Peripheral immune patterns enable robust cross-platform prediction of ALS onset and progression.

Area of Science:

• Neuroimmunology
• Systems immunology
• Biomarker discovery

Background:

• Amyotrophic lateral sclerosis (ALS) exhibits significant patient heterogeneity in progression rates.
• Lack of prognostic biomarkers hinders clinical decision-making and patient stratification for ALS therapies.

Purpose of the Study:

• To identify immune cell network disruptions associated with ALS progression.
• To develop predictive biomarkers for ALS disease status and progression rates.

Main Methods:

• Mass cytometry (CyTOF) profiling of 2.2 million immune cells from ALS patients and healthy controls.
• Analysis of immune cell type correlation patterns and network organization.
• Development of machine learning models for disease stratification.

Main Results:

• Immune cell coordination patterns, not individual cell frequencies, differentiate ALS progression.
• Observed a shift from B cell/basophil hubs in controls to neutrophil/T cell-dominated patterns in ALS.
• Machine learning models utilizing immune cell interactions outperformed those using cell frequencies for disease stratification.

Conclusions:

• ALS is characterized by a breakdown in immune coordination, offering new therapeutic targets.
• Central/effector memory CD4+ T cell interactions and plasmacytoid dendritic cell/regulatory T cell ratios are key discriminators.
• Findings suggest potential biomarkers and cell-type specific therapeutics for ALS and other neurodegenerative diseases.