Platform Communications: Abstract Book - 32nd International Symposium on ALS/MND (Complete printable file)

Fundamental research to understand how bacteria fight viral infections uncovered the function of CRISPR-Cas programmable proteins that detect and cut specific DNA or RNA sequences. I will describe our research showing how RNA-guided CRISPR-Cas proteins can be used for re-writing the DNA in cells and organisms, providing transformative technology for genome editing. Current research focuses on exploring the biochemical basis for genome editing and developing effective applications in medicine and agriculture. I will also discuss the development of CRISPR-based diagnostics technology to address the current coronavirus pandemic and improve future preparedness.

The online format brought new challenges in serving the interests of such a diverse range of delegates located across multiple time zones, but in collaboration with the MND/ALS community, we rose to the challenge. The event was a great success and hopefully we will be able to bring such enthusiasm again, as we hold a virtual meeting for the second year in a row.
In a world of noise, the Symposium continues to play a fundamental role, not only in facilitating the exchange of exciting new knowledge and information, but also serving as an interdisciplinary 'melting pot', stimulating new ideas and fostering new collaborations. We really appreciate each and every one of you playing your part in that.
We have yet again included the traditional mix of basic and clinical science, therapy development and clinical management that is synonymous with the Symposium. We extend our thanks to our plenary speakers who have graciously agreed to deliver their presentations online; and this year we have been able to review and select submitted poster presentations to be 'upgraded' to oral presentations from the outstanding pool of poster abstracts submitted, to add to the program.
As last year, the Poster Sessions remain in the virtual world and are packed with extremely high-quality content. It will be a challenge to generate the 'buzz' of the face-to-face event, but we hope to encourage a similar feel between poster presenters and delegates through live online interaction, including Q & A sessions, networking, gamification and video presentations of each poster.
Finally, we thank the Symposium Programme Committee, chaired by Professor Ammar Al-Chalabi, for their tireless work and valuable advice.
We hope you will join us online in our collective effort to understand, treat and -ultimately -defeat ALS/MND. We wish you a successful and enjoyable Symposium. The number of therapeutics in development for ALS has exploded since the ice bucket challenge in 2014. In the last 18 months the programs from pharma, biotechs and academic institutions are tackling several different pathways and utilizing biomarkers and companion diagnostics to accurately predict safety and efficacy in advance of clinical benefits. This presentation will review the current drug development event highlighting the programs which are using these techniques Until the emergence of CSF and serum neurofilament levels as biomarkers potentially indicative of response to therapy drug development programs were dependent upon clinical response to therapy to determine whether to move an investigational product forward. There is now a concerted effort in many programs to include measures of response in the early phase programs. This presentation will review the current ALS programs in the preclinical and early clinical space highlighting the key outcome measures in each program which are specific to the proposed mechanism of action in addition to neurofilament levels and to other more general aspects of trial design. Academic, Biotech and pharma sponsored trials will be covered in this presentation. Programs presented will include programs in the neurodegenerative , neuroinflammatory, hyperexcitability, toxic protein, oxidative stress, neuromuscular as well as the gene specific programs. Recommendations for future trial designs will be included in the concluding remarks.
angela.genge@mcgill.ca C3 Long-term functional benefits and safety of a fixed-dose coformulation of sodium phenylbutyrate and taurursodiol in amyotrophic lateral sclerosis Background: An oral, fixed-dose sodium phenylbutyrate (PB)/taurursodiol (TURSO) coformulation was designed to reduce neuronal death by mitigating endoplasmic reticulum and mitochondrial dysfunction. Objective: Report long-term functional and safety results from the CENTAUR trial of PB/TURSO in ALS. Methods: CENTAUR encompassed a 24-week randomized, double-blind, placebo-controlled period and up-to-132-week open-label extension (OLE) period. Adults with definite ALS (revised El Escorial criteria) 18 months from symptom onset were randomized 2:1 to PB/TURSO or placebo. Participants completing all randomized period visits were eligible to enroll in the OLE period and receive PB/TURSO. Prespecified analyses compared changes in extended slope for continuous efficacy outcomes from randomization through week 24 of the OLE period (48 weeks total) in the originally randomized treatment groups within the modified intent-to-treat (mITT) population (ie, all randomized participants receiving !1 dose of study drug with !1 postbaseline efficacy assessment), using a mixed model with repeated measures. This model was used to estimate ALS Functional Rating Scale-Revised (ALSFRS-R) total score, upper-/lower-limb Accurate Test of Limb Isometric Strength (ATLIS) scores, and slow vital capacity (SVC) in both groups at week 48. Safety was evaluated through week 24 in each period individually in the safety population (ie, all participants receiving !1 dose of study drug in the respective trial period). Results: A total of 137 participants were randomized ( The proportions of participants who experienced !1 treatment-emergent adverse event (TEAE) in the groups starting on PB/ TURSO and placebo, respectively, were 97% and 96% during the randomized period and 73% and 82% in the OLE period. Gastrointestinal TEAEs were more common in the first 3 weeks among participants receiving PB/TURSO in the randomized period and in participants switching from placebo to PB/TURSO in the OLE period. Discussion: Earlier initiation and longer duration of PB/ TURSO were associated with long-term functional benefit, as measured by slope of ALSFRS-R total score progression over 48 weeks. Improved retention of measures of upper-limb and respiratory muscle strength was also observed in those initiating PB/TURSO 24 weeks earlier. TEAEs attributable to PB/ TURSO were primarily gastrointestinal.

