value | |||
| Demographics | ||||||||
| Age, y, mean±SD | 67±13 | 64±14 | 0.24 | 0.24 | 67±14 | 66±14 | 0.05 | 0.05 |
| Female sex, n (%) | 111 (34.37) | 1135 (43.27) | <0.001 | 0.05 | 81 (35.84) | 76 (33.63) | 0.92 | 0 |
| White race, n (%) | 226 (69.97) | 1775 (67.67) | 0.34 | 0.18 | 157 (69.47) | 157 (69.47) | 0.48 | 0.05 |
| Black race, n (%) | 24 (7.43) | 234 (8.92) | 0.24 | 0.05 | 19 (8.41) | 22 (9.74) | 0.87 | 0.05 |
| Hispanic or Latino ethnicity, n (%) | 19 (5.88) | 82 (3.13) | 0.02 | 0.13 | 15 (6.64) | 10 (4.43) | 0.84 | 0.05 |
| Comorbidities | ||||||||
| Essential (primary) hypertension, n (%) | 213 (65.94) | 1507 (57.45) | <0.001 | 0.18 | 152 (67.26) | 142 (62.83) | 0.42 | 0.09 |
| Hyperlipidemia, n (%) | 184 (56.97) | 1361 (51.89) | 0.10 | 0.1 | 129 (57.08) | 127 (56.20) | 0.93 | 0.02 |
| Chronic lower respiratory diseases, n (%) | 116 (35.91) | 747 (28.48) | <0.001 | 0.16 | 86 (38.05) | 81 (35.84) | 0.34 | 0.05 |
| Chronic ischemic heart disease, n (%) | 103 (31.89) | 874 (33.32) | 0.66 | 0.03 | 77 (34.07) | 77 (34.07) | 0.37 | 0 |
| Diabetes, n (%) | 82 (25.39) | 708 (26.99) | 0.40 | 0.04 | 62 (27.43) | 55 (24.34) | 0.75 | 0.07 |
| Chronic kidney disease, n (%) | 76 (23.53) | 546 (20.82) | 0.19 | 0.07 | 55 (24.34) | 45 (19.91) | 0.91 | 0.11 |
| Cerebrovascular diseases, n (%) | 51 (15.79) | 442 (16.85) | 0.60 | 0.03 | 41 (18.14) | 36 (15.93) | 0.80 | 0.06 |
| Other peripheral vascular diseases, n (%) | 28 (8.67) | 267 (10.18) | 0.45 | 0.05 | 22 (9.74) | 23 (10.18) | 0.65 | 0.01 |
| Acute myocardial infarction, n (%) | 54 (16.72) | 435 (16.58) | 0.66 | 0 | 41 (18.14) | 40 (17.70) | 0.80 | 0.01 |
| Cardiomyopathy, unspecified, n (%) | 23 (7.12) | 332 (12.66) | <0.001 | 0.19 | 20 (8.85) | 17 (7.52) | 0.60 | 0.05 |
| Heart failure, n (%) | 66 (20.43) | 775 (29.55) | <0.001 | 0.21 | 58 (25.66) | 49 (21.68) | 0.38 | 0.09 |
| Atrial fibrillation and flutter, n (%) | 58 (17.96) | 590 (22.49) | 0.10 | 0.11 | 48 (21.24) | 37 (16.37) | 0.30 | 0.12 |
| Pericardial conditions | ||||||||
| Pericardial effusion (noninflammatory), n (%) | 10 (3.10) | 10 (0.38) | <0.001 | 0.21 | 10 (4.43) | 0 (0.00) | 1.00 | 0.3 |
| History of cardiac tamponade, n (%) | 0 (0.00) | 10 (0.38) | 0.27 | 0.09 | … | … | … | … |
| Malignant pericardial effusion, n (%) | 0 (0.00) | 0 (0.00) | … | … | 0 (0.00) | 0 (0.00) | … | … |
| Malignant neoplasms | ||||||||
| Melanoma and other malignant neoplasms of skin, n (%) | 122 (37.77) | 389 (14.83) | <0.001 | 0.54 | 75 (33.19) | 74 (32.74) | 0.55 | 0.01 |
| Malignant neoplasms of respiratory and intrathoracic organs, n (%) | 97 (30.03) | 197 (7.51) | <0.001 | 0.6 | 59 (26.11) | 58 (25.66) | 0.25 | 0.01 |
| Malignant neoplasms of urinary tract, n (%) | 73 (22.60) | 170 (6.48) | <0.001 | 0.47 | 42 (18.58) | 49 (21.68) | 0.72 | 0.08 |
| Malignant neoplasms of digestive organs, n (%) | 58 (17.96) | 195 (7.43) | <0.001 | 0.32 | 36 (15.93) | 36 (15.93) | 0.71 | 0 |
| Malignant neoplasms of breast (C50), n (%) | 26 (8.05) | 237 (9.04) | 0.50 | 0.04 | 19 (8.41) | 18 (7.97) | 0.87 | 0.02 |
| Malignant neoplasms of male genital organs, n (%) | 28 (8.67) | 221 (8.43) | 0.76 | 0.01 | 18 (7.97) | 20 (8.85) | 0.33 | 0.03 |
| Malignant neoplasms of lymphoid, hematopoietic, and related tissue, n (%) | 20 (6.19) | 316 (12.05) | <0.001 | 0.2 | 15 (6.64) | 13 (5.75) | 0.40 | 0.04 |
| Malignant neoplasms of mesothelial and soft tissue, n (%) | 24 (7.43) | 45 (1.72) | <0.001 | 0.28 | 13 (5.75) | 11 (4.87) | 0.54 | 0.04 |
| Malignant neoplasms of female genital organs, n (%) | 23 (7.12) | 80 (3.05) | <0.001 | 0.19 | 13 (5.75) | 18 (7.97) | 0.84 | 0.09 |
| Malignant neoplasms of lip, oral cavity, and pharynx, n (%) | 17 (5.26) | 50 (1.91) | <0.001 | 0.18 | 10 (4.43) | 10 (4.43) | 0.83 | 0 |
| Malignant neoplasms of bone and articular cartilage, n (%) | 10 (3.10) | 13 (0.50) | <0.001 | 0.2 | 10 (4.43) | 10 (4.43) | 1.00 | 0 |
| Malignant neoplasms of eye, brain, and other parts of the central nervous system, n (%) | 12 (3.72) | 34 (1.30) | <0.001 | 0.16 | 10 (4.43) | 10 (4.43) | 1.00 | 0 |
| Malignant neoplasms of thyroid and other endocrine glands, n (%) | 10 (3.10) | 72 (2.75) | 0.79 | 0.02 | 10 (4.43) | 10 (4.43) | 1.00 | 0 |
| Malignant neuroendocrine tumors, n (%) | 10 (3.10) | 15 (0.57) | <0.001 | 0.19 | 10 (4.43) | 10 (4.43) | 1.00 | 0 |
| Secondary neuroendocrine tumors, n (%) | 10 (3.10) | 10 (0.38) | <0.001 | 0.21 | 10 (4.43) | 10 (4.43) | 1.00 | 0 |
| Malignant neoplasms of ill‐defined, other secondary, and unspecified sites, n (%) | 223 (69.04) | 363 (13.84) | <0.001 | 1.35 | 128 (56.64) | 120 (53.10) | 0.78 | 0.07 |
| Secondary malignant neoplasm of other and unspecified sites, n (%) | 132 (40.87) | 142 (5.41) | <0.001 | 0.93 | 70 (30.97) | 59 (26.11) | 0.92 | 0.11 |
