Immune checkpoint inhibitor cardiomyotoxicity: a case series of five patients and development of a clinical management algorithm

Abstract

Immune checkpoint inhibitor (ICI) therapies have revolutionised the management of a range of malignancies. However, the syndrome of ICI induced cardiomyotoxicity, also reported as triple “M” syndrome (myasthenia gravis, myositis and myocarditis) is an unfortunate severe consequence in up to 1%–2% of patients receiving ICI, with an associated mortality of up to 50%. Based on international evidence and local experience, we have developed a clinical pathway in our institution as a guide to the early identification and targeted management of this condition.

Immune checkpoint inhibitors (ICIs) have become key therapeutics for a range of malignancies, both as first‐ and second‐line treatments as well as salvage therapies. Currently, checkpoint inhibitors in clinical use include anti‐programmed cell death protein 1 (PD‐1), anti‐programmed cell death ligand 1 (PD‐L1), anti‐cytotoxic T lymphocyte associated protein 4 (CTLA‐4) and anti‐lymphocyte activation gene 3 (LAG‐3), either as monotherapy or in combination. The iatrogenic autoimmune toxicities associated with ICI are known as immune‐related adverse events (irAEs) and may affect various organs (e.g. gastrointestinal, pulmonary) and have variable severity (mild to life threatening). While the exact pathophysiological mechanism of these adverse events remains unclear, auto‐reactive T cells and macrophages appear to play a role. ¹ , ² , ³

Recently, ICI‐associated cardiomyotoxicity, also reported as triple “M” (3 M) syndrome, reflecting the following clinical trial of disease presentation (myasthenia gravis pattern of weakness, myositis and myocarditis), was recognised as the most life‐threatening irAE. ¹ , ² , ³ , ⁴ Gougis et al. reported the onset for ICI irAEs ranged from 31 to 273 days (median 40 days), with the shortest onset for cardiomyotoxicities, including myocarditis. ² However, early identification and targeted immunosuppression do improve survival. ¹ , ²

Here we present a single‐centre case series of patients with ICI‐induced cardiomyotoxicity from a large quaternary metropolitan hospital in Sydney, Australia. Following Patient 1's outcome (irreversible respiratory muscle failure secondary to ICI cardiomyotoxicity), we formed a working group to develop a practically useful clinical pathway for the rapid identification of ICI‐induced cardiomyotoxicity, with the aim of allowing early identification and prompt, effective management of this condition. While Patient 1 in our series pre‐dated our clinical pathway guidelines, Patients 2–4 were managed while the clinical pathway was in development and Patient 5 after the guidelines had been locally disseminated. The baseline characteristics, management and outcome of those patients are outlined in Table 1. Of note, the gold standard for diagnosis of myocarditis is endomyocardial biopsy, with cardiac magnetic resonance imaging (cMRI) the best alternative non‐invasive modality. ⁵ However, given the rapid progression and unstable nature of our cases, endomyocardial biopsies were not performed, and only two cases had cMRI performed. Based on the definition of myocarditis found in previously published literature, cases 3 and 4 would meet the criteria for definitive myocarditis, while other cases would fit the definition of possible myocarditis. ⁵ Alternative diagnosis for example acute myocardial infarction from plaque rupture was considered in the diagnostic work‐up of all cases, with ICI cardiomyotoxicity deemed the most likely diagnosis for all cases.

Table 1.

Baseline characteristics of patients with ICI‐related cardiomyotoxicity at our hospital site.

Case no.	Age (years)	Sex	Ethnicity	Primary malignancy	Immunotherapy name	Onset of 3 M since last ICI	Presenting symptoms	Troponin (Trop) on presentation	Creatine kinase (CK) on presentation	NT Pro BNP on presentation	AchR and Musk antibody (*Titin antibody not tested)	Other relevant cardiac investigations
1	81	M	European	Metastatic melanoma on b/g of myasthenia gravis	Pembrolizumab (PD‐1 inhibitor)	3 weeks	Dyspnoea and myasthenia gravis pattern weakness with ocular and bulbar muscle involvement	hsTrop I 458 ng/L (≤50)	6345 U/L (40–200 U/L)	1111 ng/L (≤450)	AChR positive. AChR 3.130 nmol/L (positive >0.4). MuSK antibody negative	ECG: Sinus rhythm, left axis deviation. Echo: LVEF 43% No cMRI or relevant biopsy
2	59	F	European	Stage 4 cervical cancer	PD‐1/IL‐2 bispecific antibody fusion protein (clinical trial)	4 weeks	Dyspnoea and myalgia	Trop T 156 ng/L (≤14)	CK 1923 U/L (<150 U/L)	604 ng/L (≤125)	Both negative	ECG: Sinus rhythm, biphasic T wave V2 Echo: LVEF: 75% cMRI(day 11): patchy sub‐epicardial LGE in the basal septum No cardiac biopsy
3	54	F	Asian	Metastatic adenoid cystic carcinoma	PD‐1/IL‐2 bispecific antibody fusion protein (clinical trial)	8 weeks	Lower limb weakness	Trop T 524 ng/L (≤14)	CK 226 U/L (<150 U/L)	Not available. Day 30: 68 ng/L (≤125)	Not available	ECG: normal sinus rhythm Echo: LVEF 65% cMRI: Extensive oedema of the basal to mid inferior, inferoseptal and septal LV myocardium. Early and late gadolinium enhancements in an epicardial pattern No cardiac biopsy
4	83	M	European	Metastatic melanoma	Nivolumab (PD‐1 inhibitor) and relatlimab (LAG‐3 inhibitor)	4 weeks	Dyspnoea and myalgia	Trop T 1221 ng/L(≤14)	CK 142 U/L (<150 U/L)	4267 ng/L (≤125)	Both negative	ECG: Atrial flutter with ventricular pacing Echo: LVEF 47% No cMRI or biopsy
5	78	M	European	Advanced mesothelioma	Ipilimumab (CTLA‐4 inhibitor) and nivolumab (PD‐1 inhibitor)	1 week	Complete heart block and myasthenia gravis pattern weakness with ocular muscle involvement	3640 ng/L (0–14)	18 796 U/L (40–200 U/L)	4460 ng/L (0–450)	Both negative	ECG: complete heart block. LVEF 35%–40% No cMRI or biopsy