spaganoni@mgh.harvard.edu
Background: A randomized, placebo-controlled study (AB10015) previously demonstrated that masitinib (4.5mg/kg/ day), administered orally as an add-on to riluzole, slowed the rate of functional decline in ALS patients having an ALSFRS-R progression rate from disease-onset to baseline (DFS) of <1.1 points/month (1). At the time of final readout, overall survival (OS) data were too immature for interpretation. Objectives: Long-term OS analysis of study AB10015, testing whether a signal in OS is evident in an enriched patient population similar to that prospectively defined for ongoing confirmatory study AB19001. Methods: Survival status of all patients originally randomized in AB10015 was collected from participating investigational sites. Survival analysis (using the multivariate log-rank test and Cox proportional-hazards model, with stratification factors as covariates) was performed on the intention-to-treat population and enriched subgroups, which were defined according to initial randomization, DFS and baseline disease severity. Results: The survival analysis followed all patients originally randomized in study AB10015 for an average duration of 75 months (time of diagnosis until long-term OS analysis cutoff). Consistent with previously communicated results (1), no long-term survival advantage was observed for the overall masitinib 4.5 mg/kg/day cohort of study AB10015 (i.e. regardless of baseline disease severity or DFS) or for the low-dose (3.0 mg/kg/day) masitinib treatment-arm. Conversely, in ALS patients with mild or moderate disease severity at baseline and DFS <1.1 points/month, masitinib 4.5mg/kg/day (n ¼ 45) prolonged survival by 25 months relative to those treated with riluzole alone (n ¼ 62) (median OS of 69 versus 44 months, respectively, p ¼ 0.037) with a 47% reduced risk of death (Hazard Ratio 0.53 (95%CI [0.31-0.92]); p ¼ 0.025). People with mild or moderate ALS comprised patients that had not suffered a complete loss or severe impairment of ALSFRS-related functionality at the time of masitinib treatment initiation (i.e. patients with a score of at least 2 on each ALSFRS-R individual component). Survival data for this cohort of patients were corroborated by the effect observed in the endpoints of DALSFRS-R at week-48 and time-to-event analysis (PFS), further supporting the premise of greater treatment effect when masitinib is initiated at an earlier stage of disease. Discussion: These positive survival results provide evidence that the main efficacy outcomes of study AB10015, i.e. DALSFRS-R at week-48 (according to the primary endpoint and sensitivity analyses based on multiple imputation jumpto-reference techniques) and PFS, are legitimate surrogate endpoints for long-term OS. Results are also consistent with masitinib's mechanism of action, which is expected to provide better outcomes if administered early in disease progression (2). Confirmatory study AB19001 is ongoing with a primary endpoint of decline in ALSFRS-R from baseline to week-48 in the enriched patient population described herein.

albert.ludolph@rku.de
Background: MSC-NTF cells (NurOwn V R ) are autologous bone marrow derived mesenchymal stem cells that have been shown to favorably modify neuroprotective and neuroinflammatory cerebrospinal fluid (CSF) biomarkers following single intrathecal administration (1). We conducted a phase 3, randomized, placebo-controlled, clinical trial to evaluate the efficacy and safety of repeated administration of MSC-NTF cells in ALS patients. We have separately reported on the clinical effects of MSC-NTF cells from this trial. The goal of this analysis was to evaluate treatment effects on CSF biomarkers and their relationship to clinical outcomes. Methods: MSC-NTF cells were administered intrathecally on 3 occasions, 2-months apart. CSF samples were obtained prior to the first treatment and at 6 additional times. CSF biomarkers were analyzed using validated multiplex or singleplex assays (Simoa, Procaratplex, and Proximity Extension). Bioinformatic tools were applied to investigate relationships between CSF biomarkers and clinical outcomes. Principal component analyses (PCA) were performed on neuroinflammation, neurodegeneration, and neuroprotection pathways, identified by their contribution to clinical data models to efficiently assess the treatment effects on each pathway separately. A pre-specified stepwise regression model was used to select biomarkers that that were predictive of the primary efficacy endpoint, from all biomarkers measured, and assess the treatment effect in a combined model. Results: Robust and statistically significant CSF biomarker changes from baseline were observed with MSC-NTF treatment compared to Placebo in biomarkers related to neuroinflammation, neurodegeneration, and neuroprotection, consistent with earlier trials. For example, there were statistically significant treatment differences in VEGF-A and MCP-1 at all post-treatment time points (p < 0.05), with MSC-NTF increasing VEGF-A two-fold following treatment (p < 0.0001) and MCP-1 decreased by 74% compared to Placebo post treatment (p < 0.0001). NfL values on MSC-NTF were 82% relative to Placebo following treatment. PCA analyses retains most of the information provided across all biomarkers collected with 5 PCA components. PCA confirmed the conclusions across biomarkers that MSC-NTF increased neuroprotection and decreased neuroinflammation and neurodegeneration relative to Placebo. A statistical model demonstrated that MCP-1, Fetuin-A, VEGF-A, MSR1, and NfL were all important predictors of treatment response (all p < 0.05), as defined by the primary efficacy endpoint of the trial. The Receiver Operating Curve highlights strong model performance with 82.5% accuracy. Analysis of the primary clinical efficacy endpoint leveraging this model reveals a significant treatment response (p ¼ 0.003), with more MSC-NTF participants meeting the clinical response definition than Placebo participants. Discussion: The significant post-treatment CSF biomarker changes observed in this study and their relevance to clinical response suggest MSC-NTF cells have important effects on multiple ALS disease pathways. The data also suggests that simultaneously targeting multiple ALS disease pathways may be necessary to achieve meaningful clinical outcomes. Bioenergetic failure is increasingly considered as a unifying pathogenic mechanism in a number of ALS sub-types with energy depletion in high energy demanding neurons making them sensitive to apoptosis (1). One potential response to bioenergetic failure is to increase the energy supply to neurons and other CNS cells. There is encouraging evidence from mouse models and clinical trials that meeting the energy deficit with a high calorie diet in pwALS may affect disease progression. Malnutrition and weight loss are well recognised poor prognostic