Theme 08 - CLINICAL IMAGING AND ELECTROPHYSIOLOGY

Background: Serial high-density surface EMG (HDSEMG) has the potential to characterise voluntary and ectopic patterns of motor unit (MU) firing at different stages of ALS (1). Objectives: By distinguishing MU subtypes with variable vulnerability to disease, we aimed to evaluate compensatory neuronal adaptations that accompany disease progression. Methods: Twenty patients with ALS and five patients with benign fasciculation syndrome (BFS) underwent 1–7 assessments each. HDSEMG measurements comprised 30min of resting muscle and 1min of light voluntary activity from biceps brachii bilaterally (1). MU decomposition was performed by the progressive FastICA peel-off technique (2). Inter-spike interval, firing pattern, MU potential area, afterhyperpolarisation duration and muscle fibre conduction velocity were determined. Results: In total, 373 MUs (ALS1⁄4 287; BFS1⁄4 86) were identified from 182 recordings (3). Weak ALS muscles demonstrated a lower mean inter-spike interval (82.7ms) than strong ALS muscles (96.0ms; p1⁄4 0.00919) and BFS muscles (95.3ms; p1⁄4 0.0039). Mean MU potential area (area-underthe-curve: 487.5 vs. 98.7lV ms; p< 0.0001) and muscle fibre conduction velocity (6.2 vs. 5.1 m/s; p1⁄4 0.0292) were greater in weak ALS muscles than in BFS muscles. Purely fasciculating MUs had a greater mean MU potential area than MUs also under voluntary command (area-under-the-curve: 679.6 vs. 232.4lV ms; p1⁄4 0.00144). Discussion: These results show that first-recruited motor units fired faster and with greater amplitude as disease progressed. The development of this faster phenotype may be driven by a compensatory response to replace vulnerable fast-twitch motor units. Inhibition of this potentially maladaptive phenotypic drift may protect the longevity of the motor unit pool, stimulating a novel therapeutic avenue.

IMG-02 Resting state fMRI correlates of pseudobulbar syndrome in amyotrophic lateral sclerosis (ALS) significant differences in gray matter atrophy in patients with PBS compared to patients without PBS. Pseudobulbar index was inversely related to FC in the right MTG and positively related to FC in bilateral precuneus, right postcentral gyrus, right insula and anterior cingulate cortex. Conclusions: Our findings suggested that alterations of fronto-temporal circuits may be related to PBS in ALS. In particular, decreased FC in the right MTG and increased FC in bilateral precuneus and in right fronto-temporo-parietal areas may underlay PBS in patients with ALS. Background: Cognitive and behavioural impairments are observed in up to 50% of ALS patients (1). The ECAS is an ALS-specific multi-domain screening tool widely accepted (2). Although translated into Swedish (S-ECAS), it has not been validated in the Swedish population.
[18F]fluorodeoxyglucose positron emission tomography (18F-FDG-PET) displays a particular pattern in ALS patients, showing both hypo-and hypermetabolism in various brain areas (3). To our knowledge no studies have examined the relationship between ECAS scores and 18F-FDG-PET in ALS patients. Objectives: Validation of the S-ECAS Version A and assessment of the relationship between S-ECAS scores and glucose metabolism patterns on 18F-FDG-PET images of Swedish ALS patients. Methods: We recruited patients who received an ALS-diagnosis between January 2014 and February 2020, at the Karolinska ALS Center in Stockholm. Patients meeting the revised El Escorial criteria for definite/probable/probable laboratory-supported ALS were included and evaluated by several screening tools, including S-ECAS. We included those who had a 18F-FDG-PET measured during the study period. We used multiple regression analyses using the Statistical Parametric Mapping software to assess the relationship between glucose metabolism and cognition, including the preprocessed 18F-FDG-PET images as the dependent variable, ECAS Total scores or ECAS ALS Specific scores as independent variables, in addition to age/sex/site of onset, ALSFRS motor scores and presence of C9ORF72 mutations as covariates. Results were adjusted for multiple comparisons using family-wise error rate corrections at the cluster level (p < 0.05). Validation study has been conducted collecting S-ECAS Version A for 135 ALS patients and 35 healthy controls, all normalized on Montreal Cognitive Assessment. Results: 80 ALS patients were included in the study. Higher ECAS Total scores were associated with higher metabolism in the right orbitofrontal gyrus and lower metabolism in lateral occipital, medial occipital and cerebellar regions. The results in the analyses with ECAS ALS Specific scores were similar, although less widespread. The positive and negative correlations between both ECAS test scores and glucose metabolism overlapped with the areas previously found to show metabolic changes in ALS patients compared to controls. We are analyzing data for the validation study and will share our results at the meeting in December 2021. Conclusions: The S-ECAS seems to be sensitive for detection of cognitive dysfunction in Swedish patients as shown in our interim analyses. The metabolic pattern of 18F-FDG-PET indicates a corresponding change in the glucose metabolism in the areas responsible of the cognitive dysfunction in ALS evaluated by ECAS. juliette.foucher@ki.se