AChRc, acetylcholine receptor; CTLA‐4, cytotoxic T lymphocyte associated protein 4; Hs, high sensitivity; ICI, immune checkpoint inhibitor; LAG‐3, lymphocyte activation gene 3; MuSK, muscle‐specific kinase; PD‐L1, programmed cell death ligand 1.

Patients 2 and 3 outlined below involved clinical trial participants on PD‐1/IL‐2 bispecific antibody fusion protein leading to activation of PD‐1 CD25‐positive tumour‐specific T cells, with a proposed protective effect of increased peripheral regulatory T‐cell activation. ⁶

Discussion

The key management strategy for ICI cardiomyotoxicity is early identification. For our clinical pathway, we have outlined red flag symptoms, examination and relevant investigations (Tables S1 and S2). After identification of a suspected case of ICI‐induced cardiomyotoxicity, an urgent multidisciplinary meeting should be initiated involving medical oncology, immunology, cardiology, thoracic medicine, neurology and critical care. At this meeting, the team should decide whether the patient's symptoms are suggestive of ICI cardiomyotoxicity; if so, a stratification of disease severity should be made. An elevated troponin should not constitute a diagnosis of myocarditis unless significant change from baseline troponin, and other alternative diagnosis, for example, acute myocardial infarction from plaque rupture are ruled out. ⁷ A scoring system may be helpful in this regard, as outlined by Power et al. (EHJ 2025). ⁸

If the consensus decision is that of severe ICI cardiomyotoxicity, then urgent initiation of treatment as outlined in the flow chart is highly recommended. Given our clinical pathway, expedited approval and access to abatacept and ruxolitinib was made possible through the hospital drug committee. The management of non‐severe cases is also outlined in the flow chart. In haemodynamically stable patients able to tolerate oral therapy, high‐dose oral prednisone/prednisolone is an appropriate alternative to IV corticosteroids because oral absorption is high, and systemic exposure is comparable at bioequivalent dosing. IV (including pulse methylprednisolone) can be reserved for severe presentations. ⁸ Emerging data are also now identifying a subset of asymptomatic, non‐severe definite cases of ICI cardiomyotoxicity potentially requiring no immunosuppression but just withholding further doses of ICI and monitoring closely ⁸ (Fig. 1).

Figure 1.

Treatment flow chart for ICI cardiomyotoxicity. BD, twice daily; ICI, immune checkpoint inhibitor; IV, intravenous; PO, orally. ¹ , ² , ⁴ , ⁸ , ⁹

Due to the variable nature of this clinical syndrome, there are no standardised practices on monitoring treatment response. Treatment response should be observed within a few days of initiating treatment, especially if early remission induction therapy with abatacept and ruxolitinib is administered. It is important that there be no evidence of clinical progression (e.g. worsening respiratory muscle weakness or worsening cardiac failure). Daily monitoring of creatine kinase (CK) and troponins (particularly troponin‐T) is recommended. Periodic monitoring of N‐terminal pro‐B‐type natriuretic peptide (NT‐BNP) is recommended at baseline, at 2 weeks, with further monitoring depending on progress. Cardiac imaging including transthoracic echocardiogram and cardiac MRI are suggested.

All patients should be closely monitored for progression to respiratory muscle involvement, with spirometry, +/− ABG and respiratory physician input, and, where relevant, early intensivist assessment. Pre‐existing myasthenia gravis, +/− the presence of AChR antibodies are key risk factors for rapidly progressive respiratory failure. This was relevant to Patient 1 in this series who had a history of ocular myasthenia gravis 3 years prior to the commencement of pembrolizumab without ongoing symptoms or treatment and positive AChR antibodies.

