Predictive Model for Rituximab Response in Myasthenia Gravis Patients

A retrospective study of 117 patients treated between August 2019 and January 2025, authored by Chenlu Hou. It establishes a predictive model for minimal symptom expression after low-dose rituximab treatment in anti-acetylcholine receptor antibody-positive myasthenia gravis. The model uses five clinical variables and was validated internally via bootstrap methods.

Use Cases

• Predicting patient response to low-dose rituximab based on clinical variables like baseline MG-ADL score and B lymphocyte proportion.
• Stratifying patients into high- and low-probability groups for achieving minimal symptom expression using a threshold probability.
• Validating logistic regression-based clinical prediction models via internal bootstrap validation methods.
• Supporting personalized therapeutic decision-making for myasthenia gravis treatment.

Strengths

• Model built on data from 117 patients.
• Includes five specific predictive variables identified: new-onset MG, baseline MG-ADL score, high CD19+/CD27+ B lymphocyte proportion, high-dose prednisone, and early immunotherapy initiation.
• Model performance quantified with an AUC of 0.777 and a defined threshold probability of 0.534.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Data may reflect temporal and institutional bias inherent to a single-center retrospective study.

Provenance

Source

figshare, author Chenlu Hou.

Collection Method

Retrospective review of patients who visited the research center and received RTX treatment.

Time Range

August 2019 to January 2025.

Freshness

Last updated 2026-06-03 04:53:34; freshness should be verified.