| Secondary and unspecified malignant neoplasm of lymph nodes, n (%) | 115 (35.60) | 108 (4.12) | <0.001 | 0.86 | 56 (24.78) | 55 (24.34) | 0.51 | 0.01 |
| Secondary malignant neoplasm of respiratory and digestive organs, n (%) | 83 (25.70) | 93 (3.55) | <0.001 | 0.66 | 39 (17.26) | 36 (15.93) | 0.90 | 0.04 |
| Baseline laboratory values | ||||||||
| Hemoglobin, g/dL, mean±SD | 12.00±2.11 | 12.39±2.33 | <0.001 | 0.17 | 12.08±2.12 | 12.33±2.07 | 0.90 | 0.12 |
| 10–13.5 g/dL, n (%) | 279 (86.38) | 1573 (59.97) | <0.001 | 0.62 | 187 (82.74) | 189 (83.63) | 0.44 | 0.02 |
| 8–10 g/dL, n (%) | 151 (46.75) | 761 (29.01) | <0.001 | 0.37 | 99 (43.81) | 100 (44.25) | 0.78 | 0.01 |
| At most 8 g/dL, n (%) | 75 (23.22) | 417 (15.90) | <0.001 | 0.19 | 54 (23.89) | 53 (23.45) | 0.16 | 0.01 |
| Glomerular filtration rate, mL/min per 1.73 m2, mean±SD | 76.57±32.58 | 72.01±30.31 | 0.02 | 0.14 | 76.72±33.26 | 73.34±30.97 | 0.07 | 0.1 |
| Albumin, g/dL, mean±SD | 3.64±0.54 | 3.81±0.62 | <0.001 | 0.3 | 3.66±0.58 | 3.78±0.60 | 0.81 | 0.22 |
| Blood pressure, systolic, mm Hg, mean±SD | 123.16±21.30 | 123.84±21.31 | 0.58 | 0.03 | 122.51±22.21 | 126.82±21.72 | 0.08 | 0.2 |
| Blood pressure, diastolic, mm Hg, mean±SD | 70.73±11.54 | 71.81±12.87 | 0.11 | 0.09 | 69.88±11.69 | 73.41±11.86 | 0.04 | 0.3 |
| Hemoglobin A1c, %, mean±SD | 6.20±1.59 | 6.28±1.61 | 0.50 | 0.05 | 6.31±1.67 | 6.19±1.48 | 0.29 | 0.08 |
| C‐reactive protein, mg/L, mean±SD | 33.06±52.70 | 29.93±53.21 | 0.52 | 0.06 | 34.10±56.77 | 39.94±63.19 | 0.41 | 0.1 |
| Erythrocyte sedimentation rate, mm/h, mean±SD | 32.29±24.78 | 29.75±28.96 | 0.32 | 0.09 | 32.73±25.32 | 30.39±29.90 | 0.09 | 0.08 |
| N‐terminal‐pro‐B‐type natriuretic peptide, pg/mL, mean±SD | 2736.94±5449.48 | 4067.04±7512.44 | 0.17 | 0.2 | 3553.95±6292.52 | 3892.38±7812.75 | 0.25 | 0.05 |
| At most 400 pg/mL, n (%) | 30 (9.29) | 150 (5.72) | 0.01 | 0.14 | 18 (7.97) | 17 (7.52) | 0.74 | 0.02 |
| 400–2000 pg/mL, n (%) | 35 (10.84) | 149 (5.68) | <0.001 | 0.19 | 25 (11.06) | 25 (11.06) | 0.37 | 0 |
| At least 2000 pg/mL, n (%) | 22 (6.81) | 161 (6.14) | 0.51 | 0.03 | 18 (7.97) | 17 (7.52) | 0.15 | 0.02 |
| Troponin I (ng/mL), mean±SD | 1.04±3.15 | 1.85±9.18 | 0.54 | 0.12 | 1.29±3.55 | 0.52±1.61 | 0.31 | 0.28 |
| Left ventricular ejection fraction, %, mean±SD | 48.92±18.65 | 49.66±17.29 | 0.84 | 0.04 | 47.63±19.41 | 51.34±14.23 | 0.11 | 0.22 |
| At most 35%, n (%) | 10 (3.10) | 56 (2.14) | 0.26 | 0.06 | 10 (4.43) | 10 (4.43) | 1.00 | 0 |
| 35%–50%, n (%) | 10 (3.10) | 57 (2.17) | 0.28 | 0.06 | 10 (4.43) | 10 (4.43) | 1.00 | 0 |
| At least 50%, n (%) | 21 (6.50) | 172 (6.56) | 0.99 | 0 | 18 (7.97) | 17 (7.52) | 0.85 | 0.02 |
| ICI therapies | ||||||||
| Pembrolizumab, n (%) | 162 (50.16) | 0 (0.00) | <0.001 | 1.42 | … | … | … | … |
| Nivolumab, n (%) | 120 (37.15) | 0 (0.00) | <0.001 | 1.09 | … | … | … | … |
| Ipilimumab, n (%) | 68 (21.05) | 0 (0.00) | <0.001 | 0.73 | … | … | … | … |
| Durvalumab, n (%) | 18 (5.57) | 0 (0.00) | <0.001 | 0.31 | … | … | … | … |
| Atezolizumab, n (%) | 17 (5.26) | 0 (0.00) | <0.001 | 0.33 | … | … | … | … |
| Cemiplimab, n (%) | 10 (3.10) | 0 (0.00) | <0.001 | 0.25 | … | … | … | … |
| Dostarlimab, n (%) | 10 (3.10) | 0 (0.00) | <0.001 | 0.25 | … | … | … | … |
| Avelumab, n (%) | 10 (3.10) | 0 (0.00) | <0.001 | 0.25 | … | … | … | … |
| Injection, pembrolizumab, 1 mg, n (%) | 66 (20.43) | 0 (0.00) | <0.001 | 0.72 | … | … | … | … |
| Injection, nivolumab, 1 mg, n (%) | 47 (14.55) | 0 (0.00) | <0.001 | 0.58 | … | … | … | … |
| Injection, ipilimumab, 1 mg, n (%) | 30 (9.29) | 0 (0.00) | <0.001 | 0.45 | … | … | … | … |
| Injection, cemiplimab‐rwlc, 1 mg, n (%) | 10 (3.10) | 0 (0.00) | <0.001 | 0.25 | … | … | … | … |
| Injection, dostarlimab‐gxly, 10 mg, n (%) | 10 (3.10) | 0 (0.00) | <0.001 | 0.25 | … | … | … | … |
| Injection, atezolizumab, 10 mg, n (%) | 10 (3.10) | 0 (0.00) | <0.001 | 0.25 | … | … | … | … |
| Injection, durvalumab, 10 mg, n (%) | 10 (3.10) | 0 (0.00) | <0.001 | 0.21 | … | … | … | … |
| Injection, avelumab, 10 mg, n (%) | 0 (0.00) | 0 (0.00) | … | … | … | … | … | |
ICI indicates immune checkpoint inhibitors; and SMD, standardized mean difference.
A total of 84 patients had myopericarditis with ICI use and 877 without ICI in the unmatched cohort. After propensity matching, there were 46 patients in each group. In the matched cohort, there was no significant difference in mean age for ICI myopericarditis versus non‐ICI myopericarditis (64±14 years versus 64±12 years, P=0.82). There were no significant differences between races and ethnicities. Table 3 describes the clinical characteristics of patients with myopericarditis in detail.
Table 3.
