Assessment of corticospinal tract dysfunction and disease severity in amyotrophic lateral sclerosis. (P5.076)

Objective: Upper motor neuron dysfunction in amyotrophic lateral sclerosis (ALS) was quantified using triple stimulation (TST) and more focal transcranial magnetic stimulation (TMS) techniques that were developed to reduce recording variability. These measurements were combined with clinical and neurophysiological data to develop a novel supervised machine-learning (ML) prediction model. Background: ALS is a progressively fatal neuro-degenerative condition. It is associated with differential involvement of upper motor neuron (UMN) and lower motor neuron (LMN) pathways. To better understand differential impact of UMN-LMN components on disease related disability and mortality, objective quantification of different components in ALS disease process is needed. TMS measurements are useful for measuring UMN involvement, but, there is inherent variability associated with most commonly used TMS parameters. A simplified set of TMS parameters, easier to interpret would have better utility in a clinical setting. Design/Methods: Approval was obtained from an institutional review board. 38 subjects with clinical features and diagnosis consistent with motor neuron disease (MND) were included in this study. There were no re-imbursements for participating in this study. We independently recorded ALSFRSr, AALS and MQoLSiS for disease severity, along with raw measurement values that constituted AALS. We developed computational models to compare induced electric (E) field distributions generated by two different TMS coils: 1) circular coil (MagVenture C-100) and 2) the MagPro R-30 stimulator with MC-B70 figure-of-eight coil. Motor-evoked-potential (MEP) amplitude, central motor conduction time (CMCT), TST amplitude ratio (rAmp) and TST area ratio (rArea) and motor-threshold (MT) estimation were recorded using TMS. Results: Conclusions: Significantly, the random forest based supervised ML model was capable of predicting disease severity with an overall accuracy of >97% and a precision of >99%. Disclosure: Dr. Remanan has nothing to disclose. Dr. Sukhotskiy has nothing to disclose. Dr. Shahbazi has nothing to disclose. Dr. Furlani has nothing to disclose. Dr. Lange has received personal compensation for activities with Genzyme as a consultant.