Case Report: Pembrolizumab–induced myasthenia gravis with isolated anti-titin antibody positivity after thymectomy for thymoma

1 Introduction

Immune checkpoint inhibitors (ICIs) are a class of drugs that enhance the antitumor activity of T cells, promoting cancer cell elimination (Postow et al., 2015). Among these, pembrolizumab, a programmed cell death 1 (PD-1) inhibitor, is a key ICI; however, as their use has increased, various immune-related adverse events (irAEs) have been reported (Postow et al., 2018). These irAEs can affect various organs, including the skin, gastrointestinal tract, respiratory system, neuromuscular system, and cardiovascular system (Yin et al., 2023). Neurological complications include encephalitis, seizures, myelopathy, polyneuropathy, myasthenia gravis, and muscle weakness. Immune-related myasthenia gravis (irMG) is a rare condition that manifests with a more rapid and severe course than classic myasthenia gravis, and is associated with a high mortality rate (Huang et al., 2020).

Approximately 20% of patients diagnosed with thymoma have myasthenia gravis. When myasthenia gravis presents initially, thymoma is frequently diagnosed during evaluation (Mao et al., 2012). Furthermore, there are rare reports of myasthenia gravis occurring years after thymectomy for thymoma (Kondo and Monden, 2005). Herein, along with a literature review, we present a case of new-onset myasthenia gravis that was characterized by anti-titin antibody positivity alone and developed following pembrolizumab administration in a patient who had previously undergone a thymectomy for thymoma 12 years earlier and had remained asymptomatic in the interim.

2 Case description

A 65-year-old man presented with dysarthria, dysphagia, and dyspnea that progressively developed over the preceding 2 months. The timeline of the patient’s clinical course, including symptom onset, treatment and follow-up, is summarized in Figure 1. The patient underwent total thyroidectomy for thyroid cancer and thymectomy for thymoma 12 years before presentation. A review of the available medical records revealed no symptoms suggestive of myasthenia gravis at the time of thymectomy, and the patient reported no myasthenic symptoms during the subsequent 12-year period. He was diagnosed with bladder cancer 1 year earlier, underwent a radical cystectomy and had been receiving pembrolizumab every 3 weeks in an adjuvant immunotherapy regimen for the last 3 months. Four weeks after initiation of pembrolizumab, and approximately 1 week after the second dose, diplopia, dysarthria, dysphagia, and dyspnea developed. The symptoms gradually worsened, and interfered with activities of daily living, including eating and sleeping. No significant diurnal symptom variation was observed. Upon admission, the patient had normal vital signs and was alert. A neurological examination revealed external ophthalmoplegia, bilateral ptosis, dysarthria, and dysphagia. There was no significant weakness in the extremities, and deep tendon reflexes were normal. Arterial blood gas analysis revealed normal pCO₂ levels; however, the pO₂ level was 68.6 mmHg (normal range: 83–108 mmHg), with an oxygen saturation of 93.2% (normal range: 95%–98%). Routine blood tests, including a complete blood count and serum creatinine, were unremarkable. Biomarkers of muscle and cardiac injury were within the normal ranges: creatine kinase, 185 U/L (reference range, 37–212 U/L); creatine kinase-MB, 2.31 ng/mL (0–6.22 ng/mL); myoglobin, 33.56 ng/mL (28–72 ng/mL); and troponin I, <0.01 ng/mL. Aspartate aminotransferase and alanine aminotransferase levels were also normal at 31 U/L (5–40 U/L) and 28 U/L (5–40 U/L), respectively. The electrocardiogram was normal, and transthoracic echocardiography demonstrated normal left ventricular size and systolic function (ejection fraction, 64%). A chest CT scan indicated bilateral, mild lobar pneumonia. Pulmonary function tests demonstrated a substantial decline in forced vital capacity (FVC, 40% of predicted) and forced expiratory volume in 1 s (FEV1, 44% of predicted), indicating an extrathoracic restrictive pattern. The cerebrospinal fluid analysis showed a white blood cell count of 4/mm³, protein level of 79.8 mg/dL, and glucose level of 76 mg/dL (serum glucose: 124 mg/dL). There were no specific findings on the brain magnetic resonance imaging. The ice-cube test was negative, and the repetitive nerve stimulation test revealed a mild decremental response of 14% at low-frequency stimulation only in the trapezius muscle. No significant decremental response was observed in the orbicularis oculi or abductor digiti minimi muscles. The neostigmine test revealed a modest improvement in bilateral ptosis. Nerve conduction studies indicated normal conduction velocities for sensory and motor nerves and normal amplitudes for compound motor action potentials, without abnormal spontaneous activity on electromyography. On the second day of hospitalization, the patient’s dyspnea worsened, requiring transfer to the intensive care unit and initiation of noninvasive positive-pressure ventilation. Considering pembrolizumab-induced myasthenia gravis, the patient was prescribed a 3-day course of 500 mg methylprednisolone. The results of the anti-acetylcholine receptor (AChR), anti-muscle-specific kinase (MuSK) antibody, and anti-ganglioside antibody tests were all negative. The anti-titin antibody test in the paraneoplastic syndrome panel was positive. Despite the administration of high-dose steroid therapy, the symptoms did not improve, and immunoglobulin therapy was initiated. On Day 3 of immunoglobulin administration (2 g/kg over 5 days), the patient demonstrated clinical improvement, and was subsequently discharged after the gradual improvement of symptoms with continuation of oral prednisolone therapy. However, the steroid therapy was discontinued owing to bleeding from a duodenal ulcer, which was considered steroid-related. The patient has since remained on monthly intravenous immunoglobulin (IVIg) as maintenance therapy. At the 5-month follow-up after the initial IVIg treatment, the Myasthenia gravis-Activities of Daily Living score was 7, with sustained overall clinical improvement.