Baseline Clinical Characteristics of Patients With Myopericarditis With and Without Immune Checkpoint Inhibitors
| Characteristic | Before propensity matching | SMD | After propensity matching | SMD | ||||
|---|---|---|---|---|---|---|---|---|
| ICI myopericarditis (N=84) | Non‐ICI myopericarditis (N=877) | P value | ICI myopericarditis (N=46) | Non‐ICI myopericarditis (N=46) | P value | |||
| Demographics | ||||||||
| Age, y, mean±SD | 66±12 | 59±16 | <0.001 | 0.45 | 64±14 | 64±12 | 0.82 | 0.05 |
| Female sex, n (%) | 37 (44.05) | 393 (44.81) | 0.89 | 0.47 | 22 (47.83) | 22 (47.83) | 1.00 | 0.21 |
| White race, n (%) | 68 (80.95) | 528 (60.21) | <0.001 | 0.02 | 38 (82.61) | 34 (73.91) | 0.31 | 0.00 |
| Black race, n (%) | 10 (11.91) | 157 (17.90) | 0.17 | 0.17 | 10 (21.74) | 10 (21.74) | 1.00 | 0.00 |
| Hispanic or Latino ethnicity, n (%) | 10 (11.91) | 49 (5.59) | 0.02 | 0.23 | 10 (21.74) | 10 (21.74) | 1.00 | 0.00 |
| Comorbidities | ||||||||
| Essential (primary) hypertension, n (%) | 56 (66.67) | 572 (65.22) | 0.79 | 0.03 | 33 (71.74) | 31 (67.39) | 0.65 | 0.09 |
| Hyperlipidemia, n (%) | 45 (53.57) | 497 (56.67) | 0.58 | 0.06 | 26 (56.52) | 27 (58.70) | 0.83 | 0.04 |
| Chronic lower respiratory diseases, n (%) | 36 (42.86) | 315 (35.92) | 0.21 | 0.14 | 19 (41.30) | 19 (41.30) | 1.00 | 0.00 |
| Chronic ischemic heart disease, n (%) | 28 (33.33) | 381 (43.44) | 0.07 | 0.21 | 17 (36.96) | 19 (41.30) | 0.67 | 0.09 |
| Diabetes, n (%) | 22 (26.19) | 307 (35.01) | 0.10 | 0.19 | 10 (21.74) | 11 (23.91) | 0.80 | 0.05 |
| Chronic kidney disease, n (%) | 20 (23.81) | 303 (34.55) | 0.05 | 0.24 | 10 (21.74) | 10 (21.74) | 1.00 | 0.00 |
| Cerebrovascular diseases, n (%) | 22 (26.19) | 224 (25.54) | 0.90 | 0.01 | 11 (23.91) | 10 (21.74) | 0.80 | 0.05 |
| Other peripheral vascular diseases, n (%) | 10 (11.91) | 125 (14.25) | 0.55 | 0.07 | 10 (21.74) | 10 (21.74) | 1.00 | 0.00 |
| Acute myocardial infarction, n (%) | 23 (27.38) | 251 (28.62) | 0.81 | 0.03 | 13 (28.26) | 12 (26.09) | 0.81 | 0.05 |
| Cardiomyopathy, unspecified, n (%) | 14 (16.67) | 255 (29.08) | 0.02 | 0.30 | 10 (21.74) | 10 (21.74) | 1.00 | 0.00 |
| Heart failure, n (%) | 28 (33.33) | 429 (48.92) | 0.01 | 0.32 | 17 (36.96) | 16 (34.78) | 0.83 | 0.05 |
| Atrial fibrillation and flutter, n (%) | 15 (17.86) | 313 (35.69) | <0.001 | 0.41 | 11 (23.91) | 11 (23.91) | 1.00 | 0.00 |
| Pericardial conditions | ||||||||
| Pericardial effusion (noninflammatory), n (%) | 31 (36.91) | 290 (33.07) | 0.48 | 0.08 | 12 (26.09) | 20 (43.48) | 0.08 | 0.37 |
| History of cardiac tamponade, n (%) | 0 (0.00) | 52 (5.93) | 0.02 | 0.36 | … | … | … | … |
| Malignant pericardial effusion, n (%) | 0 (0.00) | 10 (1.14) | 0.33 | 0.15 | 0 (0.00) | 0 (0.00) | … | … |
| Malignant neoplasms | ||||||||
| Melanoma and other malignant neoplasms of skin, n (%) | 31 (36.91) | 128 (14.60) | <0.001 | 0.53 | 19 (41.30) | 10 (21.74) | 0.04 | 0.43 |
| Malignant neoplasms of respiratory and intrathoracic organs, n (%) | 35 (41.67) | 61 (6.96) | <0.001 | 0.88 | 13 (28.26) | 15 (32.61) | 0.65 | 0.09 |
| Malignant neoplasms of lip, oral cavity, and pharynx, n (%) | 10 (11.91) | 13 (1.48) | <0.001 | 0.43 | 10 (21.74) | 10 (21.74) | 1.00 | 0.00 |
| Malignant neoplasms of digestive organs, n (%) | 17 (20.24) | 61 (6.96) | <0.001 | 0.40 | 10 (21.74) | 11 (23.91) | 0.80 | 0.05 |
| Malignant neoplasms of mesothelial and soft tissue, n (%) | 10 (11.91) | 19 (2.17) | <0.001 | 0.39 | 10 (21.74) | 10 (21.74) | 1.00 | 0.00 |
| Malignant neoplasms of breast, n (%) | 10 (11.91) | 83 (9.46) | 0.47 | 0.08 | 10 (21.74) | 10 (21.74) | 1.00 | 0.00 |
| Malignant neoplasms of female genital organs, n (%) | 10 (11.91) | 27 (3.08) | <0.001 | 0.34 | 10 (21.74) | 10 (21.74) | 1.00 | 0.00 |
| Malignant neoplasms of male genital organs, n (%) | 10 (11.91) | 58 (6.61) | 0.07 | 0.18 | 10 (21.74) | 10 (21.74) | 1.00 | 0.00 |
| Malignant neoplasms of urinary tract, n (%) | 13 (15.48) | 44 (5.02) | <0.001 | 0.35 | 10 (21.74) | 10 (21.74) | 1.00 | 0.00 |
| Malignant neoplasms of eye, brain, and other parts of the central nervous system, n (%) | 10 (11.91) | 11 (1.25) | <0.001 | 0.44 | 10 (21.74) | 10 (21.74) | 1.00 | 0.00 |
| Malignant neoplasms of thyroid and other endocrine glands, n (%) | 10 (11.91) | 17 (1.94) | <0.001 | 0.40 | 10 (21.74) | 10 (21.74) | 1.00 | 0.00 |
| Malignant neoplasms of lymphoid, hematopoietic, and related tissue, n (%) | 10 (11.91) | 213 (24.29) | 0.01 | 0.33 | 10 (21.74) | 10 (21.74) | 1.00 | 0.00 |
| Malignant neoplasms of bone and articular cartilage, n (%) | 10 (11.91) | 10 (1.14) | <0.001 | 0.45 | 0 (0.00) | 10 (21.74) | <0.001 | 0.75 |
| Malignant neuroendocrine tumors, n (%) | 10 (11.91) | 10 (1.14) | <0.001 | 0.45 | 0 (0.00) | 10 (21.74) | <0.001 | 0.75 |
| Malignant neoplasms of ill‐defined, other secondary, and unspecified sites, n (%) | 68 (80.95) | 157 (17.90) | <0.001 | 1.62 | 32 (69.57) | 36 (78.26) | 0.34 | 0.20 |
| Secondary neuroendocrine tumors, n (%) | 10 (11.91) | 10 (1.14) | <0.001 | 0.45 | 0 (0.00) | 10 (21.74) | <0.001 | 0.75 |
| Secondary and unspecified malignant neoplasm of lymph nodes, n (%) | 34 (40.48) | 57 (6.50) | <0.001 | 0.87 | 18 (39.13) | 17 (36.96) | 0.83 | 0.04 |
| Secondary malignant neoplasm of other and unspecified sites, n (%) | 40 (47.62) | 80 (9.12) | <0.001 | 0.94 | 17 (36.96) | 23 (50.00) | 0.21 | 0.27 |
| Secondary malignant neoplasm of respiratory and digestive organs, n (%) | 37 (44.05) | 45 (5.13) | <0.001 | 1.01 | 16 (34.78) | 18 (39.13) | 0.67 | 0.09 |
| Baseline laboratory values | ||||||||