factors in amyotrophic lateral sclerosis (ALS). In people living with ALS (pwALS) weight loss of 5% or more there is associated with a two-fold increase in risk of death. The weight loss is multifactorial and is compounded by hypermetabolism with resting energy expenditure being, on average, 20% higher than in healthy individuals (2). A lower pre-morbid BMI in ALS patients and a 40% risk reduction of ALS in obese patients support the role of weight and energy homeostasis in influencing ALS pathogenesis. Current nutritional management is often not evidence-based with little guidance on nutritional management regarding assessment of nutritional status, total daily energy expenditure calculations, appropriate dietary intake or oral nutritional supplementation. We have identified a lack of knowledge in healthcare professions regarding the nutritional management of ALS (3). The poor evidence base explains the poor nutritional outcomes observed. If we accept increasing energy intake is a potential therapeutic option, then there are complex challenges to be overcome; not least a need to consider how to influence the nutritional behaviours and knowledge of pwALS and also the staff and services that support them. We have undertaken a user centred design approach to create OptiCALS (Optimal Calories in ALS), a complex nutritional intervention. OptiCALS supports pwALS to increase their calorie intake using a food first approach and behaviour change techniques. Additionally, the OptiCALS tool provides an evidenced based framework for healthcare professionals to assess nutritional status and set calorie goals for pwALS. The effectiveness of the OptiCALS intervention is being assessed in a large randomised controlled trial (ISRCTN -No. 30588041). Assessment of nutritional status is largely based on weight which is problematic given the challenges of using weighing scales as disease progresses. Therefore more practical assessment methods are needed. Advances in bioimpedance techniques and nutritional biomarkers may provide alternative information about body composition and energy status. Objectives: Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease lacking curative therapeutic approaches. Clinical diagnosis of ALS has an average 11month latency due to complication of clinical symptoms. In recent years, the concept of gut-brain axis inspires novel perspectives of diagnosing and curing neurological diseases. However, studies from several groups worldwide reported the gut microbiota results for ALS patients with either small sample size or inconsistent conclusions. This study aims to explore the link between gut microbiota and pathological mechanisms of ALS. Methods: In current study, we conducted the 16S ribosomal RNA gene sequencing microbiota sequencing in fecal samples of 35 ALS diagnosed early-to-middle stage ALS patients and 35 condition-matched healthy controls, a subgroup of samples (29 ALS and 23 controls) was conducted untargeted metabolomics mapping. Results: b-diversity of gut microbiota indicated significant difference between ALS and controls (p < 0.05). A higher median ration of Firmicutes/Bacteroidetes (F/B) was found in ALS patients (ALS: 29.07 vs HC: 18.32). With untargeted metabolomics mapping, significantly different metabolites between ALS and HC were enriched in nicotinate and nicotinamide metabolism pathway, retrograde endocannabinoid signaling, inflammatory mediator regulation of TRP channels, sphingolipid metabolism and thiamine metabolism (p < 0.05). Nicotinate and nicotinamide metabolism pathway was downregulated in ALS patients, while retrograde endocannabinoid signaling, inflammatory mediator regulation of TRP channels, sphingolipid metabolism and thiamine metabolism were upregulated in ALS patients. Conclusions: The major findings of the current study indicate that, gut microbe in ALS changes in a pro-inflammation direction, oxidative stress in ALS may be aggravated by the dysregulation of bacterial metabolites. Our study provides an in-depth view of gut microbiota and substance characteristics based on decent sample sizes with multi-omics strategies. It would strengthen the link between gut microbe/metabolic substance and ALS pathogenesis. Subsequently, we evaluated the relationship between the lipid profile and survival time in a large population-based registry. We used Cox proportional hazards models to assess the prognostic value of total cholesterol (TC), LDL-, HDL-cholesterol, and triglycerides (TRI), adjusted for known prognosticators. Results: The systematic review comprised eight articles; four studies (50%) found a beneficial effect on survival in patients with high levels of TC, LDL/HDL ratio or TRI, corrected in each study for different confounders. Therefore, it is difficult to draw an overall conclusion. Our prospective cohort study included 1,346 consecutively patients with ALS of whom a lipid profile was determined at diagnosis. A dose-response relationship between HDL-cholesterol and survival (HR 1.35 (95% CI 1.15-1.57, p < 0.001)) was found, increasing the risk of death with 35% per every point increase in HDL-cholesterol. None of the other lipid components were significantly associated with survival when adjusted for known prognosticators. Conclusions: Our large population-based study shows that HDL-cholesterol is an independent prognosticator of survival. Gaining further insight in cholesterol metabolism, HDL functioning, and metabolic stress, might reveal novel aspects of ALS pathogenesis and potential targets for treatment. Protein folding is managed by the 'protein homeostasis' or 'proteostasis' machinery, which comprises the cellular pathways that control the biogenesis, trafficking, folding, degradation of proteins. These integrated pathways are distributed across several subcellular compartments that interact in a coordinated way. Efficient and balanced proteostasis is essential to maintain cellular health and viability and to protect against environmental challenges. Furthermore, compared to other cell types, neurons are large, long-lived cells with complex morphologies. These characteristics can present unique demands on the proteostasis network to dynamically regulate the motor neuronal proteome. Loss of proteostasis is central to protein misfolding diseases and not surprisingly, dysfunction of this network is implicated in pathogenesis of ALS. Furthermore, collapse of proteostasis also occurs during normal aging, which is the main risk factor for ALS. Here I will review recent studies linking the intricate proteostasis mechanisms to ALS, and how this is related to other emerging cellular mechanisms implicated in ALS pathophysiology. Therapeutic strategies based on restoring proteostasis or on targeting specific components of the proteostasis network, therefore, offer potential for future approaches for ALS and other protein misfolding diseases.
