Modeling Variability in the Progression of Huntington's Disease A Novel Modeling Approach Applied to Structural...

Area of Science:

• Neuroscience
• Genetics
• Biostatistics

Background:

• Huntington's disease (HD) is a fatal neurodegenerative disorder.
• HD is caused by an expanded cytosine-adenine-guanine (CAG) triplet repeat in the huntingtin gene.
• Predicting HD progression is crucial for clinical management and therapeutic development.

Purpose of the Study:

• To introduce a novel, general class of disease progression models for Huntington's disease.
• To predict continuous markers of HD state using age and CAG repeat length.
• To incorporate various sources of variability into the models for improved accuracy.

Main Methods:

• Mixed-effects models were employed, fitting to structural imaging markers from the TRACK-HD database.
• Flexible regression splines were used to model the effects of age and CAG length.
• Random walk terms and a Kalman filter were utilized to model disease dynamics and estimate variances.

Main Results:

• The developed models effectively predict continuous markers of HD progression.
• The models account for age, CAG length, measurement error, population variability, and disease dynamics.
• The approach provides a flexible framework for analyzing HD progression.

Conclusions:

• The novel disease progression models offer a robust tool for predicting Huntington's disease trajectory.
• These models can be applied to various continuous markers of HD state.
• The findings contribute to a better understanding of HD pathogenesis and progression.