IMG-04 PAS induced recovery of intracortical inhibition in patients with ALS
A. Lackmy-Vallee 1 , I. Peyre 1,2 , G. Querin 3 Background: Evidence from transcranial magnetic stimulation (TMS) studies indicated that cortical hyperexcitability is an early mechanism involved in selective motor neuron death in amyotrophic lateral sclerosis (ALS) (1). In this view, electrophysiogical methods used to modulate cortical excitability may help to improve the balance between excitation and inhibition in the motor cortex in attempt to slow down the disease progression. Paired associative stimulation (PAS) combines low-frequency TMS applied over the motor cortex with stimulation of peripheral nerve afferents. In healthy subjects, PAS induces changes in synaptic transmission in cortical and spinal neuron networks that outlast the period of application (2). To date, only one study has revealed modifications in motor evoked potentials (MEPs) after PAS in ALS (3). Objective: This study aims to explore effects induced by PAS on intracortical inhibitions (ICI) evoked in the motor cortex and described as impaired in sporadic and familial ALS (1). Methods: Fourteen newly-diagnosed ALS patients were included. Short and long-latency ICI were explored in wrist muscles using electromyogram recordings and paired-pulse TMS methods. The degrees of both inhibitions were estimated before and 60 min after one session of repeated PAS (N ¼ 200, TMS over arm motor area combined with electrical stimulations of radial nerve afferents). Results: One session of PAS failed to modulate long-latency ICI. However, PAS enhanced short-latency ICI, especially in patients exhibiting the more impaired inhibition at baseline. This supports that PAS may help to recover GABAA-mediated ICI to counteract cortical hyperexcitablity in ALS. Background: Recent human imaging and postmortem studies suggest relative sparing of brain in SOD1 ALS disease biology (1,2). However, a recent in vivo imaging study suggests increased brain glial activation in ALS individuals with SOD1 variants of unclear significance (3). It remains unknown if brain is involved or spared in symptomatic SOD1 ALS individuals. TSPO-PET is a putative marker of neuroinflammation. Prior studies show increased TSPO-PET signal in ALS motor cortex, which correlates with clinical markers of upper motor neuron burden (UMNB) (4,5).
Objectives: To measure and compare brain glial activation and motor cortical atrophy in SOD1 ALS participants versus historical matched healthy controls, using [11C]PBR28 TSPO-PET radioligand. Methods: Ten ALS participants with "likely-pathogenic" or "pathogenic" SOD1 variants underwent clinical assessments and a combined PBR28-PET/MR scan at Massachusetts General Hospital. Using our group's previously published PET/ MRI analyses protocols (4) and historical age/sex/TSPO binding affinity matched healthy control data (HCs), whole-brain voxelwise and apriori defined region of interest (ROI) (primary motor cortices) analyses of PBR28-PET uptake signal were performed. The primary endpoint, PBR28-PET uptake metric SUVRwb [¼mean standard uptake values in ROI normalized to mean whole brain standard uptake values] was computed and compared between groups. MRI-based cortical thickness was measured in the same ROI and compared between groups. T-tests were used for all comparisons and p value <0.05 for ROI and Z > 2.3 Pcluster ¼0.05 for voxelwise analysis. Results: Of the ten enrolled SOD1þ ALS participants (age 55 þ 11 years, 7 males), nine participants were included in final analyses. Seven participants had rapid [A5V, G93S, G94A, A5T, I114T] and two slow [D90A, L145] progressive SOD1 variants. Consistent with known SOD1 ALS phenotype, 8/9 participants had clinically lower motor neuron predominant phenotype [Median (range) MGH-UMNB (5) score was 20.5 (9, 29)]. The median (range) disease duration at scan was 8.3 (3,170) months. Whole brain voxel-wise and ROI analyses showed no detectable differences in PBR28-PET SUVRwb in SOD1 ALS (N ¼ 9) versus HCs (N ¼ 8). The Mean þ SD PBR28-PET SUVRwb in ROI were SOD1 ALS (0.9866 þ 0.04) and HCs (0.9927 þ 0.04), p ¼ 0.8. SOD1 ALS (2.4083 þ 0.1008 mm) and HCs (2.4416 þ 0.1151 mm) showed no cortical thickness differences in ROI. Conclusion: There were no significant brain glial activation changes or cortical thinning in SOD1 ALS cohort. This is inkeeping with the known clinical phenotype of lower motor neuron predominance as well as recent