Azathioprine Toxicity in Patients With Myasthenia Gravis: A Single Center Experience

INTRODUCTION/AIMS: Azathioprine (AZA) is a first-line immunosuppressive agent for myasthenia gravis (MG), yet most toxicity data for AZA are from gastrointestinal literature. The aim of the current study is to add a large MG patient cohort with a comprehensive evaluation of the adverse event profile of AZA and associations between patient demographics and drug reactions. METHODS: Patients were identified from the Duke Electronic Medical Record using EPIC's SlicerDicer tool. INCLUSION CRITERIA: MG diagnosis confirmed by antibody/electrodiagnostic testing, evaluation by a Duke neuromuscular physician, and AZA use between January 1, 2014 and January 7, 2023. Drug reaction was defined as an abnormal lab result or symptomatic side effect after AZA initiation. RESULTS: Among 145 patients, 96 (66%) developed a drug reaction; 63 (43.5%) required AZA dose reduction or discontinuation, and 23 (16%) had symptomatic side effects. Most reactions improved with dose reduction or observation. Leukopenia, macrocytosis, and hepatotoxicity were the most frequent. For every additional comorbidity, the odds of drug reaction increased by 30% (OR 1.3; 95% CI 1.12-1.51; p < 0.001). Flu-like reactions occurred in 10% of patients. Cancer incidence (4.1%) was rare and there were no cases of Pneumocystis jirovecii pneumonia despite rare use of trimethoprim-sulfamethoxazole prophylaxis. DISCUSSION: AZA drug reactions were common but infrequently severe or symptomatic. Higher comorbidity burden predicted drug reactions. Most laboratory abnormalities resolved without AZA discontinuation, supporting individualized monitoring and dose adjustment rather than automatic cessation. Lower doses should be considered for balancing efficacy with dose-dependent drug reactions.