julie.atkin@mq.edu.au C10 Sporadic ALS disease initiation and targeted therapy: nuclear accumulation of CHMP7 initiates nuclear pore complex injury and subsequent TDP-43 dysfunction in sporadic and C9orf72 ALS J. Rothstein 1 , A. Coyne 1 , F. Rigo 2 , F. Bennett 2 and C. Lusk 3 1 Johns Hopkins University, Baltimore, MD, USA; 2 Ionis Pharmaceuticals, Carlsbad, CA, USA; 3 Yale University, New Haven, CT, USA Alterations in the components (nucleoporins, Nups) and function of the nuclear pore complex (NPC) have been implicated as contributors to the pathogenesis of genetic forms of neurodegeneration including C9orf72 Amyotrophic Lateral Sclerosis/Frontotemporal Dementia (ALS/FTD). We hypothesized that Nup alterations and the consequential loss of NPC function may lie upstream of TDP-43 dysfunction and mislocalization widely observed in ALS, FTD, and related neurodegenerative diseases. We now have accumulated evidence that CHMP7, a critical mediator of NPC quality control, is increased in nuclei of C9orf72 and sporadic ALS induced pluripotent stem cell (iPSC) derived spinal neurons (iPSNs) and postmortem human motor cortex prior to the emergence of Nup alterations. Using a large cohort of Answer ALS supplied sporadic ALS iPSC (n > 30) and human autopsy specimen, we also determined that CHMP7 has enhanced nuclear localization, in the same neurons with nuclear TDP43 depletion. Inhibiting the nuclear export of CHMP7, triggered Nup reduction and TDP-43 dysfunction and pathology in human neurons. Knockdown of CHMP7 alleviated disease associated Nup alterations, deficits in Ran GTPase localization, defects in TDP-43 associated mRNA expression, and alleviated downstream glutamate induced neuronal death. Analytics of the pathway over time revealed that CHMP7 nuclear location, preceded nuclear pore complex degradation which in turn preceded the loss of nuclear tDP43 in human spinal neurons. Thus, our data support a role for altered CHMP7 mediated Nup homeostasis as a prominent initiating pathomechanism for sporadic as well as C9orf72 ALS/FTD and highlights the potential for CHMP7 as therapeutic target. Importantly, these studies identify a molecular target for sporadic ALS and one that may repair the common molecular defect in nuclear pores as well as TDP43. Background: Apart from the well-defined neuron-centric factors, few reports consider that variability of sporadic ALS progression can depend on the less-defined contributions from non-neuronal cell types including glia and blood vessels (1). Nonetheless, variability of ALS patient survival continues to confound clinical trial design and estimations of disease dynamics remain based only on neuronal cell derived outcomes. Objectives: To better understand how non-neuronal cells contribute to ALS aetiology. Methods: We inferred cell activity in ALS spinal cord transcriptomes using single-cell guided profiling, vascular histopathology and plasma profiling.
