Photo Credit: Kateryna Kon
The following is a summary of “Disentangling Neurodegeneration From Aging in Multiple Sclerosis Using Deep Learning: The Brain-Predicted Disease Duration Gap,” published in the November 2024 issue of Neurology by Pontillo et al.
Distinguishing brain aging from disease-related neurodegeneration in patients with multiple sclerosis (PwMS) is essential to understanding disease progression.
Researchers conducted a retrospective study to determine whether a disease-specific model could enhance the brain-age gap (BAG) by capturing aspects unique to MS.
They collected 3D T1-weighted brain MRI scans from PwMS to build a cross-sectional multicenter cohort for age and disease duration (DD) modeling, and a longitudinal single-center cohort focusing on early MS. A 3D DenseNet model predicted DD from minimally preprocessed images, while DeepBrainNet was used to predict age. The brain-predicted DD gap (predicted minus actual DD) was introduced as a DD-adjusted measure of MS-specific brain damage. Model predictions were examined for the effects of lesions and brain volume, and the DD gap was biologically and clinically validated within a linear model, assessing its relationship with the BAG and Expanded Disability Status Scale (EDSS).
The results showed MRI scans from 4,392 PwMS (69.7% female, age: 42.8 ± 10.6 years, DD: 11.4 ± 9.3 years) from 15 centers, with the early cohort of MS consisting of 749 sessions from 252 patients (64.7% female, age: 34.5 ± 8.3 years, DD: 0.7 ± 1.2 years). The model predicted DD better than chance (mean absolute error = 5.63 years, R2 = 0.34) and was nearly orthogonal to the brain-age model (correlation between DD and BAG: r = 0.06 [0.00–0.13], P=0.07). Variations influenced predictions in brain volume and showed sensitivity to MS lesions (difference between unfilled and filled scans: 0.55 years [0.51–0.59], P<0.001). The DD gap significantly explained EDSS changes (B = 0.060 [0.038–0.082], P<0.001), enhancing BAG (ΔR2 = 0.012, P<0.001). A higher DD gap was linked to greater annualized EDSS change (r = 0.50 [0.39–0.60], P<0.001), adding explanatory power for disability progression over BAG alone (ΔR2 = 0.064, P<0.001).
They concluded that the brain-predicted DD gap was sensitive to MS lesions and brain atrophy and serves as a potential MS-specific biomarker for disease severity and progression, complementing the BAG.
Source: neurology.org/doi/10.1212/WNL.0000000000209976
