A large-scale multicentre cerebral diffusion tensor imaging study in amyotrophic lateral sclerosis

Objective: Damage to the cerebral tissue structural connectivity associated with amyotrophic lateral sclerosis (ALS), which extends beyond the motor pathways, can be visualized by diffusion tensor imaging (DTI). The effective translation of DTI metrics as biomarker requires its application across multiple magnetic resonance imaging scanners and patient cohorts. A multi-centre study was undertaken to assess structural connectivity in ALS at a large sample size.
Methods: Four-hundred-and-forty-two DTI data sets from patients with ALS (N=253) and controls (N=189) were collected for this retrospective study from eight international ALS-specialist international clinic sites. Equipment and DTI protocols varied across the centres. Fractional anisotropy (FA) maps of the control subjects were used to establish correction matrices to pool data, and correction algorithms were applied to the FA maps of the control and ALS patient groups.
Results: Analysis of data pooled from all centres using whole-brain-based statistical analysis of FA maps confirmed the most significant alterations in the corticospinal tracts, and captured additional
significant white matter tract changes in the frontal lobe, brainstem and hippocampal regions of the ALS group that coincided with post mortem neuropathological stages. Stratification of the ALS group for disease severity (ALS functional rating scale) confirmed these findings.
Interpretation: This large-scale study overcomes the challenges associated with processing and analysis of multi-platform, multi-centre DTI data, and effectively demonstrates the anatomical fingerprint patterns of changes in a DTI metric that reflect distinct ALS disease stages. This success paves the way for the use of DTI-based metrics as read-out in natural history, prognostic stratification and multi-site disease-modifying studies in ALS.