Results: Here we report that perivascular fibroblast cell gene activity during presymptomatic disease stage remodels blood vessel matrix and provides distinct plasma protein biomarker that can independently predict short ALS patient survival at diagnosis (2). We determined that sporadic ALS patients present cellular changes consistent with two mouse models in which gene expression patterns from vascular cells precede the blood-brain barrier dysfunction and microglial response. Notably, perivascular fibroblast cells elicited the strongest pre-onset gene enrichments and their marker proteins SPP1 and COL6A1 accumulated in enlarged perivascular spaces in sporadic ALS patients. Moreover, in 574 ALS patients from four independent cohorts, increased plasma levels of SPP1 at disease diagnosis repeatedly predicted shorter survival with a stronger effect than established indications of bulbar onset or neurofilament levels in cerebrospinal fluid. Discussion: Although vascular dysfunction has previously been proposed as an epidemiological risk factor in humans (3) and contribute to the multi-step ALS hypothesis (4), our results instead provide a plausible molecular and cellular explanation for these associations and a potential target for therapy. We propose that the activity of the recently-discovered perivascular fibroblast can predict ALS patient survival and provide a novel conceptual framework to re-evaluate definitions of ALS aetiology. In Switzerland, assistance to suicide is legal, unless the assisting person incites a person for selfish reasons (Article 115 of the Swiss Penal Code). Euthanasia, thus active killing, is forbidden. Assisted suicide is increasingly accepted in Swiss society. Since the 1980s, right-to-die organisations have institutionalized the process of assisted suicide for adult patients. Whereas the rate of suicide without assistance has decreased in the years of 2000-2018 (from 19.8 to 12 per 100,000 inhabitants), the rate of medically assisted suicide has risen continuously, from 1.5 to 14.3. So far, the national monitoring of suicide acts provides little information about underlying diagnoses and patients' origin (suicide tourism). We will present data about the development and circumstances of assisted suicide in patients of the ALS Clinic Basel, led by Kathi Schweikert and localized in the University Hospital and the REHAB Basel, respectively. In this setting, patients have access to a multiprofessional network of care, including advance care planning and early palliative care, and parallely to right-to-die organizations. Trends for assisted suicide of patients cared for by the ALS Clinic Basel are compared to those nationwide. Wishes to die in patients with incurable disease are grounded on personal motivations, but also on trajectoryrelated patterns, problems and concerns (1). Wishes to die may be a mean to preserve agency in a situation of uncertainty and loss of control. In ALS patients, wishes to die statements seem to be more frequent than for example in patients with cancer. In our interview-study on 62 patients with incurable disease (7 patients with ALS) wishes to die in ALS were directed to a future that is likely to come along with suffocation or heavy dependency, but also with psychosocial and spiritual obstacles, as hopelessness, loss of control, loss of role or the self-perception of being a burden (SPB). Patients with SPB showed an altered self-understanding that did not meet mutual expectations within their relationships (2). Family caregivers felt deeply touched by SPB and tried to unburden patients by giving care and compassion. SPB were not only associated with feelings of self-doubt, shame or guilt, but also with important feelings of solicitude and love. Acknowledging this seemed to unburden patients from wishes to die. The relational aspect is important in ALS, since patients' extensive care and cognitive impairment can foster unbalanced relationships. The relational dimension of a wish to die implicates that the situation and concerns of family caregivers should be addressed early in dialogue and decisionmaking. Supporting  The ENCALS survival prediction model offers patients with ALS the opportunity to receive a personalized prognosis of survival at diagnosis (1). We developed a communication guide to support physicians in tailoring prognostic discussion to the individual needs and preferences of patients with ALS (2). Objectives: To explore experiences of patients with ALS, caregivers, and physicians with discussing personalized prognosis in ALS. Methods: We conducted a qualitative study using semi-structured interviews with patients and caregivers, and a focus group with physicians. We invited recently diagnosed patients with ALS, referred to one of three ALS care teams in the Netherlands and with whom the personalized prognosis was discussed. The focus group consisted of five physicians involved in the recruitment of participants. Interviews and focus group were conducted by researchers not involved in the care of patients (RvE, LK, AB), and transcribed and analysed thematically (RvE, LK). Results: Data saturation was reached after interviewing twelve patients and nine caregivers. Patients' prognosis ranged from short to very long. Three overarching themes with eight subthemes emerged. I Tailored communication.
Physician's tailoring of communication style and information provision mediated the emotional impact of prognostic disclosure and increased satisfaction with communication.
Regardless of good or bad news, most patients and caregivers were satisfied with the communication. II Personal factors. Coping style, illness experiences, and information needs affected how patients and their caregivers coped with the prognosis and were largely independent of role (patient or caregiver), gender, or prognosis. III Regaining control over the future. The emotional impact on patients and caregivers ranged from happy and reassuring to regret. They found prognostic disclosure helpful looking towards the future and emphasized the importance of quality over quantity of time left to them. These themes were confirmed by the focus group. Physicians described prognostic disclosure with nonwestern patients or a language barrier as challenging. In parallel, protocols for the directed differentiation of hPSCs into a myriad of cell types have been established, and it is foreseeable to such protocols will become available for accessing literally any cell type on demand. In the current presentation, I will discuss the state of hPSC-based models for neurodegeneration with a particular focus on ALS.
Human PSC technology has matured to the level where some of the very first clinical trials based on results obtained in iPSC-based studies have been initiated, and where iPSCderived lineages are becoming a routine tool for genomewide genetic screens or in drug discovery studies. However, many serious challenges remain to realize the potential and make iPSC technology truly predictive of human neurodegenerative disease. Those include cell line to cell line variability both between and within individuals that can impact differentiation potential and induce or mask the emergence of disease phenotypes. Another major drawback is the immature (fetal to neonatal) stages of neuronal maturation that can be achieved using current iPSC-differentiation approaches, that is a mismatch from the adult or aged state at which neurodegenerative disease typically occurs. This is reflected in immature physiological properties of human iPSC-derived lineages and the lack of adult or age-related genetic and epigenetic signatures. Finally, most iPSC-based studies are focused on modeling disease features in a given neuronal lineage such as spinal motoneurons in ALS, midbrain dopamine neurons in PD or cortical neurons in AD. However, it is important to capture the role of glial cells including the modeling of neuroinflammatory interactions in iPSC-based models. Furthermore, it will be critical to study neuronal function in the context of an appropriate target such as neuromuscular interactions in ALS or nigrostriatal interactions in PD. In my presentation, I will provide an update on the technologies developed in my lab to address some of those key challenges including our goal to achieve a better control of cell type composition and studying neuronal/glial interactions in defined proportions using a novel tri-culture system. I will also present strategies under development to induce enhanced levels of maturation in iPSC-derived lineages.
