Henry Kaminski, MD, Department of Neurology, and Linda Kusner, PhD, Department of Pharmacology & Physiology at George Washington University, discuss predicting response outcomes with proteomic and machine learning analyses in treating patients with myasthenia gravis (MG).
MG is a chronic autoimmune neuromuscular disease characterized by weakness of the skeletal muscles. It mostly develops in adults but children can also develop this rare condition. Common symptoms include weakness of the muscles that control the eye and eyelid, facial expressions, chewing, talking, and swallowing. The condition results from a defect in the transmission of nerve impulses to muscles usually due to the presence of antibodies against the acetylcholine receptor. The exact reason this occurs is not known.
A study published in Springer Nature analyzed serum proteomes collected at baseline from participants in a phase 3 randomized trial comparing thymectomy plus prednisone versus prednisone alone, along with matched controls using liquid chromatography–mass spectrometry.
Due to the variable treatment responses present in patients with MG, there is a need for biomarkers to guide therapeutic decision making. Proteomic profiling, coupled with machine learning, was utilized to offer a hypothesis-free approach to identify multi-protein signatures associated with treatment response.
Results showed that baseline serum proteomes distinguished MG from controls, with pathway enrichment implicating complement activation, immunoglobulin production, and T-cell receptor signaling. Distinct protein panels predicted 6-month clinical improvement within each treatment arm. In the thymectomy-plus-prednisone group, models captured non-linear relationships of predictive proteins in contrast with the predominant additive patterns observed in the prednisone-alone group. Predictive proteins were enriched for T-cell signaling and leukocyte trafficking functions, providing insight into treatment-specific biology.
While Drs. Kaminski and Kusner explain that there is still much work to be done, this method captured core disease characteristics of MG and predicted short-term clinical response in a treatment-specific manner. These findings could enable biomarker-guided selection of treatment, refine risk stratification, and furnish mechanistic readouts for future MG trials and clinical care.
To learn more about gMG and other rare neurological conditions, visit https://checkrare.com/diseases/neurology-nervous-system-diseases/