| Hemoglobin, g/dL, mean±SD | 11.40±2.23 | 11.38±2.53 | 0.94 | 0.01 | 11.48±2.22 | 10.75±2.13 | 0.12 | 0.34 |
| 10–13.5 g/dL, n (%) | 76 (90.48) | 698 (79.59) | 0.02 | 0.31 | 43 (93.48) | 43 (93.48) | 1.00 | 0.00 |
| 8–10 g/dL, n (%) | 52 (61.91) | 496 (56.56) | 0.34 | 0.11 | 27 (58.70) | 34 (73.91) | 0.12 | 0.33 |
| At most 8 g/dL, n (%) | 31 (36.91) | 322 (36.72) | 0.97 | 0.00 | 17 (36.96) | 20 (43.48) | 0.52 | 0.13 |
| Albumin, g/dL, mean±SD | 3.48±0.59 | 3.60±0.69 | 0.15 | 0.18 | 3.45±0.59 | 3.44±0.72 | 0.97 | 0.01 |
| Blood pressure, systolic, mm Hg, mean±SD | 123.52±20.89 | 117.05±22.97 | 0.02 | 0.29 | 127.56±18.53 | 115.31±22.17 | 0.01 | 0.60 |
| Blood pressure, diastolic, mm Hg, mean±SD | 72.51±12.75 | 69.16±13.99 | 0.04 | 0.25 | 75.00±13.36 | 66.38±8.48 | <0.001 | 0.77 |
| Glomerular filtration rate, mL/min per 1.73 m2, mean±SD | 77.09±35.27 | 71.84±35.70 | 0.21 | 0.15 | 76.37±37.03 | 86.26±46.27 | 0.28 | 0.24 |
| C‐reactive protein, mg/L, mean±SD | 42.48±65.92 | 52.35±72.63 | 0.40 | 0.14 | 30.77±57.03 | 58.43±54.59 | 0.11 | 0.48 |
| Hemoglobin A1c, %, mean±SD | 5.96±1.20 | 6.04±1.34 | 0.70 | 0.06 | 5.87±1.41 | 6.10±1.00 | 0.52 | 0.19 |
| Erythrocyte sedimentation rate, mm/h, mean±SD | 40.16±36.85 | 37.64±31.62 | 0.33 | 0.07 | 39.00±37.97 | 34.47±31.59 | 0.14 | 0.13 |
| Troponin I, ng/mL, mean±SD | 0.82±2.79 | 0.85±2.52 | 0.67 | 0.01 | 0.24±0.66 | 0.95±2.17 | 0.70 | 0.44 |
| N‐terminal‐pro‐B‐type natriuretic peptide, pg/mL, mean±SD | 3554.89±4927.39 | 3433.38±5725.85 | 0.96 | 0.02 | 4514.25±5872.87 | 1631.67±1371.12 | 0.29 | 0.68 |
| At most 400 pg/mL, n (%) | 10 (11.91) | 99 (11.29) | 0.93 | 0.02 | 10 (21.74) | 10 (21.74) | 0.11 | 0.00 |
| 400–2000 pg/mL, n (%) | 10 (11.91) | 142 (16.19) | 0.86 | 0.12 | 10 (21.74) | 10 (21.74) | 1.00 | 0.00 |
| At least 2000 pg/mL, n (%) | 10 (11.91) | 145 (16.53) | 0.30 | 0.13 | 10 (21.74) | 10 (21.74) | 1.00 | 0.00 |
| Left ventricular ejection fraction, %, mean±SD | 56.80±10.47 | 51.79±17.47 | 0.27 | 0.35 | 57.90±9.12 | 56.88±11.17 | 1.00 | 0.10 |
| At most 35%, n (%) | 10 (11.91) | 52 (5.93) | 0.22 | 0.21 | 10 (21.74) | 10 (21.74) | 0.82 | 0.00 |
| 35%–50%, n (%) | 10 (11.91) | 58 (6.61) | 0.03 | 0.18 | 10 (21.74) | 10 (21.74) | 1.00 | 0.00 |
| At least 50%, n (%) | 18 (21.43) | 139 (15.85) | 0.07 | 0.14 | 10 (21.74) | 10 (21.74) | 1.00 | 0.00 |
| ICI therapies | ||||||||
| Nivolumab, n (%) | 39 (46.43) | 0 (0.00) | 0.19 | 1.32 | … | … | … | … |
| Pembrolizumab, n (%) | 37 (44.05) | 0 (0.00) | <0.001 | 1.25 | … | … | … | … |
| Ipilimumab, n (%) | 22 (26.19) | 0 (0.00) | <0.001 | 0.84 | … | … | … | … |
| Atezolizumab, n (%) | 10 (11.91) | 0 (0.00) | <0.001 | 0.52 | … | … | … | … |
| Avelumab, n (%) | 10 (11.91) | 0 (0.00) | <0.001 | 0.52 | … | … | … | … |
| Durvalumab, n (%) | 10 (11.91) | 0 (0.00) | <0.001 | 0.52 | … | … | … | … |
| Cemiplimab, n (%) | 0 (0.00) | 0 (0.00) | … | … | … | … | … | … |
| Dostarlimab, n (%) | 0 (0.00) | 0 (0.00) | … | … | … | … | … | … |
| Injection, nivolumab, 1 mg, n (%) | 20 (23.81) | 0 (0.00) | … | 0.79 | … | … | … | … |
| Injection, pembrolizumab, 1 mg, n (%) | 18 (21.43) | 0 (0.00) | <0.001 | 0.74 | … | … | … | … |
| Injection, durvalumab, 10 mg, n (%) | 10 (11.91) | 0 (0.00) | <0.001 | 0.52 | … | … | … | … |
| Injection, ipilimumab, 1 mg, n (%) | 12 (14.29) | 0 (0.00) | <0.001 | 0.58 | … | … | … | … |
| Injection, cemiplimab‐rwlc, 1 mg, n (%) | 0 (0.00) | 0 (0.00) | <0.001 | … | … | … | … | … |
| Injection, dostarlimab‐gxly, 10 mg, n (%) | 0 (0.00) | 0 (0.00) | … | … | … | … | ||
| Injection, atezolizumab, 10 mg, n (%) | 10 (11.91) | 0 (0.00) | … | 0.52 | … | … | … | |
| Injection, avelumab, 10 mg, n (%) | 0 (0.00) | 0 (0.00) | <0.001 | … | … | … | ||
ICI indicates immune checkpoint inhibitors; and SMD, standardized mean difference.
Outcomes
Table 4 describes the 90‐day, 1‐year, and 5‐year outcomes of patients with pericarditis, myocarditis, and myopericarditis with and without ICI. In the pericarditis cohort, there were 41 (1.12%), 16 (0.43%), and 0 (0.0%) patients on ICI therapy at 90‐day, 1‐year, and 5‐year follow‐up, respectively. Similarly, in the myocarditis cohort, there were <10 patients at 90‐day and 1‐year follow‐up, and no patients on ICI therapy at 5‐year follow‐up. In the myopericarditis cohort, there were no patients on ICI therapy at 90‐day, 1‐year, and 5‐year follow‐up. In 90 days, there was no significant difference in all‐cause mortality when comparing ICI versus non‐ICI pericarditis (26.6% versus 25.2%, P=0.173). There was a higher ICI‐related mortality at 1 year (44.3% versus 37.3%, P<0.001) and 5 years (54.3% versus 46.9%, P<0.001). Similarly, there was a higher 90‐day mortality in the group with ICI‐myocarditis (28.1% versus 15.6%, P=0.001), 1‐year (38.9% versus 20.8%, P<0.001), and 5‐year (47.8% versus 34.3%, P=0.002). No difference in 90‐day all‐cause mortality was witnessed between ICI and non‐ICI myopericarditis (43.5% versus 26.1%, P=0.08), but there was a higher 1‐year (56.5% versus 28.3%, P<0.001) and 5‐year (63.0% versus 28.3%, P=0.01) mortality. Post hoc power calculations were performed for mortality outcomes across the cohorts with pericarditis, myocarditis, and myopericarditis. For the cohort with pericarditis, the statistical power was 16.6% for 90‐day mortality and 100% for both 1‐year and 5‐year mortality. In the cohort with myocarditis, the power was 92.2% for 90‐day mortality, 98.9% for 1‐year mortality, and 80.6% for 5‐year mortality. In the cohort with myopericarditis, the power was 40.2% for 90‐day mortality and 100% for both 1‐year and 5‐year mortality.
Table 4.