Those are just a few of the remaining challenges to be addressed for realizing the full potential of iPSC-technology in modeling disease and in personalize medicine.
studerl@MSKCC.ORG C16 ALS drug discovery using AI platform with patient iPSC panel Background: Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease that causes progressive loss of motor neurons. The disease progresses quickly, and the development of effective medicine is urgently required. The discovery of induced pluripotent stem cell (iPSC) technology enabled us to analyze the pathophysiology of ALS and screen therapeutic drugs using patient motor neurons. We have constructed a phenotypic screening system to evaluate the efficacy of compounds with a readout of motor neuron survival using ALS patient iPSCs. Methods: To find a potent drug for many ALS patients, we developed an artificial intelligence (AI)-based drug discovery algorithm with patient iPSCs. Although it is still a challenge to apply machine learning to achieving sufficiently complex phenotypic drug screening due to imbalanced datasets, nonlinear prediction, and unpredictability of new chemotypes, we could overcome these issues by constructing a prediction model based on a heat diffusion equation (PM-HDE), a novel non-linear approach that uses a partial differential equation describing heat distribution for prediction of the efficacy of compounds.
Results: The algorithm of PM-HDE was verified as being feasible for virtual compound screening using biotest data of 946 assay systems registered with PubChem as benchmark sets. PM-HDE presented high prediction performance showing high AUC levels even in datasets with a bias of the active ratio. Furthermore, the performance of PM-HDE was compared with the k-nearest neighbor (kNN), random forests (RF), and support vector machine (SVM), and showed higher accuracy compared to those of the well-known in silico hit predictors. PM-HDE was applied to actual screening to evaluate compounds to rescue ALS motor neurons. Based on supervised learning of the data of about 50,000 compounds from biological phenotypic screening with ALS motor neurons derived from patient iPSCs, virtual screening of >1.6 million compounds was implemented, followed by a second screening which confirmed that PM-HDE enriched the hit compounds and identified new chemotypes. Finally, the compounds identified by PM-HDE demonstrated broad and potent effectiveness against motor neuron death in ALS patient iPSC panels.
Discussion: This AI platform using patient iPSCs could be a powerful tool for discovering promising compounds for drug development. It may be possible to screen unlimited numbers of compounds, and spread of the application to the drug discovery for various other diseases is also expected. Results: Chronic activation of hM3Dq receptors in cortical motor neurons induced significant hyperexcitability, progressive motor deficits and degeneration of cortical motor neurons, spinal motor neurons, axons and neuromuscular junctions in mice. Cortical hyperexcitability also lead to astrocyte and microglia activation in brain and spinal cord. Importantly, cortical hyperexcitability triggered TDP-43 phosphorylation and cytoplasmic aggregation of TDP-43 in both cortical and spinal motor neurons. Discussion: These results establish that cortical hyperexcitability is sufficient to drive TDP-43 pathology in the brain and spinal cord, associated with sequential degeneration of cortical motor neurons, spinal motor neurons and their peripheral synapses, consistent with a 'dying-forward' mechanism of ALS onset. C18 AI-augmented search for disease-modifying treatments: a new era in ALS drug discovery J. Hunter

BenevolentAI, London, United Kingdom
The complex aetiology and heterogeneity of neurodegenerative diseases, such as ALS, has proved a formidable barrier to the translation of basic research into effective treatments. With a rapidly aging global population, the need to address this unmet medical need becomes ever more pressing. The exponential increase in biomedical data generation and research output, however, offers confidence that we are on the cusp of making substantial progress. BenevolentAI combines advanced AI and machine learning with cutting edge science to decipher complex disease biology, generate novel insights and discover more effective medicines. The company's unique AI-applied drug discovery platform spans every step of the drug discovery process, powering an in-house pipeline of over 25 drug programmes from early discovery towards clinical phases. This presentation will examine how the application of this technology is helping to enhance the mechanistic understanding of human disease, therapeutic target identification and uncover more effective disease-modifying treatments for ALS and other neurodegenerative diseases.
Jackie.Hunter@benevolent.ai C19 How important are biomarkers in drug development?

R. Bowser
Barrow Neurological Institute, Phoenix, AZ, USA ALS biomarker research has been ongoing for over 30 years, with early studies focused on targeted amino acids or proteins that may be elevated in the blood or cerebrospinal fluid (CSF) of ALS patients when compared to healthy controls or disease control groups. Newer technologies permitted large, unbiased studies to identify protein, metabolic or genomic changes in ALS patients. These discoveries uncovered numerous biologic pathways altered in ALS that have been targets for therapeutic development. Beyond identifying pathways targeted for drug development, biomarkers contribute in many ways to the ALS drug development process. Pathologic markers identified in ALS models of disease that recapitulate the pathology of the human disease (i.e. TDP-43 pathology) permit preclinical studies that target pathology and hopefully can be translated to the human disease. An important use of biomarkers to demonstrate target engagement in preclinical and clinical studies has only more recently been incorporated into ALS drug development. Biomarkers that can be measured and monitored in biofluids (blood, CSF, urine) may highlight downstream impact of drug treatments both in preclinical models and during a clinical trial. More recently, biomarkers have been used to select a patient population that may best respond to a particular treatment, or to monitor the efficacy of the treatment during a clinical trial. This impacts the design of a trial and may also lower the number of participants necessary to evaluate the efficacy of the treatment. Finally, the use of biomarkers to identify ALS at a pre-symptomatic phase or to monitor disease progression has great potential to impact future clinical trials. While much work remains till ALS biomarkers are used by regulatory agencies to make drug approvals, there is little doubt that biomarker research has greatly accelerated ALS drug development and has been a key factor in the expanding pipeline of ALS clinical trials during the past decade.