postmortem findings of relative brain sparing in SOD1 ALS. Background: Although Amyotrophic Lateral Sclerosis (ALS) primarily affects motor neurons, non-neuronal cells also play an important role in disease progression. The inflammatory cascade, which follows motor neuronal injury, is mediated in part by chronic microglia activation and proliferation in the central nervous system. Neuroinflammation has been established as a contributing factor to the pathophysiology of ALS. Microglial survival and proliferation are controlled by the receptor tyrosine kinase CSF-1R, and disease progression in ALS animal models is associated with altered CSF-1R signaling. BLZ945 is a synthetically derived and potent inhibitor of the protein receptor CSF 1R. In a study in SOD1G93A ALS mice, BLZ945 administration resulted in histological evidence of dose-dependent microglia clearance as well as in-vivo maintenance of normal body weight gain and a dose-dependent delay of disease-related motor impairments. Therefore, selective inhibition of CSF-1R to deplete microglia may represent a therapeutic strategy for the slowing of ALS progression. Objectives: To evaluate the potential therapeutic application of a CSF-1R inhibitor for disease modification in ALS using [11C] PBR28 PET imaging. Methods: The expression of the Translocator Protein (TSPO) is highly upregulated during microglial activation. TSPO PET imaging, using the PET tracer [11C] PBR28, is the current gold standard technique for in vivo assessment of neuroinflammation in humans. In a study of the safety and tolerability of BLZ945 in individuals with ALS, (NCT04066244), 20 study participants will receive four days of daily BLZ945 in five dose escalation cohorts. The change in TSPO PET imaging signal in the motor cortex, from pre to post dose will be used to determine proof of mechanism for BLZ945 in ALS. Conclusion: Specific targeting of neuroinflammation may have a therapeutic benefit in ALS. The use of TSPO PET imaging has the potential to serve as an effective tool for the early evaluation of experimental compounds that target neuroinflammation.

Ronenn.roubenoff@novartis.com
Background: Early detection of bulbar impairment is critical in ALS for differential diagnosis from ALS mimics. However, an assessment for dysarthria is often confusing to physicians particularly in those with mild bulbar impairment. We aimed to characterize acoustic properties of speech disturbance in Korean ALS patients and investigate the relationship between acoustic parameters and clinical features of ALS. We further analyzed early changes of acoustic parameters detected in ALS patients with perceptually normal voice. Method: Speech samples were collected from 43 patients with an ALS spectrum disorder (27 patients with dysarthria (ALSwD), 16 patients without dysarthria (ALSwoD)) and 20 healthy controls. The presence or absence of dysarthria was determined by the speech sub-score (4 points) of ALSFRS-R. The speech material comprised three corner vowels, /a, i, u/. The syllables of speech samples include the six repetitions of a single open syllable /pi, pu, pa/ and 39 closed monosyllabic words. We identified the duration and formant frequency of each vowel and calculated the vowel space area (VSA) from three corner vowels.
Results: Vowel durations of the three corner vowels were significantly longer in the ALSwD group than those in the ALSwoD and healthy control groups. F1 frequencies of vowel /a/ and the vowel space area (VSA) of ALSwoD group were significantly lower compared to healthy controls. The VSA exhibited an association with speech score of ALSFRS-R.  (3,4). Objectives: To understand whether: (a) ALS-FTD and ALS is on a continuum, (b) ALS-FTD and classic ALS (cALS) are distinct clinical subtypes with potentially distinct sites of neurodegeneration.
Methods: An exploratory whole brain grey matter (GM) analysis using voxel based morphometry (VBM) and white matter (WM) network analysis were performed using graph theory approach. Diffusion tensor imaging and T1-weighted MRI data were obtained for 14 neurological controls, 25 cALS, and 15 ALS-FTD patients.
Results: Significant differences in nodal degree measures of the WM network were observed between ALS patients and controls in frontal and motor areas, extramotor regions, subcortical regions and cerebellum. WM networks changes were significantly different between ALS-FTD and cALS patient groups. Significant differences in GM volume were observed only between ALS-FTD patients and controls but not between cALS patients and controls. Discussion: Distinct and non-overlapping regions of involvement between ALS-FTD and cALS patient subgroups do not support a continuum of degeneration. Our GM VBM results and graph theory WM network analysis indicate that these two clinical phenotypes of ALS have distinct patterns and potentially unique mechanisms of neurodegeneration.

