A Validated Model to Predict Severe Weight Loss in Amyotrophic Lateral Sclerosis

Area of Science:

• Neurology
• Clinical Medicine
• Biostatistics

Background:

• Severe weight loss is a frequent and serious complication in amyotrophic lateral sclerosis (ALS).
• This weight loss is multifactorial and significantly correlates with reduced patient survival.
• Predicting severe weight loss is crucial for timely clinical management.

Purpose of the Study:

• To develop and validate a predictive model for severe weight loss in ALS patients.
• To identify key clinical predictors associated with severe weight loss.
• To provide a tool for stratifying ALS patients based on weight loss risk.

Main Methods:

• Utilized longitudinal weight data from over 6000 ALS patients across three distinct cohorts.
• Constructed an accelerated failure time model incorporating five clinical predictors: symptom duration, revised ALS Functional Rating Scale, site of onset, forced vital capacity, and age.
• Evaluated model performance and generalizability using internal-external cross-validation and random-effects meta-analysis.

Main Results:

• The predictive model achieved a concordance statistic of 0.71 (95% CI 0.63-0.79).
• Model calibration was robust, with a slope of 0.91 (0.69-1.13) and intercept of 0.05 (-0.11-0.21).
• Identified specific clinical factors significantly associated with the risk of severe weight loss in ALS.

Conclusions:

• The study successfully identified key clinical factors predicting severe weight loss in amyotrophic lateral sclerosis.
• The developed model demonstrates good performance and generalizability for risk prediction.
• This research lays the foundation for a clinical tool to stratify ALS patients by weight loss risk, potentially improving patient outcomes.