Robert.bowser@barrowneuro.org In contrast to the traditional project-teams that focus on the development of a single novel drug candidate, the ALS team is instead indication focused. This team works internally in close alignment with the respective molecule-focused teams in order to leverage learnings across programs; establish biomarkers and align trial design. And externally, the team works to build relationships with ALS patient groups and clinical consortia, in order to ensure patient-centric clinical trial designs and optimal research, and clinical study efficiency. Results: In the absence of disease-related PD biomarkers, TSPO PET imaging, using the PET tracer [11C] PBR28, is being used as an early PD biomarker to determine proof of mechanism of neuroinflammatory agents. Such proof of mechanism approaches are needed to select promising drug candidates and justify the costs of larger, later-phase studies.
Home-based digital biomarkers that allow for the collection of digital endpoints by study participants and their caregivers in their own home environment are being evaluated in a dedicated observational study. These digitally collected endpoints have the potential to increase the sensitivity of clinical trial endpoints and improve trial efficiency, in addition to decreasing trial participant burden. A master study design protocol is being developed that can be applied across multiple drug candidates with the goal of facilitating clinical operations. For Ph2 and 3 studies, platform designs that maximize the ability of patients to receive potentially helpful treatment and reduce placebo exposure are important. The establishment of strategic alliances with ALS consortia including collaborating with existing platform trials will similarly facilitate clinical operations. An ALS patient advisory board, convened during the trial protocol drafting stage, provided guidance related to the patient-centric design consideration, which increased the attractiveness of clinical trials to prospective study participants. , what is normal and abnormal and how to measure change over time. Furthermore interpretation can be hazardous, there is a difference between cognitive and capacity assessment. Measurement has enabled understanding of pathological cerebral changes (5) and progression (6), the relationship between frontotemporal spectrum disorder and disease stage (7), differentiating between different dementias (8), and has had an impact on clinical care (9). s.abrahams@ed.ac.uk Background: The conceptualisation of ALS has changed dramatically over recent decades from one that was traditionally considered a pure motor disorder, to what is now deemed a multisystem neurodegenerative condition. The overlap across ALS and Frontotemporal Dementia (FTD) has been studied extensively, however the extent of similarities (or differences) across this spectrum in terms of non-motor impairment and associated modifying factors remains less well-explored.
Objective: This study aimed: (1) to identify the pattern of non-motor impairment (including neuropsychiatric symptoms, sleep and mood disorders) across behavioural and cognitive ALS phenotypes compared to behavioural-variant FTD (bvFTD); (2) to determine the overlapping non-motor features across ALS and FTD subtypes to ultimately confirm the extent of the ALS-FTD spectrum; and (3) to establish the contribution of age, sex, disease state and cognitive status on the severity of non-motor features. Methods: Consecutive participants were recruited and underwent a detailed clinical, cognitive, behavioural, neurophysiological and neuroimaging assessment. Neuropsychiatric and other non-motor symptoms were determined using the Cambridge Behavioural Inventory (CBI-R). The scores were converted to define impairment in terms of symptom severity (mild, moderate and severe) for each subscale. Initially rate and severity of symptoms were determined and compared across groups, then contribution of age, sex and disease states were analysed and finally a regression model identified predictors of symptom severity. Results: In total, 250 participants (115 ALS, 98 bvFTD, and 37 ALS-FTD) were recruited. A similar pattern of neuropsychiatric symptom severity was identified (apathy, disinhibition and stereotypical behaviour) for all behavioural phenotypes of ALS (ALS with behavioural impairment ([ALSbi], ALS with cognitive and behavioural impairment [ALScbi] and ALS-FTD) compared to bvFTD (all p > 0.05). Neuropsychiatric symptoms were also present in cases defined as ALS without cognitive and behavioural impairment (ALSPure) and the cognitive phenotype of ALS (ALSci), although they were less frequent and at the milder end of the spectrum. Disordered sleep and disrupted mood were common across all phenotypes (all p < 0.05). The severity of sleep dysfunction was influenced by both sex and age (all p < 0.05). All non-motor symptoms were common early in the disease process and deteriorated in line with progression on the Amyotrophic Lateral Sclerosis Functional Rating Scale-Revised (ALSFRS-R; all p < 0.05). Diagnostic phenotype, disease duration and global cognition scores were the strongest predictors of non-motor and neuropsychiatric impairments.