venkatnrp78@gmail.com
Background: Previous studies using paired-pulse Transcranial Magnetic Stiumulation showed, in both sporadic (sALS) and familial (fALS) forms of Amyotrophic Lateral Sclerosis (ALS), an early cortical hyperexcitability in the motor cortex. This is a form of cortical dysfunction that preceds lower motor neuron signs, suggesting to be a poweful biomarker promoting early diagnosis of ALS. Eletroencephalography (EEG) can be a valid alternative approach to detect and monitor cortical dysfunction through investigation of the interaction between neural oscillations at different frequencies, known as Cross-Frequency Coupling (CFC) analysis, more precisely the study of phase-amplitude coupling (PAC) between slow and fast oscillations. CFC is established to be highly dependent on a proper cortical excitation/inhibition balance and it is known to be altered in neurodegenerative conditions such as Parkinson's and Alzheimer's Diseases or Fronto-Temporal Dementia (1,2). Objective: Our aim is to investigate whether there is an altered CFC in resting state EEG (rsEEG) in ALS patients and if this alteration can represent an early biomarker of cortical dysfunction in ALS. Our research hypothesis is to find a perturbed PAC in ALS patients rsEEG as compared to healthy controls.
Methods: Data were collected on 26 sALS patients with mild motor deficits (median ALSFRS-r score ¼40) and 27 healthy controls. An high density EEG recording protocol was used (74 channels, 4 kHz sampling rate, bandpass 0.03-1330 Hz), consisting in 5 min recording with eyes closed and 5 min with eyes open. RsEEG data was pre-processed to erase nonneural artefacts using an independent component analysis approach and filters to select EEG signal below 60 Hz. PAC analysis was performed on 5 channels around the cranial vertex (Fz, Cz, Pz, C3, C4 according to the international 10-20 system) and it was evaluated using the Tort  Background: Threshold-Tracking Transcranial Magnetic Stimulation (TT-TMS) refers to the variation of magnetic pulse intensity to achieve a motor evoked potential (MEP) of target amplitude. Lower short intracortical inhibition (SICI, a measure of interneuronal, GABA-Aergic inhibition of the corticospinal tract) in ALS compared to mimic disease and healthy populations, has been identified using TT-TMS and has been proposed as a diagnostic biomarker (1). These abnormalities now requires replication in other patient population samples. Furthermore, all previous TMS studies of ALS cortical pathophysiology have applied stimulation via coils positioned in the typical posteroanterior (PA) orientation. However, stimulation with different coil orientations, such as anteroposterior (AP) orientation, is established to preferentially engage different axons within the motor cortex and, as a result, affect the characteristics of the MEPs generated. For example, SICI is significantly greater in healthy individuals when AP coil orientation is used compared to PA orientation (2). This indicates that alternative coil orientations may reveal additional, and potentially more specific, biomarkers of ALS.
Objective: To investigate the sensitivity and reproducibility of TT-TMS based measures, measured with different orientations of magnetic field, in detecting ALS-associated cortical network disruption.
Methods: EMG was recorded from the APB muscle while fully-automated TT-TMS protocols (3) were applied over the contralateral motor cortex using a 50 mm figure-of-eight coil with PA and AP directions of induced current flow. SICI, intracortical facilitation (ICF), long intracortical inhibition (LICI) and short and long IHI were measured. Data from 19 patient and 31 controls has been recorded to date. Statistical comparison was undertaken with two-sided t-tests as data were confirmed normal by Shapiro-Wilks testing. Background: Interrogation of network-level dysfunction in ALS holds promise in developing quantitative markers for diagnosis, prognosis, and disease progression. The objectives of this study were to examine effective connectivity (EC) during motor tasks (i.e. motor planning (MP) and motor execution (ME)) in ALS and compare with age-matched controls, and to provide novel neurophysiology based biomarkers of motor network disruption in ALS. Methods: Electroencephalography (EEG, 128-channels) was recorded from 16 healthy controls and 22 ALS patients during motor tasks (i.e. isometric pinch grip tasks between thumb and index finger of the right hand to maintain 10% of maximum voluntary contraction). Source data were estimated from sensor data using linearly constrained minimum variance (LCMV) beamformers. Effective connectivity between left (contralateral, c)/right (ipsilateral, i) primary motor (cM1/ iM1), primary sensory (cS1/iS1), supplementary motor area (cSM/iSM), prefrontal (cPF/iPF), anterior cingulate (cAC/iAC), and superior parietal (cSP/iSP) areas were estimated using partial directed coherence (PDC) for 8 frequency bands and compared between ALS and healthy controls. Causal flow (CF), the difference between causal outflow and inflow, was calculated for each cortical region of interest and compared between ALS and healthy controls.