3 Discussion

ICI-related myasthenia gravis represents a rare but potentially lethal irAE that frequently manifests as an overlap syndrome with myositis or myocarditis (Safa et al., 2019). In the current literature, anti-striational antibodies are recognized as a heterogeneous group of autoantibodies that target various muscle proteins, including titin, ryanodine receptors, and Kv1.4 (Yamamoto et al., 2001). Among them, anti-titin antibodies are frequently observed in thymoma-associated myasthenia gravis, and their presence is likely associated with severe disease progression and muscle involvement (Meriggioli and Sanders, 2012). However, in this case, despite the presence of anti-titin antibodies, no elevated CK levels or clinical evidence indicative of myocarditis were observed. This finding suggests that anti-titin antibody positivity alone does not necessarily indicate systemic muscle involvement and highlights the clinical heterogeneity of ICI-related myasthenia gravis.

Another noteworthy aspect of this case is that myasthenia gravis developed 12 years after thymectomy following a prolonged asymptomatic period. A previous report described pembrolizumab-induced myasthenia gravis occurring 15 years after thymectomy; however, only anti-striational antibodies were reported and anti-titin antibody status was not specified (Lara et al., 2019). In contrast, our patient demonstrated isolated anti-titin antibody positivity without detectable anti-AChR or MuSK antibodies. These findings suggest that persistent thymoma-associated immune dysregulation may remain clinically silent for many years and become unmasked following immune checkpoint inhibition.

To better characterize this rare phenotype, we systematically reviewed published cases of ICI-related myasthenia gravis with anti-striational antibody positivity. Clinical characteristics, antibody profiles, and thymoma history were summarized (Table 1). In the majority of cases, there was no history of thymoma, and clinical details were limited in several cases. Only a single case exhibited a history of thymoma analogous to the present case. Moreover, among cases that tested positive for anti-striational antibodies, the specific antibody subtypes, including anti-titin antibodies, were rarely clearly described. In particular, there is no report of a case with anti-titin antibodies and a history of thymoma. Therefore, our case expands the immunological spectrum of ICI-related myasthenia gravis and provides additional insight into the relationship between thymoma-associated autoimmunity and immune checkpoint inhibition.

TABLE 1. Summary of previously reported cases of ICI-induced myasthenia gravis with anti-striational antibody positivity.

Ab, antibody; AChR, acetylcholine receptor; CS, corticosteroid; F, female; ICI, immune checkpoint inhibitor; IVIg, intravenous immunoglobulin; IVMP, intravenous methylprednisolone; M, male; MMF, mycophenolate mofetil; NSCLC, non-small cell lung cancer; PSL, prednisolone; SCLC, small cell lung cancer.

Notably, this patient exhibited no neurological indications of myasthenia gravis prior to the initiation of ICI treatment. This finding suggests the potential for ICI-induced new autoimmune disease as well as the possibility that an underlying thymoma-associated autoimmune predisposition that remained quiescent for an extended period was subsequently triggered by immune checkpoint inhibition. Although causality cannot be established from a single case, our findings support the hypothesis that autoreactive T cells or long-lived plasma cells persisting after thymectomy may remain dormant and become reactivated following PD-1 blockade (Huang et al., 2020; Safa et al., 2019; Berrih-Aknin and Le Panse, 2014).

This case also provides important therapeutic insights. Although high-dose corticosteroid therapy did not result in meaningful clinical improvement, the patient showed gradual improvement in symptoms following IVIg administration. Although conclusions regarding treatment efficacy cannot be drawn from a single case, this experience supports previous reports suggesting that early use of IVIg or plasma exchange may be beneficial in selected patients with ICI-related myasthenia gravis (Safa et al., 2019). Furthermore, steroid-related duodenal ulcer bleeding prompted discontinuation of corticosteroids and modification of the treatment strategy, underscoring the importance of an individualized therapeutic approach based on the patient’s overall clinical condition and treatment-related complications.

As a single case report, this study cannot establish a direct causal relationship between isolated anti-titin antibody positivity and the clinical phenotype of ICI-induced myasthenia gravis. In addition, anti-titin antibody testing was qualitative, precluding assessment of antibody titers or their relationship with disease activity. Nevertheless, this case highlights a rare immunological phenotype of ICI-induced myasthenia gravis, raising the possibility that long-standing thymoma-associated autoimmunity may be unmasked by ICI therapy. These findings underscore the importance of careful neurological monitoring in patients with a history of thymoma receiving ICIs. Further studies are needed to clarify the clinical significance of anti-titin antibodies and the utility of pre-treatment antibody screening in this population.

4 Patient perspective

The patient reported that the neurological symptoms caused substantial distress and had a greater impact on his daily life than he had anticipated from his cancer treatment. Following clinical improvement, he expressed satisfaction with his recovery despite the need for continued hospital visits for maintenance treatment.

Generative AI statement

The author(s) declared that generative AI was used in the creation of this manuscript. The authors used ChatGPT (OpenAI) solely for grammar checking and language editing.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Ahmed Sheriff, Charité University Medicine Berlin, Germany

Reviewed by: Vikrant Rai, Western University of Health Sciences, United States

Edison K Miyawaki, Mass General Hospital Brigham, United States