Ninety‐Day, 1‐Year, and 5‐Year Outcomes of Immune Checkpoint Inhibitor‐Associated Pericarditis, Myocarditis, and Myopericarditis
| 90‐d outcomes | 1‐y outcomes | 5‐y outcomes | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Pericarditis | |||||||||
| Outcomes | Pericarditis with ICI (N=3661), n (%) | Pericarditis without ICI (N=3661), n (%) | P value | Pericarditis with ICI (N=3661), n (%) | Pericarditis without ICI (N=3661), n (%) | P value | Pericarditis with ICI | Pericarditis without ICI | P value |
| (N=3661), n (%) | (N=3661), n (%) | ||||||||
| All‐cause mortality | 933 (26.6%) | 883 (25.2%) | 0.173 | 1635 (44.3%) | 1375 (37.3%) | <0.001 | 1988 (54.3%) | 1717 (46.9%) | <0.001 |
| New tamponade | 177 (5.0%) | 174 (5.0%) | 0.87 | 233 (6.3%) | 222 (6.0%) | 0.594 | 241 (6.6%) | 248 (6.8%) | 0.743 |
| Constriction | 16 (0.5%) | 14 (0.4%) | 0.714 | 21 (0.6%) | 22 (0.6%) | 0.878 | 23 (0.6%) | 22 (0.6%) | 0.238 |
| Need for pericardiocentesis | 174 (5.0%) | 176 (5.0%) | 0.913 | 219 (5.9%) | 219 (5.9%) | 1.000 | 231 (6.3%) | 233 (6.4%) | 0.924 |
| Myocarditis | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| Outcomes | Myocarditis with ICI (N=226), n (%) | Myocarditis without ICI (N=226), n (%) | P value | Myocarditis with ICI (N=226), n (%) | Myocarditis without ICI (N=226), n (%) | P value | Myocarditis with ICI (N‐268), n (%) | Myocarditis without ICI (N=268), n (%) | P value |
| All‐cause mortality | 65 (28.1%) | 36 (15.6%) | 0.001 | 88 (38.9%) | 47 (20.8%) | <0.001 | 128 (47.8%) | 92 (34.3%) | 0.002 |
| New tamponade | 0 (0%) | ≤10 | NA | 0 (0%) | ≤10 | NA | 0 (0%) | ≤10 | NA |
| Constriction | 0 | 0 | NA | 0 | 0 | NA | 0 | 0 | NA |
| Need for pericardiocentesis | 0 | 0 | NA | 0 | 0 | NA | 0 | 0 | NA |
| Myopericarditis | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| Outcomes | Myopericarditis with ICI (N=46), n (%) | Myopericarditis without ICI (N=46), n (%) | P value | Myopericarditis with ICI (N=46), n (%) | Myopericarditis without ICI (N=46), n (%) | P value | Myopericarditis with ICI (N=46), n (%) | Myopericarditis without ICI (N=46), n (%) | P value |
| All‐cause mortality | 20 (43.5%) | 12 (26.1%) | 0.08 | 26 (56.5%) | 13 (28.3%) | <0.001 | 29 (63.0%) | 13 (28.3%) | 0.001 |
| New tamponade | 0 (0%) | ≤10 | NA | 0 (0%) | ≤10 | NA | 0 (0%) | ≤10 | NA |
| Constriction | 0 | 0 | NA | 0 | 0 | NA | 0 | 0 | NA |
| Need for pericardiocentesis | 0 | 0 | NA | 0 | 0 | NA | 0 | 0 | NA |
ICI indicates immune checkpoint inhibitor; and NA, not applicable.
Outcome events <10 could not be documented due to the risk of exposing patient health information.
A multivariable Cox proportional hazard model showed a higher ICI‐related pericarditis mortality rate in 1‐year (hazard ratio [HR], 1.20 [95% CI, 1.12–1.29], P<0.001) and 5‐year follow‐up (HR, 1.26 [95% CI, 1.18–1.35], P<0.001) with no significant differences in 90‐day mortality (HR, 1.03 [95% CI, 0.94–1.12], P=0.601). For myocarditis, there was a higher ICI‐related 90‐day (HR, 1.82 [95% CI, 1.21–2.73], P<0.001), 1‐year (HR, 2.04 [95% CI, 1.43–2.91], P<0.001), and 5‐year (HR, 1.64 [95% CI, 1.26–2.16], P<0.001) mortality. Similarly, for myopericarditis, there was a higher 1‐year (HR, 2.24 [95% CI, 1.15–4.37], P<0.001) and 5‐year (HR, 2.78 [95% CI, 1.44–5.36], P<0.001) mortality with no significant difference in 90‐day mortality (HR, 1.77 [95% CI, 0.86–3.62], P=0.113). Figures 1A, 1B, and 1C, Figures 2A, 2B, and 2C, and Figures 3A, 3B, and 3C describe the Kaplan–Meier survival curves for 90‐day‐, 1‐year‐, and 5‐year follow‐up for pericarditis, myocarditis, and myopericarditis, respectively. Corresponding to Table 4, Figures 4A, 4B, and 4C depict the 90‐day, 1‐year, and 5‐year outcomes of patients with pericarditis, myocarditis, and myopericarditis with and without ICI. A multivariable Cox proportional hazard model subgroup analysis was also performed comparing ICI pericarditis without effusion and non‐ICI related pericarditis without effusion, showing that ICI‐related pericarditis without effusion had higher 5‐year mortality (HR, 1.65 [95% CI, 1.06–2.57], P<0.001) as compared with non‐ICI‐related pericarditis without effusion with no significant difference between 90‐day (HR 0.99 [95% CI, 0.54–1.82], P=1.00) and 1‐year (HR 0.93 [95% CI, 0.57–1.53], P=0.752) mortality. The Kaplan–Meier survival curve is shown in Figure S1.
Figure 1. Kaplan–Meier survival curves for (A) 90‐day, (B) 1‐year, and (C) 5‐year follow‐up of immune checkpoint inhibitor‐induced pericarditis versus nonimmune checkpoint inhibitor‐induced pericarditis.
Figure 2. Kaplan–Meier survival curves for (A) 90‐day, (B) 1‐year, and (C) 5‐year follow‐up of immune checkpoint inhibitor‐induced myocarditis versus nonimmune checkpoint inhibitor‐induced myocarditis.
Figure 3. Kaplan–Meier survival curves for (A), 90‐day, (B), 1‐year, and (C), 5‐year follow‐up of immune checkpoint inhibitor‐induced myopericarditis versus nonimmune checkpoint inhibitor‐induced myopericarditis.
Figure 4. Bar graph representing 90‐day, 1‐year, and 5‐year outcomes of all‐cause mortality, cardiac tamponade, constrictive pericarditis, and need for pericardiocentesis in (A) pericarditis, (B) myocarditis, and (C) myopericarditis with immune checkpoint inhibitors versus nonimmune checkpoint inhibitor.
ICI indicates immune chekpoint inhibitor.
Subgroup analyses to assess 5‐year mortality were performed according to ICI class, cancer type, and treatment modalities compared with the general cohort with non‐ICI pericarditis. Among ICI classes, the PD‐1/CTLA‐4 combination showed the highest risk with an HR of 4.39 (95% CI, 3.15–6.11, P<0.001), followed by PD‐1 alone (HR, 3.06 [95% CI, 2.70–3.48], P<0.001) and PD‐L1 (HR, 2.44 [95% CI, 1.94–3.06], P<0.001). For cancer types, gastrointestinal malignancies exhibited the highest mortality risk (HR, 4.28 [95% CI, 2.29–8.00], P<0.001), followed by respiratory/intrathoracic cancers (HR, 2.75 [95% CI, 2.07–3.64], P<0.001) whereas lymphoid/hematopoietic cancers showed a lower and not statistically significant risk (HR, 1.41 [95% CI, 0.72–2.74], P=0.840). Regarding treatment, NSAID use was associated with an HR of 2.97 (95% CI, 2.58–3.43, P<0.001), steroids with an HR of 2.82 (95% CI, 2.48–3.21, P<0.001), and colchicine with an HR of 2.33 (95% CI, 1.70–3.18, P<0.001). Combination therapies showed slightly lower risks, with NSAID plus colchicine at an HR of 2.44 (95% CI, 1.68–3.56, P<0.001) and triple therapy (NSAID, colchicine, and steroid) at an HR of 1.72 (95% CI, 1.11–2.65, P=0.01), suggesting a potential moderation of risk with more intensive treatment regimens. The forest plots are presented in the Figures S2A, S2B, and S2C. The central illustration describes the core findings of our study.