Results
Results: Comparison of effective networks between ALS and healthy controls revealed significant differences (p Ã < 0.05): During MP, cSM!cM1 EC at delta band was weaker whereas iS1!cSP EC at gamma-high was stronger in ALS; during ME, cSP!iS1 and iSM!iS1 ECs at theta band were stronger in ALS, whereas cM1!cAC EC at beta-high was weaker. During MP, iPF (p Ã ¼ 0.012) showed increased CF at the theta band, whereas cSP (p Ã ¼ 0.033) showed decreased CF at the gamma-high band in ALS. During ME, ALS showed decreased CF in iPF (p Ã ¼ 0.034) and increased CF in cSP (p Ã ¼ 0.041) at delta band compared to healthy controls. ALS also showed increased CF in both cSP (p Ã ¼ 0.034) and cAC (p Ã ¼ 0.010) at theta and higher-beta frequency bands, respectively, during the ME task.

001). Comparison of patients in different
Phases showed further decrease of cortical thickness along with disease progression (p < 0.001). However, there were no differences in surface measures between patients with low and high disease aggressiveness (p < 0.001). Regression analyses identified negative correlations between cortical thickness and individual disease covered (p < 0.001). Discussion and conclusions: Using SBM we were able to reveal structural changes of grey matter in patients with ALS compared to controls that worsen in parallel to clinical disease progression. By application of the D50 model, we could thus identify robust correlations between the decrease of cortical thickness and ALS-related functional disability, but not with disease aggressiveness. This indicates that cortical thickness could be a biomarker representing individual disease covered independent from the overall speed of disease progression. Our findings support the applicability of cortical thickness as a biomarker to monitor disease Phase in ALS, which is particularly valuable for clinical trials. Objectives: To develop and provide preliminary validation for a multifactorial facial surface EMG framework as a novel objective bulbar assessment tool. Methods: Participants included 13 patients with ALS, with or without clinical bulbar symptoms, and 10 healthy controls, each performed a speech task three times (total N ¼ 69). All participants provided informed consent in accordance with the Institutional Review Board approval by the University of Kansas Medical Center. Fifty-four linear and nonlinear features were extracted from the surface EMG recordings for masseter, temporalis, and anterior belly of digastric to characterize the attributes of mandibular muscular performance, including complexity, regularity, frequency, amplitude, and intermuscular coordination. These features were factor analyzed. The resulting factors were evaluated in terms of internal consistency, correlation with the established functional speech measures, and efficacy for patient-control classification. Results: Five internally consistent, interpretable factors were derived (Cronbach's alpha !0.89), representing the functioning of masseter, temporalis, digastric, antagonist muscles, and agonist muscles, respectively. These factors explained 40-43% of the variance in the functional speech outcomes, based on linear regressions, and were !95% accurate (sensitivity, specificity !90%) in patient-control classification, based on three supervised machine learning algorithms. The factors related to antagonists coordination and temporalis activity were the most contributive to both the regressions and classifications. Discussion: Two important bulbar muscular changes were identified in ALS, related to both UMN and LMN pathologies. These changes reflected (1) decreased motor unit recruitment and synchronization for jaw antagonists and (2) increased recruitment for temporalis, attributable to a potential neuromuscular adaptation. The mandibular surface EMG framework therefore shows promise as an objective, quantifiable means of assessing both UMN and LMN bulbar involvement in ALS.

prong@ku.edu
Background: Cortical thickness is a potential marker of brain neurodegeneration in ALS. Previous studies have shown cortical thinning in the motor cortex in ALS; however, it remains unknown whether speech impairment in ALS is associated with neurodegeneration involving other brain regions responsible for speech production. Understanding regional thickness differences will provide a better understanding of clinical phenotyping of speech abnormalities and functional changes in ALS.
