Evaluation of the predictive value of sCD200 and sCD200R for disease severity and relapse in myasthenia gravis.

BACKGROUND: Myasthenia gravis (MG) is an autoimmune disorder with fluctuating muscle weakness and abnormal cytokine profiles. Relapse remains a major challenge despite therapeutic advances. The CD200-CD200R signaling axis delivers potent inhibitory signals to maintain immune homeostasis, but the clinical significance of its soluble forms, soluble CD200 (sCD200) and soluble CD200 receptor (sCD200R), in MG remains unclear. METHODS: We enrolled healthy controls and MG patients at the acute phase or in remission. Plasma sCD200/sCD200R levels were compared among the three groups, and correlations with clinical data were assessed in initial-onset patients. Initial-onset patients were then followed to remission for paired comparisons of sCD200/sCD200R. Patients in remission were followed prospectively, clinical data and plasma sCD200/sCD200R levels were compared between relapsed and non-relapsed patients. Multivariate logistic regression identified independent risk factors for relapse, and a combined predictive model was constructed. RESULTS: sCD200 and sCD200R levels were significantly elevated in acute-phase MG, reduced in remission, and lowest in healthy controls. sCD200 and sCD200R were positively correlated with QMG scores, whereas sCD200R additionally showed a positive correlation with IL-10. Paired comparisons confirmed significant declines from acute to remission. Within the remission cohort, elevated sCD200 and sCD200R predicted increased relapse risk during follow-up. A combined model of sCD200, sCD200R, and IL-10 for predicting relapse achieved an AUC of 0.916, with good calibration and positive net benefit on decision curve analysis. CONCLUSIONS: sCD200 and sCD200R reflect MG activity and relapse risk. The combined model provides notable predictive performance for MG relapse, supporting its potential clinical utility.