A predictive model for low-dose rituximab response in anti-acetylcholine receptor antibody-positive myasthenia gravis: establishment and validation.

Background: Rituximab (RTX) represents an established therapeutic option for myasthenia gravis refractory to conventional immunotherapy. However, its heterogeneous efficacy across populations creates a challenge in clinical application, necessitating the development of robust predictive models to better identify patients likely to derive meaningful benefit from this treatment. Methods: This study retrospectively reviewed the patients who visited the research center and received RTX treatment between August 2019 and January 2025. The primary outcome measure was minimal symptom expression (MSE). A predictive model was constructed using logistic regression and presented in a nomogram. The performance of the model was evaluated by calculating the area under the receiver operating characteristic curve (AUC), with internal validation performed via the bootstrap method. The study population was divided into high- and low- probability groups according to Youden index. Calibration curves with 1000 replications bootstrap resampling were plotted to visualize the calibration of the nomogram. Decision curve analyzes (DCA) with 1000 replications bootstrap resampling were per formed to evaluate the clinical usefulness of the model. Results: A total of 117 patients were included. Five variables were screened including new-onset MG, baseline MG-ADL score, high CD19 Conclusion: This nomogram helps identify AChR-MG patients likely to attain MSE 6 months after low-dose rituximab, thereby supporting personalized therapeutic decision-making.