Objective: To evaluate thickness distribution patterns in ALS individuals with speech impairments versus healthy controls (HCs). Methods: Thirty-seven participants with ALS and 22 sex and age-matched HCs underwent clinical and a T1-weighted 3-Tesla qMRI evaluation in a cross-sectional study at MGH. The ALS participants were divided into two groups based on their ALSFRS-R speech subscore: preserved speech (ps) (N ¼ 16 ps-ALS participants had normal speech score ¼4); and deteriorated speech (ds) (N ¼ 21 ds-ALS participants had speech scores 0-3). Fifteen apriori-defined and automatically segmented cortical and subcortical brain regions of interest (ROIs) were selected based on involvement in motor speech. Three analyses were performed: (i) whole-brain vertex-wise analyses in ps-ALS and ds-ALS versus HCs (cluster -clusterforming threshold ¼1.3, p ¼ 0.05); (ii) ROIs comparison in ps-ALS and ds-ALS versus HCs; and (iii) association between M1 thinning and ALSFRS-R total score in the entire ALS group. All analyses were adjusted for age and sex. Results: The ps-ALS and ds-ALS subgroups were similar in their baseline clinical characteristics, except for (a) ds-ALS group had lower ALSFRS-R bulbar subscores (p < 0.0001) and (b) ps-ALS had lower ALSFRS-R gross motor subscores (p ¼ 0.04). The whole-brain analyses showed significant cortical thinning in the ds-ALS group in the left pre-and postcentral gyri (lip and tongue movements) and the right inferior parietal lobule (silent [p] [t] repetition) compared to the HC (p < 0.05). The ROI analyses showed significant thinning in the ds-ALS group compared to HCs in the left cingulum gyrus (R 2 ¼ 0.35, p < 0.0001, laryngx opening), bilateral insula (R 2 > 0.20, p < 0.04, articulatory planning), left middle-dorsal and bilateral middle pre-motor cortex (R 2 > 0.27, ps <0.02, speech motor initiation), left dorsal pars opercularis (R 2 ¼ 0.22, p ¼ 0.02, Broca's area/speech articulation) regions, in addition to bilateral M1(R 2 > 0.23, ps <0.01, articulatory movementslips, tongue, jaw). There was no differences in the ps-ALS group compared to HCs in the whole-brain and ROI analyses except for the right pars opercularis (R 2 ¼ 0.22, p ¼ 0.04). The ALSFRS-R total scores correlated positively with M1 thickness (Right M1 R 2 ¼ 0.23 p ¼ 0.003, Left M1 R 2 ¼ 0.15, p ¼ 0.02). Conclusion: Speech deterioration in ALS is associated with cortical atrophy extending beyond the motor cortex and involving multiple cortical areas of motor speech production. Future longitudinal studies in early ALS will help understand temporal progression patterns in these cortical areas, filling critical gaps of motor speech impairment and prediction modeling for speech in ALS. Background: The split elbow sign was recently reported to be a clinical feature of ALS, characterised by preferential weakness of biceps brachii compared to triceps muscle (1,2). Specifically, the biceps brachii muscle was shown to exhibit a greater degree and frequency of weakness, as reflected by Medical Research Council (MRC) score, when compared to the triceps muscle. Objectives: A novel neurophysiological index, termed the split elbow index (SEI), was developed to quantify the splitelbow sign, and assess its utility in ALS. Methods: Clinical and neurophysiological assessment was prospectively undertaken on 34 ALS patients and 34 ALS mimics. Compound muscle action potential (CMAP) amplitude was recorded from biceps brachii and triceps muscles from which the SEI was calculated using the following formula: SEI ¼ CMAP amplitude (BICEPS BRACHII)/CMAP amplitude (TRICEPS BRACHII). Results: The split elbow sign was significantly more common in ALS patients when compared to ALS mimic patients (p < 0.05). The SEI was significantly reduced in ALS patients when compared to ALS mimics (p < 0.01). This reduction was evident in both spinal and bulbar onset ALS. A SEI cut-off value of 0.62 exhibited a sensitivity of 72% and specificity of 62%. Discussion: The split elbow sign is significantly more common in ALS patients, and was supported by a reduction in the SEI. The SEI may be utilised as a surrogate biomarker of the split elbow sign in future ALS studies. Background: Burgeoning evidence highlights the critical role of cortical hyper-excitability in ALS pathogenesis. Previous work (1) using Transcranial Magnetic Stimulation (TMS) demonstrates cortical hyper-excitability is produced by dysfunction across both facilitatory (short interval intra-cortical facilitation or SICF) and inhibitory (short interval intra-cortical Inhibition or SICI) interneuron circuits. Critically a direct invivo demonstration of cortical hyper-excitability in ALS has not yet been achieved. Objectives: Recent technical advances allow for the direct capture of responses from cortical interneuron populations in the form of the Transcranial Evoked Potential (TEP), generated through a combination of TMS, electroencephalography (EEG) and advanced signal processing. Earlier work (2) demonstrated that the P60 amplitude of the TEP is reduced through activation of the GABA-ergic inhibitory circuits of SICI. Here we demonstrate a loss of SICI and P60 TEP inhibition in ALS patients and advance the technique further by demonstrating an increase in SICF potentiation of the TEP P60 potential amongst ALS patients, thereby directly demonstrating the shift toward cortical hyper-excitability is delineated by dysfunction across both circuits. Methods: TMS evoked potentials were studied in 12 ALS patients and 15 age-matched healthy controls (HC). Motor Evoked Potentials (MEP) were recorded using a belly-tendon montage over the Abductor Pollicus Brevis muscle, while TEP were recorded using a standardized 64 channel EEG array. TMS responses were studied under three conditionsa single pulse delivered at 110% of Resting Motor Threshold (RMT), an inhibitory paired pulse paradigm (SICI) with a 70% RMT conditioner delivered 3 ms before a 110% RMT test pulse, and a facilitatory paired pulse paradigm (SICF) with a conditioner delivered 3 ms after the 110% RMT test pulse. MEP and TEP recordings were processed offline, with TEP processed using an established signal processing protocol (3). Results: There was a significant reduction in TEP P60 inhibition in ALS patients (À14.1% ± 16.9) compared to controls (À55.5% ± 13.9, p ¼ 0.037) when assessing the SICI paradigm. Additionally, P60 facilitation was enhanced in ALS (165.0% ± 69.3) compared to controls (24.7% ± 41.3%, p ¼ 0.0392) within the SICF paradigm. These changes were accompanied by a reduction of SICI (ALS ¼ À6.6% ± 11.6, HC ¼ À74.0% ± 5.3, p ¼ 0.0001), and increase in SICF (ALS ¼228.7% ± 80.1, HC ¼61.9% ± 24.3, p ¼ 0.0385) recorded by MEPs. Discussion: The data from TMS-EEG TEPs directly demonstrate cortical hyperexcitability is present in ALS patients and characterized by dysfunction across both inhibitory and facilitatory interneuron circuits. Interventions aimed at modulating these cortical circuits may prove therapeutically useful. Additionally, methods that capture the signature of cortical hyper-excitability (TEP and MEP changes) may be of diagnostic utility.