Conclusion:
The current findings support the current nuanced concept of the ALS-FTD spectrum by identifying a strikingly similar pattern of non-motor symptoms across the subgroups of ALS and bvFTD. These findings further highlight the impact of non-motor, including neuropsychiatric symptoms, on both the disease trajectory and quality of life for people living with ALS. This advanced understanding of the ALS-FTD spectrum may accelerate the early identification of patient needs and improve clinical awareness of these symptoms. ALS is a rare disease, and diagnosis is currently made by exclusion of other conditions. The diagnostic pathway is determined by presentation to a healthcare professional with new symptoms; recognition of "red flag" symptoms /signs; appropriate investigations to exclude other diagnoses; timely and accurate disclosure of diagnosis; and referral to a specialist clinic. Patient journey studies have shown that those with ALS are seen by 3-5 healthcare professionals and average 4-5 investigations prior to definitive diagnosis. Early referral to a general Neurologist may not reduce diagnostic delay but reduces the costs of investigation. Ideally, all patients with ALS should be offered the opportunity to participate in clinical trials. However, with some notable exceptions, the majority of specialist clinics continue to describe a mean diagnostic delay of 12-15 months, which represents up to 30% of life expectancy for the majority of those with ALS and provides a mere 3-6-month window for eligibility for many trials. However, improving diagnostic delay is challenging. Increasing public awareness of ALS and education of primary care practitioners of "red flags" such as tongue fasciculations can trigger early referrals to specialist clinics. But this risks "false positive" presentations, limiting the availablility of expensive specialist services for those with definite diagnoses and increasing waiting lists in public clinics. Conversely, failure to recognize the variability of presentation of ALS, including cognitive/behavioural change, and excessive reliance on neurophysiologic evidence, particularly in bulbar onset disease, can delay referrals from general to specialist clinics. Reluctance by non-specialist neurologist to disclose a terminal diagnosis, nihilistic perspectives about clinical trials, and excessive and inappropriate investigations or treatments to exclude rare mimic syndromes can also delay definitive diagnosis. How can we do better? Diagnostic biomarkers based in biochemical, imaging and neurophysiology will be helpful. But whether current technologies reduce diagnostic delay remains to be determined. In the meantime, there is limited evidence that increasing awareness among general practitioners reduces diagnostic delay. Increased referrals from primary care will require that clinics accept higher rates of "non-ALS" for diagnostic exclusion. This will likely require an expansion of ALS clinics. Additional, expedited diagnosis will require a recognition of the heterogeneity of ALS, the inclusion of cognitive and behavioural change as part of the clinical phenotype, and the identification of "at risk" groups (e.g. members of kindreds with familial ALS). There is scope for educating secondary and tertiary care clinicians in appropriate investigations including the judicious use of emerging biomarkers, and the value of early referral to specialist clinics. Ultimately, expedited referral to specialist neurologists of those with suspected ALS remains the best ways to reduce diagnostic delays and improve access for patients to clinical trials. Background: Home-based monitoring of spirometry allows for more frequent assessments of respiratory function in patients with motor neuron disease (MND) and saves travel and clinic burden, compared to regular MND care. However, use of home-based spirometry in MND care is still lacking.
Objective: Therefore, we aim to assess the reliability and feasibility of unsupervised home-based vital capacity (VC) testing in patients with MND. Methods: We included 33 patients with MND who had access to a smartphone, and did not use non-invasive or invasive ventilation during daytime. Patients completed a 12week home monitoring protocol, consisting of 4-weekly unsupervised home assessments of the upright VC with a full-face mask, functional impairment (ALSFRS-R) and dyspnea-related symptoms (MND Dyspnea Scale). In total 9 patients were assisted with home-monitoring by a caregiver. At baseline, during a home visit, patients and caregivers were trained in performing a VC test, and the investigator performed a supervised VC test, which was repeated at final follow-up during a second home visit. Reminders were sent at each follow-up through text message or email. Inter-rater reliability was determined through the 95% limits of agreement and the intraclass correlation coefficient (ICC) between the unsupervised and supervised VC tests. Sensitivity and specificity were assessed to determine whether the unsupervised VC could detect a supervised VC <80%, and feasibility was assessed through a survey on user-experiences, including items on user-friendliness, burden of monitoring and complexity of measuring. With the recent approval of Zolgensma for the childhood motor neuron disorder Spinal Muscular Atrophy (SMA) and the exponential growth in companies committed to gene therapy development for rare and neurological disorders, there is much excitement and anticipation for this therapeutic modality for adult motor neuron disease/ amyotrophic lateral sclerosis (ALS). Whilst SMA is a monogenic disorder, multiple genes contribute to the demise of neurons in ALS. Mutations in the gene encoding Cu/Zn superoxide dismutase (SOD1) account for 2% of ALS. Pre-clinical studies in SOD1 mouse models provide the preliminary rationale to block production of SOD1 protein and hence the gain of a toxic function, an approach now in phase III clinical trials using antisense technology. The discovery of hexanucleotide expansions C9orf72 accounts for 10% of ALS and is potentially amenable to a gene therapy approach. Gene therapy provides a one-time therapeutic approach directly to the root cause of the disease, as opposed to multiple ongoing administration of small molecules or antisense oligonucleotides. With growing evidence that neurofilaments provide a strong early predictor of disease several years before clinical onset one could envision treatment over an extended period, favoring the gene therapy approach.
Whilst the field has learned numerous lessons from the development of gene therapy for a childhood disease, SMA, and we can leverage these learnings for ALS, there are significant considerations when comparing a childhood disorder to an adult disorder. The route of administration and an understanding of the cells types transduced is critical. In addition to challenges in delivery, immunogenicity remains a concern and various approaches are in consideration to circumvent these challenges. Furthermore, with advances in capsid technology, the hope is that we will eventually be able to regulate gene expression and have the ability to switch the expression of a gene product on and off as needed. The significant investment into gene therapy development demands rigorous trial design and biomarker development. A clear correlation between target engagement and a change in a clinically meaningful endpoint is important for the approval of this therapeutic modality. Whilst there is progress in the development of biomarkers for ALS, there are limited data in the field showing a correlation between these promising biomarkers (soluble and imaging biomarkers) and clinical outcome measures. Technology, for people with ALS and other disabilities, in many ways is the absolute cutting edge of the technology industry. Technology for disabled people is a training ground, which will eventually impact all others. Investing in these technologies is not only important because it is for the right reasons, but because of this 'collision' it is valuable in the global marketplace. clare@AnswerALS.org