Human Neurodegeneration: A Spectral EEG and TMS based Approach in Amyotrophic Lateral Sclerosis

Abstract

Amyotrophic lateral sclerosis (ALS) is defined by the neurodegeneration of upper and lower motor neurons of the corticospinal tract, resulting in progressive, terminal and incurable decline in movement, speech and swallowing functions. Although motor neuron degeneration provides a unifying characteristic for this diagnosis, individual patients experience extensively heterogeneous motor and non-motor symptoms and progression rates. This not only results in distressing uncertainty for those diagnosed, but introduces unpredictable variation to patient cohorts in clinical trials, diminishing the power of these trials to detect therapeutic effects of novel drug candidates. Extensive imaging, psychology and physiology research to date has illustrated that ALS symptoms and progression rates are driven not only by the motor neurons, but by broader cortical network dysfunction and atrophy. Quantification of the spatiotemporal patterns of cortical network dysfunction in ALS and their relationships to disease symptoms may therefore provide a basis for subcategorising patients early in disease. These physiological measurements can then be clinically implemented to facilitate more specific prediction of individual prognoses. In addition, such measures could substantially improve clinical trial design by enabling stratification of cohorts and more objective, quantitative measurement of drug effects based on fundamental ALS pathophysiology. In this project, threshold tracking transcranial magnetic stimulation (TMS) and electroencephalography (EEG) were implemented to interrogate ALS-related cortical network pathology. Threshold tracking TMS with electromyography was used to investigate corticospinal tract function and the effects of ALS on intracortical and interhemispheric motor networks which regulate the upper motor neurons. In addition to attempting to reproduce previous reports of GABAAergic decline as a biomarker of ALS, the effects of ALS on indices associated with the glutamatergic and GABABergic interneuronal and corpus callosal function in motor networks were investigated by paired pulse paradigms. Single pulse TMS was also used to investigate the latency of signal transmission from cortex to muscle, while peripheral nerve stimulation was applied to quantify lower motor neuron impairment. To interrogate the nature and location of cognitive, sensory and motor cortical dysfunction, EEG was recorded in ALS patients during performance of auditory oddball and sustained attention to response tasks. Cross-sectional and longitudinal EEG signal analyses, in addition to source analysis, were applied to characterise changes in cortical activation using event related potentials, as well as cortical communication using event related spectral perturbations. This work has revealed that ALS drives dynamic patterns of hypo- and hyper- activation and synchronisation in both motor and non-motor cortical circuitry. Namely, TMS and EEG studies indicate that the motor cortex is initially hyperactive in ALS, even during non-motor tasks, potentially due to loss of inhibitory interneuronal function, and that this hyperactivity wanes with disease progression. Similar patterns of early hyperactivity were observed in the dorsolateral prefrontal and posterior parietal cortices, in addition to impaired event related beta oscillation desynchronization being recorded over these areas, suggesting that GABAergic interneuronal decline also occurs in frontoparietal cognitive networks early in disease. By contrast, other sensory and cognitive areas, including the temporal and inferior frontal cortex are initially suppressed, becoming hyperactive later in the disease. Temporal regions also display alpha and beta band hypersynchrony during auditory sensation, which may reflect excessive bottom-up suppression, accounting for observed reduction in auditory cortex activation early in disease. Many of these cortical networking abnormalities correlated with impairments in associated disease symptoms, including cognitive, behavioural and motor decline at the time of recording or in future, as well as survival times. These findings highlight the ability of threshold tracking TMS and EEG to objectively capture the pathology underpinning ALS heterogenous symptoms. These measures might now be further developed to define clinically-relevant, network-based subphenotypes of ALS and to improve clinical trial design.