mehdivandenbos@gmail.com
Background: Previous studies of asymptomatic carriers of ALS-associated C9orf72 gene expansions have reported the occurrence of cognitive and behavioural changes, as well as dysfunctional brain metabolism and structure degeneration (1,2). Little is known, however, about the neurophysiological changes that may occur before clinical disease manifestation. Early detection and characterisation of disease manifestations in asymptomatic C9orf72 expansion carriers ('C9orf72-positive') can underpin new therapeutic strategies that are based on early, targeted treatments. Objectives: To evaluate changes in cortical networks responsible for attention and behavioural inhibition in asymptomatic C9orf72 expansion carriers (AEC) using electroencephalography (EEG). Methods: High-density EEG (128 channels) was recorded during the randomised sustained attention to response task (SART) in 14 AEC (C9orf72-positive ALS family members) and 17 age-matched controls (12 population-based controls and five C9orf72-negative ALS family members). The task requires participants to respond every time they see a digit (1-9, Go) except for the digit 3 (NoGo). We have assessed five behavioural measures of performance: NoGo accuracy (percentage of three-digit stimuli followed by response omission), Go accuracy (percentage of non-three digit stimuli followed by a response in the permitted time window), anticipation error (clicking less than 150 ms after a Go stimulus) and response time. Using advanced source-localisation methods, we evaluated the SART-generated event related potentials associated with behavioural inhibition: N2 (200-350 ms post-stimulus) and overt attention: P3 (350-500 ms); both known to be impaired in ALS (3). For both groups, a difference in activations during the two potentials were estimated for the NoGo condition. Statistical comparisons were done using Mann-Whitney test. Results: Controls and AEC did not differ significantly in response time (p ¼ 0.54) or NoGo accuracy (p ¼ 0.1). However, AEC committed significantly more anticipation errors (p ¼ 0.02), while Go accuracy was marginally significant (p ¼ 0.06). The right superior and middle frontal gyri were significantly more active during P3 compared to controls (p < 0.05). The right inferior parietal lobule was significantly less active during N2 compared to controls (p < 0.05). Discussion: These results show the potential of EEG to capture functional changes associated with the asymptomatic carriers of C9orf72. The identification and characterisation of biomarkers that can be linked to the early development of ALS, can in turn help in the early diagnosis and enhance the understanding of causal physiological processes. This approach is important for early treatment strategies and to make ALS a potentially preventable disease.