DISCUSSION
In the largest study to date of 3661 patients with ICI‐associated pericarditis, we showed that patients with ICI‐associated pericarditis experience increased long‐term all‐cause mortality compared with propensity‐matched patients with pericarditis not associated with ICIs. Similarly, among the 226 ICI‐related myocarditis there was higher short‐ and long‐term all‐cause mortality in patients with ICI‐related myocarditis as compared with matched patients with non‐ICI‐related myocarditis. Among 46 patients with ICI‐related myopericarditis, there was a higher long‐term all‐cause mortality when compared with non‐ICI‐related myopericarditis. There were no significant differences in 90‐day, 1‐year, and 5‐year outcomes of tamponade, constriction, and the need for pericardiocentesis between the cohorts with pericarditis with and without ICI. ICIs have become increasingly used in the pharmacotherapeutic armamentarium of cancer treatment. However, ICIs are associated with numerous systemic irAEs that have become increasingly recognized. The cumulative incidence of irAEs varies widely and has been reported to range from 15% to 90%. 14 , 15 The mechanism of irAEs is incompletely understood but is thought to be related to autoreactive T cells leading to autoimmune reactions affecting both on‐target and off‐target organs. 16 , 17 , 18 The pathophysiological underpinnings of myocarditis or pericardial disease related to ICI use are also not fully elucidated. The current hypothesis is related to a shared antigen between cardiomyocytes and tumor cells. 19 , 20 Despite the growing popularity, utility, and awareness of ICIs and irAEs, there are limited data regarding ICI‐associated pericardial or myocardial disease. The observed associations between ICI and myocarditis or pericarditis can be explained by certain key immunopathological mechanisms. ICIs block inhibitory pathways (CTLA‐4, PD‐1, PD‐L1), leading to unchecked T‐cell activation. This can result in cytotoxic T‐cell infiltration of cardiac tissue, particularly by CD8+ T cells, which are implicated as central mediators of myocardial injury. Both murine models and human studies have demonstrated clonal expansion of effector CD8+ T cells in the myocardium during ICI‐associated myocarditis, with evidence that these T cells can recognize cardiac‐specific antigens such as α‐myosin, a protein not expressed in the thymus and thus not subject to central tolerance. 21 , 22 , 23 The immune response unleashed by ICIs may target cardiac antigens that share homology with tumor antigens, or may simply break tolerance to cardiac self‐antigens, leading to autoimmunity. This is supported by the identification of cardiac myosin‐specific T cells in both animal models and patients with ICI myocarditis. 22 , 23 Moreover, ICIs promote upregulation of proinflammatory cytokines (eg, IFN‐γ [interferon‐gamma], TNF‐α [tumor necrosis factor alpha], IL [interleukin]‐1β, IL‐6, IL‐17), which can exacerbate cardiac inflammation, endothelial dysfunction, and potentially contribute to pericardial involvement. 24 , 25
Numerous observational studies, case series, and case reports of ICI‐associated pericardial disease have been established. 26 Sharma et al recently performed a meta‐analysis of 37 randomized controlled trials and 19 observational studies separately and showed an increased risk of pericardial disease in patients on ICIs when compared with patients on non‐ICI treatment (odds ratio [OR], 2.73 [95% CI, 1.57–4.77], P<0.01). 27 A systematic review of 31 studies by Mudra et al identified 38 patients who developed ICI‐associated pericarditis. 28
Although there are numerous studies assessing the risk of ICI‐associated pericardial disease, there are fewer investigations of outcomes in those patients who develop pericarditis or myocarditis. Our study showed a higher mortality risk with ICI‐induced myocarditis, which indicates a significant risk. According to an analysis of patients with this condition, the onset of symptoms typically occurs early, with a median interval of 34 days after the initiation of therapy. The mortality rate in these cases is notably high, at approximately 50%. 10 The high mortality associated with ICI‐induced myocarditis is further supported by several studies. A pooled analysis of trials sponsored by the National Cancer Institute‐Cancer Therapy Evaluation Program found that myocarditis accounted for 45% of ICI‐related major adverse cardiac events, with a mortality rate of 22.5% among those affected. 5 Another retrospective study from a single institution reported that 42% of patients with ICI‐related myocarditis experienced major adverse cardiac events, which significantly affected overall survival. 29 A meta‐analysis highlighted that the incidence of myocarditis was higher with dual ICI therapies compared with monotherapies, and the mortality rate for myocarditis was ∼37.7%. 30 Additionally, a pharmacovigilance study using the World Health Organization’s VigiBase database reported that myocarditis was associated with a high mortality rate of 50% among patients treated with ICIs. 31
Several other studies have also observed higher mortality in ICI‐induced pericarditis and myopericarditis. A study by Kwan et al found that dual ICI therapy was associated with a 2.0‐fold increased risk of myopericarditis, which in turn was associated with a 2.9‐fold increased risk of mortality. 32 Gong et al reported that patients who developed pericardial disease while on ICIs had a trend toward increased all‐cause mortality compared with those who did not develop pericardial events (HR, 1.53 [95% CI, 0.99–2.36],P=0.05). 33 This suggests a potential association between ICI‐induced pericarditis and higher mortality, although the statistical significance was marginal. Additionally, Ma et al found that ICI‐associated pericardial toxicities can result in serious outcomes, with fatal cases occurring earlier than nonfatal cases (median time to onset of 25 days versus 42 days, P=0.003). 34 Tang et al conducted a retrospective study that showed a nearly 5‐fold increase in 1‐year all‐cause mortality in patients with cancer who developed pericardial effusion after PD‐1 inhibitor therapy compared with those who did not (64.10% versus 13.46%, P<0.001). 35 Lastly, a systematic review by Inno et al of case reports and series found that although pericardial disease was reversible in the majority of cases, it was life‐threatening in 53.6% of cases and severe in 21.4%, with 2 deaths reported. 36
There are a few studies that align with our result of no significant differences in 90‐day, 1‐year, and 5‐year outcomes of tamponade, constriction, and the need for pericardiocentesis between pericarditis with and without ICI exposure. The retrospective cohort study by Palaskas et al directly compared patients with cancer who developed pericardial effusions requiring pericardiocentesis while on ICI therapy versus those on other cancer therapeutics. The study found that survival durations after pericardiocentesis were similar between the ICI and non‐ICI groups, suggesting no significant difference in short‐ or intermediate‐term outcomes, including recurrence and need for repeat pericardiocentesis, between these cohorts. 37 Additionally, Gong et al performed a large retrospective analysis comparing pericardial events in ICI‐treated patients versus matched controls. Although ICI use was associated with a higher incidence of pericardial disease, the study did not demonstrate significant differences in outcomes such as tamponade, constriction, or procedural interventions when comparing pericarditis cases with and without ICI exposure. There was no evidence of increased rates of tamponade or constrictive physiology specifically attributable to ICI use. 33
These findings have implications both with respect to clinicians understanding the prognosis of these patients and also caring for patients who already have underlying cardiovascular disease risk factors and other comorbidities. Early recognition and management of ICI‐related pericardial and myocardial disorders are crucial to improving outcomes as delayed treatment can lead to worse outcomes. The American Society of Clinical Oncology (ASCO) guidelines also highlight the severe nature of cardiovascular toxicities, including pericarditis, associated with ICIs, noting that these events can lead to high mortality due to complications such as refractory arrhythmia or cardiogenic shock. 38 The ASCO guidelines also recommend discontinuation of ICIs upon suspicion of myocarditis or pericarditis and initiation of high‐dose corticosteroids, with procedural interventions like pericardiocentesis in cases of cardiac tamponade. 38 For myocarditis, early cardiology consultation and continuous telemetry monitoring are essential due to the risk of life‐threatening arrhythmias. Diagnostic evaluation should include an ECG, troponin measurements, echocardiogram, and potentially cardiac magnetic resonance imaging or endomyocardial biopsy if the diagnosis is uncertain. The 2025 European Society of Cardiology guidelines recommend diagnostic triage within 24 hours in patients with a high suspicion of ICI‐induced myocarditis with immediate cessation of ICI and administration of high‐dose corticosteroids. 39 For pericarditis, NSAIDs and colchicine may be used for symptom control, but corticosteroids remain the mainstay of treatment. These recommendations are based on the high morbidity and mortality associated with these conditions, emphasizing the need for early recognition and aggressive management. These recommendations underscore the importance of early recognition and aggressive management to improve patient outcomes and reduce the high morbidity and mortality associated with these cardiovascular toxicities. 38 Further research is necessary to help elucidate predictors of pericarditis in patients initiating ICIs to help clinicians mitigate these undesirable cardiovascular events.
