Neuroimaging in ALS: illuminating disease progression and phenotypical heterogeneity

Amyotrophic Lateral Sclerosis (ALS) is a devastating neurodegenerative disorder characterized by the progressive loss of motor neurons, leading to muscle weakness, respiratory failure, and death. The disease is highly heterogeneous; individual survival ranges from months to over a decade, and up to 50% of patients experience cognitive or behavioural changes, with 5-15% developing frontotemporal dementia (FTD). This clinical and biological diversity complicates therapeutic development, as heterogeneous patient populations frequently mask drug efficacy in clinical trials.

This dissertation demonstrates that Magnetic Resonance Imaging (MRI) is a valuable, objective, and widely accessible tool to map both motor and non-motor dimensions of neurodegeneration in ALS. By utilizing T1-weighted (T1w) sequences to measure grey matter volume and cortical thickness, alongside Diffusion Tensor Imaging (DTI) to assess white matter tract integrity, this research captures the diverse structural footprints of the disease.

Key findings
• Mapping motor degeneration: T1w imaging reveals non-uniform cortical thinning in the precentral gyrus. By analysing the motor cortex along its somatotopic organization (mouth, arms, legs), the study shows that region-specific thinning directly correlates with clinical central motor neuron (CMN) symptoms.
• Superior sensitivity in detecting CMN involvement: Utilizing advanced, tensor-free DTI methods, the research detected reduced corticospinal tract fibre density even in patients lacking clinical symptoms. The study concludes that MRI is more sensitive than traditional neurological examinations, which are often confounded when severe peripheral motor neuron (PMN) muscle atrophy masks underlying CMN dysfunction.
• Neuroanatomical basis for cognitive decline: Cognitive deficits assessed via the Edinburgh Cognitive and Behavioural ALS Screen (ECAS) correlate with specific patterns of grey and white matter decay, mapping specific cognitive features to different regions of the brain. While neuropsychological scores stagnated in longitudinal analyses, MRI successfully detected ongoing structural decline over time. Notably, memory impairments correlated with reduced hippocampal volume, proving they are a primary disease manifestation rather than secondary to executive dysfunction.
• Presymptomatic detection: In asymptomatic C9orf72 carriers, longitudinal MRI detected structural brain atrophy up to six years before symptom onset, allowing accurate predictions of age-dependent disease penetrance. Remarkably, this six-year neuroimaging window outperforms the biochemical biomarker Neurofilament light chain (NfL), which only rises three years prior to symptom onset.
• Characterizing subgroups: Data-driven clustering of MRI data identified three distinct neurodegenerative ALS subtypes: Pure Motor (ALS-PM), Motor/Fronto-Temporal (ALS-FT), and a widespread Motor/Cingulate/Parietal/Temporal/Cerebellar (ALS-CPT) subtype. The latter challenges the conventional theory that ALS spreads along a fixed, linear anatomical pathway.
• Genetic insights: The UNC13A risk allele (rs12608932) was linked to shorter survival, bulbar onset, cognitive dysfunction, and specific temporal lobe thinning. Later evidence points towards UNC13A related cryptic exon inclusion in the context of TDP-43 pathology, only for specific types of neurons. The findings suggest that genetic risk profiles may be linked to local brain vulnerability and phenotypical heterogeneity.

Conclusion
MRI serves as a highly sensitive, multidimensional biomarker capable of monitoring disease progression, classifying distinct pathophysiological subgroups, and detecting neurodegeneration years before clinical symptoms manifest. Future research should integrate neuroimaging into diagnostic, screening, and staging frameworks to optimize early, targeted therapeutic interventions.