S.Dukic@umcutrecht.nl
Background: ALS is a progressive motor neuron disorder that leads to eventual death (1). Although neuroimaging seems to be a reliable potential biomarker especially several studies showed significant MRI metrics changes between ALS patients and healthy controls, there is an unmet need for reliable biomarkers, not only for better diagnosis but also to provide a reliable assessment of disease progression (2). Therefore, the aim of this study is to measure the change of spinal cord MRI metrics over time based on a prospective, longitudinal, multipoint study. Methods: This study is an ancillary analysis using data collected from the Paris Center which is part of the PULSE study. PULSE is an ongoing observational and prospective multicentric cohort in ALS patients. We included 40 ALS patients who underwent a structural and diffusion MRI. Magnetic resonance imaging (MRI) scans were acquired on 3T Siemens scanner, and clinical variables collected over three-time points. Spinal cord toolbox (SCT) was used to treat the structural and diffusion images to compute cross-sectional area (CSA) per-level and DTI parameters (FA, MD, RD, AD) at the lateral corticospinal tract and the dorsal columns at the cervical level. Clinical and demographic data will be then evaluated for correlations with cervical spinal cord imaging findings. Results: At the inclusion timepoint, MRI damage parameters, including CSA per-level and the DTI parameters at the lateral corticospinal tract and posterior dorsal columns at the cervical level, showed significant difference within the cohorts when we divided them regarding age at onset and site of onset. A significant difference in the MRI damage parameters was found within subgroups regarding the rate of progression as measured by the ALSFRS. Background: Motor unit loss frequently precedes the reduction of compound muscle action potentials (CMAP) in neurodegenerative diseases, as reinnervation masks early motor unit loss (1). Similarly, clinical weakness is masked, making qualitative scores insensitive in early disease stages. The CMAP scan is a practical bedside test from which motor unit number estimates (MUNE) can be reliably derived (2). Hence, this technique has the potential to quantify motor unit loss in distal muscles, providing earlier markers for motor neuron diseases (MND). Routine screening with this technique could also aid in quantifying disease progression. Objective: To determine the association between CMAP scan MUNE and well-established clinical scores in patients with suspected MND. Methods: We prospectively recruited 92 patients who were referred to our outpatient clinic under suspicion of MND and 12 age-matched controls. Patients underwent routine diagnostic workup for MND, after which at the same day CMAP scans were recorded from the abductor pollicis brevis (APB). We derived a MUNE from the CMAP scan with MScanFit (2). We subsequently assessed the association between MUNE and well-established clinical scores, including the ALSFRS-R, fine motor function (FMF) and MRC score of the APB. Results: The patient population consisted of 66 (72%) ALS, 4 (4%) PMA, 2(2%) PLS and 20 (22%) patients who eventually got another diagnosis. Median (range) of MUNE in patients was 49 . Associations of MUNE with clinical scores were best modeled with 3 knot cubic splines. Best association by MUNE were MRC scores (R 2 ¼ 0.35, p < 0.001), followed by FMF scores (R 2 ¼ 0.23, p < 0.001) and ALSFRS-R scores (R 2 ¼ 0.08, p < 0.05). Ceiling effects for clinical scores became apparent from MUNE >36, well below MUNE identified in age-matched controls (lower limit of normal defined as 5th percentile ¼52). Discussion: Our results confirm the notion that a decreased MUNE was observed in patients who still had normal functional scores. As such, CMAP scan MUNE is a potential diagnostic and prognostic marker which more closely follows the pathophysiological process of neurodegeneration. Future work should focus on longitudinal assessment of multiple muscles to further identify the potential benefits of this useful quantitative technique. Background: Threshold-tracking transcranial magnetic stimulation (TT-TMS) measures have been shown to be of diagnostic and prognostic value in ALS (1). Specifically, a decrease in short intracortical inhibition, which captures interneuronal GABAA-ergic inhibition of the corticospinal tract has repeatedly been demonstrated to discriminate ALS patients from controls and mimic disorders (2). TT-TMS is, therefore, increasingly being used to measure central motor network pathophysiology in ALS. These TMS research studies require participants to remain relaxed enough to consistently maintain very low amplitude EMG in the target muscle for several minutes at a time. However, unlike MRI studies (3), TMS study protocol typically does not allow any audio or visual input to occupy the participant as a precaution against the potential influence of sensory input on these motor measures (4). As a result, study participation can be difficult. Where diseaserelated muscle tension and stiffness causes high baseline EMG amplitudes, data must be discarded and replaced, prolonging session duration and potentially limiting the number of measures that can be recorded. Additionally, patients may become distressed or overly attentive to the stimuli and resulting in contractions in the absence of distraction. This in turn may influence cortical excitability and the TMS measures being investigated (5). Objective: To investigate whether watching and/or listening to a documentary that does not relate to movement affects commonly-studied TMS measures of motor cortical excitability. Methods: Data are being collected from 10 healthy control volunteers. EMG is recorded from both dominant and nondominant APB muscles while fully automated TT-TMS is used to measure resting motor threshold (RMT), threshold hunting target (THT), SICI, long intracortical inhibition (LICI) and interhemispheric inhibition (IHI). Stimuli are delivered to the motor cortex contralateral to the dominant hand. In the case of IHI, the conditioning stimuli are delivered to the motor cortex ipsilateral to the dominant hand. Recordings are first taken with no sensory stimulation, followed by visual and auditory stimulation, only auditory stimulation, and finally once more with no sensory stimulation. Figure-of-eight coils with 50 mm windings are being used for all stimuli. The effects of auditory and/or visual stimulation on baseline EMG amplitudes, the number of rejected trials and measures of motor cortical excitability will be investigated by comparison to data collected without distractor stimuli present. Linear mixed-effects modelling will be used to account for the effects of sensory stimuli, age and order of sessions. Results: Data collection is ongoing until August 2021 and results will be reported at the 2021 MNDA Symposium. tadjiney@tcd.ie