Basic Science and Pathogenesis.

Background: Amyotrophic lateral sclerosis (ALS) shares pathological and genetic underpinnings with frontotemporal dementia (FTD). ALS manifests with diverse symptoms, including progressive neuro-motor degeneration, muscle weakness, but also cognitive-behavioural changes in up to half of the cases. Resting-state EEG measures, particularly spectral power and functional connectivity, have been instrumental for discerning abnormal motor and cognitive network function in ALS [1]-[3]. Based on our recent findings using EEG microstates [4], we hypothesized that dynamic analysis of time-varying spectral EEG measures at source-level can further interrogate the altered functional network that underpin the cognition and behaviour in ALS as a motor-dominant neurodegeneration.

Method: For this purpose, we aimed to identify transient brain states associated with specific functional networks and to characterise the spatio-temporal and spectral alterations in these brain states using high-density resting-state EEG, recorded from 99 individuals with ALS and 78 healthy controls (HC). Using a time-delay embedded Hidden-Markov-Model on source-reconstructed EEG data [5], we identified transient and recurrent brain states characterised by spectral power and coherence. The model was trained to convert source-reconstructed time courses into sequences of functional networks (brain states).

Result: Twelve brain states were identified with distinct patterns of spectral power and coherence for individuals with ALS and HC. The Brain States 1, 3, 7 and 9 showed significant association between brain states fractional occupancy and behavioural decline (Beaumont Behavioural Inventory [7]; >0.25, q<0.03, >0.65), while state 5 showed association with fluency decline (Edinburgh Cognitive and Behavioural ALS scale [8]; = -0.3, q = 0.004, = 0.83). States 1, 7 and 10 were characterised by frontal lobe activation (spectral power higher than the average within the state), while state 3 exhibited activation in the sensorimotor network. Highest spectral power during state 5 was in the supplementary motor area, a region primarily associated with motor planning but also possibly associated with language [9].

Conclusion: This study demonstrated the utility of resting-state EEG and its temporal decomposition based on spatio-spectral dynamics for investigating and interrogating the less-dominant cognitive-behavioural decline in ALS. It shows clear potentials for applications in cognate conditions such as FTD and Alzheimer's disease with a dominant cognitive-behavioural decline.