Limitations
Our study strength is due to its use of the TriNetX database, which is a large conglomerate of health care organizations allowing the investigation of large real‐world cohorts. This allows for the identification and investigation of diagnoses that are relatively rare in the broader population, such as irAEs related to ICI use. However, due to its dependence on ICD‐10 codes, it does not include clinical or other confounding variables, despite propensity matching procedures. The dependence on ICD‐10 codes without confirmation against clinical guidelines could lead to diagnostic errors. Although neoplasm was one of the variables used for propensity matching, we did not match for each individual malignancy. Therefore, the outcomes could, theoretically, be influenced by the type of malignancy the patient was diagnosed with. There is a potential selection bias from including ICI‐treated patients with more aggressive cancers or greater disease severity, which may confound outcomes due to unmeasured variables like cancer stage or tumor burden. There could also be a potential selection bias, as Black patients were underrepresented in the ICI group with pericarditis, possibly due to differences in access to ICI therapy. However, there were no significant differences after propensity matching in the pericarditis group. Additionally, we reported only therapies at the time of acute pericarditis diagnosis, lacking data on follow‐up. There is a lack of data on disease recurrence and the number of documented flares. Furthermore, the study’s retrospective design prevents inference of causal associations. There is limited availability of data on prior or concurrent radiation therapy. Although propensity score matching improved covariate balance, including for age, small residual differences remained, which could reflect variance‐related artifacts rather than true imbalance but may also suggest residual confounding, a known limitation of observational studies. Although our study was adequately powered to detect differences in 1‐year and 5‐year mortality outcomes across all cohorts, the statistical power for 90‐day mortality was limited in the pericarditis (16.6%) and myopericarditis (40.2%) groups. This raises the possibility of a Type II error in short‐term outcomes. Conversely, the post hoc power for longer‐term mortality was 100% in several comparisons; however, such estimates, being based on observed effect sizes, may overstate true power. Thus, while our findings are statistically robust for long‐term mortality, caution is warranted in interpreting null results, especially in early outcomes or when generalizing more broadly. Nevertheless, our study is the largest to date to evaluate ICI‐associated pericarditis and the first to investigate these patients’ outcomes as compared with the general population with pericarditis.
CONCLUSIONS
In the largest cohort study to date of 2394 patients with ICI‐associated pericarditis, we demonstrated an increased risk of all‐cause mortality, up to ~2‐fold increased mortality at 5 years, tamponade incidence, and need for pericardiocentesis as compared with the general population with pericarditis. Further investigations are warranted to better evaluate predictors or worse outcomes and preventive measures to avoid these outcomes in an already sick population.
Sources of Funding
None.
Disclosures
Dr Klein reports research funding from Kiniksa Pharmaceuticals Ltd., Ventyx BioSciences, and Cardiol Therapeutics, and has served on scientific advisory boards from Kiniksa Pharmaceuticals Ltd., Ventyx BioSciences, and Cardiol Therapeutics.
Supporting information
Data S1 List of ICD‐10 codes used and associated diagnoses, variables adjusted for the multivariable Cox proportional hazard model.
Figures S1–S2
This article was sent to Tochukwu M. Okwuosa, DO, Associate Editor, for review by expert referees, editorial decision, and final disposition.
Supplemental Material is available at https://www.ahajournals.org/doi/suppl/10.1161/JAHA.125.041163
For Sources of Funding and Disclosures, see page 18.
References
- 1. Cancer immunotherapy using checkpoint blockade|Science. Accessed January 11, 2025. https://www.science.org/doi/10.1126/science.aar4060.
- 2. Frontiers|Enhancing anti‐tumor immune responses through combination therapies: epigenetic drugs and immune checkpoint inhibitors. Accessed January 11, 2025. https://www.frontiersin.org/journals/immunology/articles/10.3389/fimmu.2023.1308264/full. [DOI] [PMC free article] [PubMed]
- 3. Cardiovascular toxicities associated with immune checkpoint inhibitors: an observational, retrospective, pharmacovigilance study—PubMed. Accessed January 11, 2025. https://pubmed.ncbi.nlm.nih.gov/30442497/. [DOI] [PMC free article] [PubMed]
- 4. Ball S, Ghosh RK, Wongsaengsak S, Bandyopadhyay D, Ghosh GC, Aronow WS, Fonarow GC, Lenihan DJ, Bhatt DL. Cardiovascular toxicities of immune checkpoint inhibitors: JACC review topic of the week. J Am Coll Cardiol. 2019;74:1714–1727. doi: 10.1016/j.jacc.2019.07.079 [DOI] [PubMed] [Google Scholar]
- 5. Naqash AR, Moey MYY, Cherie Tan X‐W, Laharwal M, Hill V, Moka N, Finnigan S, Murray J, Johnson DB, Moslehi JJ, et al. Major adverse cardiac events with immune checkpoint inhibitors: a pooled analysis of trials sponsored by the National Cancer Institute‐Cancer Therapy Evaluation Program. J Clin Oncol. 2022;40:3439–3452. doi: 10.1200/JCO.22.00369 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Dolladille C, Akroun J, Morice P‐M, Dompmartin A, Ezine E, Sassier M, Da‐Silva A, Plane A‐F, Legallois D, L’Orphelin J‐M, et al. Cardiovascular immunotoxicities associated with immune checkpoint inhibitors: a safety meta‐analysis. Eur Heart J. 2021;42:4964–4977. [DOI] [PubMed] [Google Scholar]
- 7. Nesfeder J, Elsensohn AN, Thind M, Lennon J, Domsky S. Pericardial effusion with tamponade physiology induced by nivolumab. Int J Cardiol. 2016;222:613–614. doi: 10.1016/j.ijcard.2016.08.023 [DOI] [PubMed] [Google Scholar]
- 8. de Almeida DVP, Gomes JR, Haddad FJ, Buzaid AC. Immune‐mediated pericarditis with pericardial tamponade during nivolumab therapy. J Immunother. 2018;41:329–331. doi: 10.1097/CJI.0000000000000217 [DOI] [PubMed] [Google Scholar]
- 9. Addison D, Neilan TG, Barac A, Scherrer‐Crosbie M, Okwuosa TM, Plana JC, Reding KW, Taqueti VR, Yang EH, Zaha VG, et al. Cardiovascular imaging in contemporary cardio‐oncology: a scientific statement from the American Heart Association. Circulation. 2023;148:1271–1286. [DOI] [PubMed] [Google Scholar]
- 10. Basso C. Myocarditis. N Engl J Med. 2022;387:1488–1500. doi: 10.1056/NEJMra2114478 [DOI] [PubMed] [Google Scholar]
- 11. Writing Committee , Drazner MH, Bozkurt B, Cooper LT, Aggarwal NR, Basso C, Bhave NM, Caforio ALP, Ferreira VM, Heidecker B, et al. 2024 ACC expert consensus decision pathway on strategies and criteria for the diagnosis and management of myocarditis: a report of the American College of Cardiology Solution set Oversight Committee. J Am Coll Cardiol. 2025;85:391–431. [DOI] [PubMed] [Google Scholar]
- 12. Klein AL, Wang TKM, Cremer PC, Abbate A, Adler Y, Asher C, Brucato A, Chetrit M, Hoit B, Jellis CL, et al. Pericardial diseases: international position statement on new concepts and advances in multimodality cardiac imaging. JACC Cardiovasc Imaging. 2024;17:937–988. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Chiabrando JG, Bonaventura A, Vecchié A, Wohlford GF, Mauro AG, Jordan JH, Grizzard JD, Montecucco F, Berrocal DH, Brucato A, et al. Management of acute and recurrent pericarditis: JACC state‐of‐the‐art review. J Am Coll Cardiol. 2020;75:76–92. doi: 10.1016/j.jacc.2019.11.021 [DOI] [PubMed] [Google Scholar]