Once clinical remission is achieved, ongoing care typically involves outpatient follow‐up with de‐escalation of therapies. There are no standard treatment duration guidelines for most of the therapies used. However, corticosteroids are typically tapered over a few weeks (generally a 30% dose decrease every 10–20 days), with ruxolitinib typically used for 4–6 weeks in total (typically tapering ruxolitinib from 30 to 10 mg/day as intermediate dose). Further details of treatment and outcome in this case series are outlined in Table 2.

Table 2.

Relevant treatment and outcome patients with ICI‐related cardiomyotoxicity at our hospital site.

Case	Steroid	Ruxolitinib	Abatacept	Other treatment	2 weeks’ ICI cardiomyotoxicity outcome	1 month ICI cardiomyotoxicity outcome	6 months ICI cardiomyotoxicity outcome	Cause of death
1	Yes (day 1)	Yes (day 11)	No	PLEX, RTX, IVIG	Intubated complicated by HSV‐1 infection	GI bleeding with palliation on day 35	Not applicable	Gastrointestinal bleeding and ICI cardiomyotoxicity
2	Yes (day 1)	Yes (day 2)	Yes (day 7)	Nil	Improved muscle strength, dyspnoea and cardiac markers	Discharged home on day 21. Improved symptoms	Off corticosteroid (3 months treatment in total). Ruxolitinib ceased (6 months after treatment). Residual proximal muscle weakness	Not applicable
3	Yes (day 1)	Yes (day 6)	Yes (day 9)	MMF	Improved muscle strength and cardiac markers	Discharged home on day 21. Improved symptoms. Normal cardiac and muscle enzyme	Off corticosteroid (3 months of treatment). Off ruxolitinib and MMF in (2 months treatment). No residual symptoms	Not applicable
4	Yes (day 1)	Yes (day 1)	Yes (day 1)	IV heparin	Improved dyspnoea and muscle symptom. Improved cardiac markers. Retroperitoneal bleeding complications	Colonic perforation likely secondary to diverticular disease (day 21). Palliated Day 24 and passed away Day 25	Not applicable	Colonic perforation
5	Yes (day 1)	Yes (day 1)	Yes (day 1)	Permanent pacemaker insertion and pyridostigmine	Improved dyspnoea and ocular myasthenia gravis symptoms and Trop/CK	Discharged to private rehab with improved symptoms	Not applicable	Not applicable

Note: Day next to the medication is day of initiation post ICI identification. We developed the clinical pathway over a period of 6 months; case 5 used the fully formalised clinical pathway, whereas cases 2–4 used the clinical pathway in development. IVIG, intravenous immunoglobulin; MMF, mycophenolate; PLEX, plasma exchange; RTX, rituximab.

There is a paucity of evidence in the literature to guide maintenance therapy. There are ongoing clinical trials evaluating maintenance therapy by Salem et al. ⁴ The objective after 3 weeks of intensive initial treatment is to wean patients off all immunosuppression. In their trial, a weaning reassessment visit is conducted every 7 to 14 days, starting at day 21. ⁴ At each clinic visit, the need for further abatacept is assessed. For the patients in our cohort, no further abatacept was required after the initial three doses (Table 2). Given the high morbidity and mortality associated with ICI‐induced cardiomyotoxicity, further doses of ICI are generally contraindicated in most current guidelines, even after complete recovery. Though this is still an area of unmet need, particularly for patients who derived benefit from their ICI therapy, further research is being conducted examining the risks and benefits of rechallenging with ICI in certain subsets of patients.

In the ICI myocarditis cohort study by Salem et al., early initiation of intensive combination therapy with corticosteroids, abatacept and ruxolitinib led to improved survival and outcomes. The 100‐day overall survival was 77% in the early initiation cohort (n = 23/30), in comparison to the prior standard of care regimen (without abatacept and ruxolitinib) where the 100‐day overall survival was 30% (n = 3/10). ¹ For our cases, 6‐month overall survival was 50% of the cases with 6‐month follow‐up (4/5). The gender distribution in our cohort was similar to that of the cohort study; however, the median age of our cohort was 78 years old, which is older than the median age (72 years) in the study by Salem et al. ⁶ It is unclear whether a slightly older cohort might have influenced the outcome of our cases. The size of our cohort was also significantly smaller than that of Salem et al.'s.

ICI cardiomyotoxicity is a severe irAE with high associated mortality of up to 50%; however, prompt identification and intervention, with a multi‐disciplinary team approach, is crucial in managing this serious irAE. The management strategies of this condition differ from those of other causes of myocarditis, with targeted therapies such as abatacept and ruxolitinib improving its outcome. Further studies are ongoing to optimise treatment regimens according to stratified disease severity.

Supporting information

Table S1. Identification of ICI‐induced cardiomyotoxicity/triple “M” syndrome.

Table S2. Relevant investigations (bold indicating urgent).

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.