- 14. Eggermont AMM, Chiarion‐Sileni V, Grob J‐J, Dummer R, Wolchok JD, Schmidt H, Hamid O, Robert C, Ascierto PA, Richards JM, et al. Prolonged survival in stage III melanoma with ipilimumab adjuvant therapy. N Engl J Med. 2016;375:1845–1855. doi: 10.1056/NEJMoa1611299 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Wang DY, Salem J‐E, Cohen JV, Chandra S, Menzer C, Ye F, Zhao S, Das S, Beckermann KE, Ha L, et al. Fatal toxic effects associated with immune checkpoint inhibitors: a systematic review and meta‐analysis. JAMA Oncol. 2018;4:1721–1728. doi: 10.1001/jamaoncol.2018.3923 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Postow MA, Sidlow R, Hellmann MD. Immune‐related adverse events associated with immune checkpoint blockade. N Engl J Med. 2018;378:158–168. doi: 10.1056/NEJMra1703481 [DOI] [PubMed] [Google Scholar]
- 17. Brahmer JR, Lacchetti C, Schneider BJ, Atkins MB, Brassil KJ, Caterino JM, Chau I, Ernstoff MS, Gardner JM, Ginex P, et al. Management of immune‐related adverse events in patients treated with immune checkpoint inhibitor therapy: American Society of Clinical Oncology clinical practice guideline. J Clin Oncol. 2018;36:1714–1768. doi: 10.1200/JCO.2017.77.6385 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Achim A, Liblik K, Gevaert S. Immune checkpoint inhibitors—the revolutionary cancer immunotherapy comes with a cardiotoxic price. Trends Cardiovasc Med. 2024;34:71–77. doi: 10.1016/j.tcm.2022.09.004 [DOI] [PubMed] [Google Scholar]
- 19. Reuben A, Petaccia de Macedo M, McQuade J, Joon A, Ren Z, Calderone T, Conner B, Wani K, Cooper ZA, Tawbi H, et al. Comparative immunologic characterization of autoimmune giant cell myocarditis with ipilimumab. Onco Targets Ther. 2017;6:e1361097. doi: 10.1080/2162402X.2017.1361097 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Matzen E, Bartels LE, Løgstrup B, Horskær S, Stilling C, Donskov F. Immune checkpoint inhibitor‐induced myocarditis in cancer patients: a case report and review of reported cases. Cardio‐Oncology. 2021;7:27. doi: 10.1186/s40959-021-00114-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Gong J, Neilan TG, Zlotoff DA. Mediators and mechanisms of immune checkpoint inhibitor‐associated myocarditis: insights from mouse and human. Immunol Rev. 2023;318:70–80. doi: 10.1111/imr.13240 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Axelrod ML, Meijers WC, Screever EM, Qin J, Carroll MG, Sun X, Tannous E, Zhang Y, Sugiura A, Taylor BC, et al. T cells specific for α‐myosin drive immunotherapy‐related myocarditis. Nature. 2022;611:818–826. doi: 10.1038/s41586-022-05432-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Won T, Kalinoski HM, Wood MK, Hughes DM, Jaime CM, Delgado P, Talor MV, Lasrado N, Reddy J, Čiháková D. Cardiac myosin‐specific autoimmune T cells contribute to immune‐checkpoint‐inhibitor‐associated myocarditis. Cell Rep. 2022;41:111611. doi: 10.1016/j.celrep.2022.111611 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Gergely TG, Drobni ZD, Sayour NV, Ferdinandy P, Varga ZV. Molecular fingerprints of cardiovascular toxicities of immune checkpoint inhibitors. Basic Res Cardiol. 2025;120:187–205. doi: 10.1007/s00395-024-01068-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Rubio‐Infante N, Ramírez‐Flores YA, Castillo EC, Lozano O, García‐Rivas G, Torre‐Amione G. A systematic review of the mechanisms involved in immune checkpoint inhibitors cardiotoxicity and challenges to improve clinical safety. Front Cell Dev Biol. 2022;10:851032. doi: 10.3389/fcell.2022.851032 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Palaskas N, Lopez‐Mattei J, Durand JB, Iliescu C, Deswal A. Immune checkpoint inhibitor myocarditis: pathophysiological characteristics, diagnosis, and treatment. J Am Heart Assoc. 2020;9:e013757. doi: 10.1161/JAHA.119.013757 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Sharma A, Alexander G, Chu JH, Markopoulos A, Maloul G, Ayub MT, Fidler MJ, Okwuosa TM. Immune checkpoint inhibitors and cardiotoxicity: a comparative meta‐analysis of observational studies and randomized controlled trials. J Am Heart Assoc. 2024;13:e032620. doi: 10.1161/JAHA.123.032620 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Mudra SE, Rayes DL, Agrawal A, Kumar AK, Li JZ, Njus M, McGowan K, Kalam KA, Charalampous C, Schleicher M, et al. Immune checkpoint inhibitors and pericardial disease: a systematic review. Cardio‐Oncology. 2024;10:29. doi: 10.1186/s40959-024-00234-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Qin Y, Zhang T, Du Z, Chen S, Li Y, Lv Y, Du X, Hu Y, Liu Z. Prognosis of immune checkpoint inhibitor‐related myocarditis: retrospective experience of a single institution. Int Immunopharmacol. 2024;136:112385. doi: 10.1016/j.intimp.2024.112385 [DOI] [PubMed] [Google Scholar]
- 30. Nielsen DL, Juhl CB, Nielsen OH, Chen IM, Herrmann J. Immune checkpoint inhibitor‐induced cardiotoxicity: a systematic review and meta‐analysis. JAMA Oncol. 2024;10:1390–1399. doi: 10.1001/jamaoncol.2024.3065 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Salem J‐E, Manouchehri A, Moey M, Lebrun‐Vignes B, Bastarache L, Pariente A, Gobert A, Spano J‐P, Balko JM, Bonaca MP, et al. Cardiovascular toxicities associated with immune checkpoint inhibitors: an observational, retrospective, pharmacovigilance study. Lancet Oncol. 2018;19:1579–1589. doi: 10.1016/S1470-2045(18)30608-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Kwan JM, Shen M, Akhlaghi N, Hu J‐R, Mora R, Cross JL, Jiang M, Mankbadi M, Wang P, Zaman S, et al. Adverse cardiovascular events and cardiac imaging findings in patients on immune checkpoint inhibitors. PLoS One. 2024;19:e0314555. doi: 10.1371/journal.pone.0314555 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Gong J, Drobni ZD, Zafar A, Quinaglia T, Hartmann S, Gilman HK, Raghu VK, Gongora C, Sise ME, Alvi RM, et al. Pericardial disease in patients treated with immune checkpoint inhibitors. J Immunother Cancer. 2021;9:e002771. doi: 10.1136/jitc-2021-002771 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Ma Z, Pei J, Sun X, Liu L, Lu W, Guo Q, Lyu J, Liu Y, Zhang Y, Zhao Z. Pericardial toxicities associated with immune checkpoint inhibitors: a pharmacovigilance analysis of the FDA Adverse Event Reporting System (FAERS) database. Front Pharmacol. 2021;12:12. doi: 10.3389/fphar.2021.663088 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Tang M, Dang P, Liu T, Yang K, Wang Y, Tse G, Liu H, Liu Y, Chan JSK, Liu C, et al. Risk factors and outcomes of pericardial effusion in cancer patients receiving PD‐1 inhibitors. Int J Cardiol. 2024;407:132029. doi: 10.1016/j.ijcard.2024.132029 [DOI] [PubMed] [Google Scholar]
- 36. Inno A, Maurea N, Metro G, Carbone A, Russo A, Gori S. Immune checkpoint inhibitors‐associated pericardial disease: a systematic review of case reports. Cancer Immunol Immunother. 2021;70:3041–3053. doi: 10.1007/s00262-021-02938-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Palaskas N, Morgan J, Daigle T, Banchs J, Durand J‐B, Hong D, Naing A, Le H, Hassan SA, Karimzad K, et al. Targeted cancer therapies with pericardial effusions requiring pericardiocentesis focusing on immune checkpoint inhibitors. Am J Cardiol. 2019;123:1351–1357. doi: 10.1016/j.amjcard.2019.01.013 [DOI] [PubMed] [Google Scholar]
- 38. Schneider BJ, Naidoo J, Santomasso BD, Lacchetti C, Adkins S, Anadkat M, Atkins MB, Brassil KJ, Caterino JM, Chau I, et al. Management of immune‐related adverse events in patients treated with immune checkpoint inhibitor therapy: ASCO guideline update. JCO. 2021;39:4073–4126. doi: 10.1200/JCO.21.01440 [DOI] [PubMed] [Google Scholar]
- 39. Schulz‐Menger J, Collini V, Gröschel J, Adler Y, Brucato A, Christian V, Ferreira VM, Gandjbakhch E, Heidecker B, Kerneis M, et al. 2025 ESC guidelines for the management of myocarditis and pericarditis: developed by the task force for the management of myocarditis and pericarditis of the European Society of Cardiology (ESC) Endorsed by the Association for European Paediatric and Congenital Cardiology (AEPC) and the European Association for Cardio‐Thoracic Surgery (EACTS). Eur Heart J. 2025;46:3952–4041. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data S1 List of ICD‐10 codes used and associated diagnoses, variables adjusted for the multivariable Cox proportional hazard model.
Figures S1–S2