CHDI FOUNDATIONCHDI

Accelerating Therapeutic Development for Huntington's Disease

2020 Conference

CHDI’s 15th Annual HD Therapeutics Conference took place February 24 – February 27, 2020, in Palm Springs. This unique conference series focuses on drug discovery and development for Huntington’s disease, and draws participants and speakers from the biotech and pharmaceutical sectors as well as academia and research institutions. The conference is intended as a forum where all participants can share ideas, learn about new disciplines, network with colleagues and build new collaborative partnerships. We are indebted to all of the conference speakers, and especially grateful to those who are able to make their presentations available here for a wider audience.

Epigenetic control of myelination and functional regeneration

Seth A Ament, PhD University of Maryland School of Medicine

Progressive striatal gene expression changes and epigenetic alterations are a prominent feature of Huntington’s disease (HD), but the mechanisms by which these changes occur remain poorly described. I will describe recent and ongoing experiments in genetically precise mouse models of the HD mutation, addressing two key unanswered questions: First, to understand direct relationships between HTT and chromatin, we performed high-resolution chromatin immunoprecipitation and sequencing (ChIP-seq) to characterize HTT genomic occupancy in the mouse striatum. These experiments revealed reproducible HTT occupancy at thousands of genomic loci, many of which were differentially occupied in striatal tissue from HD knock-in mice versus wildtype controls. Second, to gain insight into cell type-specific disease processes, we sequenced the nuclear transcriptomes of thousands of cells from the striatum of HD knock-in mice versus wildtype controls. These experiments revealed transcriptional changes distributed across nearly all striatal cell types and single-cell trajectories from wildtype to disease-associated transcriptional states.

Top

Spatiotemporal dynamics of molecular pathology in ALS using spatial transcriptomics

Hemali Phatnani, PhD, New York Genome Center

Paralysis occurring in amyotrophic lateral sclerosis (ALS) results from denervation of skeletal muscle as a consequence of motor neuron degeneration. Interactions between motor neurons and glia contribute to motor neuron loss, but the spatiotemporal ordering of molecular events that drive these processes in intact spinal tissue remains poorly understood. We used spatial transcriptomics to obtain gene expression measurements of mouse spinal cords over the course of disease, as well as of postmortem tissue from ALS patients, to characterize the underlying molecular mechanisms in ALS. We identified novel pathway dynamics, regional differences between microglia and astrocyte populations at early time-points, and found perturbations in several transcriptional pathways shared between murine models of ALS and human postmortem spinal cords. Here, we will describe these findings and outline how we plan to expand these studies more broadly.

Top

Single cell approaches to explore immune cell mechanism in neurodegenerative disease

Suman Jayadev, MD, University of Washington Medical Center

Neurodegenerative disease pathogenesis is influenced by numerous factors that become increasingly relevant with age such as somatic genetic variation, loss of neural cell homeostasis, and environmental and experiential exposures. Identifying candidate cellular mechanisms that could be tractable targets for drug intervention requires not only characterizing the molecular pathways, but also the cells in which those pathways are disease relevant. Recent advances in cell type specific molecular phenotyping can further these efforts. As the role of immune dyshomeostasis in neurodegeneration becomes increasingly clear, so has the diversity of immune cells mediating the observed variety of immune dysfunctional responses. Our group leverages developing single cell approaches to characterize the repertoire of neural cells from normal to disease brain. Furthermore, through novel bioinformatics approaches we model molecular disease progression using single cell sequencing data from autopsy brain tissue allowing us to extract dynamic information about early disease events from “static” tissue.

Top

A reference striatal gene coexpression map for scalable evaluation of perturbation molecular signatures

X William Yang, MD, PhD, David Geffen School of Medicine at UCLA

A major challenge in studying the mammalian brain is its enormous complexity in terms of cell types and numbers, molecular interactions within and between cells, and dynamic changes through development, aging, and diseases. To begin understanding these processes, it is crucial to have tools that allow the molecular and cellular biology of the intact brain to be studied in a reproducible, interpretable, and scalable manner. Here, I will present a novel approach that involves the construction and visualization of a highly reproducible reference striatal gene coexpression map (CoExMap). We have annotated the stable striatal coexpression modules with both known and new data to uncover its biological underpinnings. Importantly, the utility of CoExMap is demonstrated by analysis and interpretation of large-scale perturbation transcriptomic signatures from genetic, viral, and pharmacological perturbation studies in both wildtype and Huntington’s disease models. This new systems biology tool has the potential to enhance the scale, reproducibility, and interpretability of transcriptomic molecular signatures in understanding mammalian brain biology and pathology, and in therapeutic discovery.

Top

Somatic expansion: A driver of disease and therapeutic target in myotonic dystrophy type 1 and Huntington disease

Darren G Monckton, PhD, University of Glasgow

Although the symptoms and downstream pathology in Huntington disease (HD) and myotonic dystrophy type 1 (DM1) appear to be very different, the underlying genetics of these two disorders are remarkably similar. Both disorders are caused by the expansion of an unstable CAG•CTG repeat beyond the ~35 repeats usually observed in the general population. In both disorders repeat length is positively correlated with disease severity and inversely with age at onset. The repeat tract is genetically unstable and is biased toward expansion, explaining the anticipation observed in both disorders. Likewise, the repeat is somatically unstable in each disorder, in a process that is expansion-biased, tissue-specific, allele-length- and age-dependent. Notably, very large expansions up to 1,000s of repeats are observed in striatal neurons of HD patients and in the skeletal muscle cells of DM1 patients. In approximately 5% of DM1 patients the CTG repeat expansion is interrupted by the presence of CCG variant repeats. These variants have dramatic stabilizing effects on expansion in both the germline and soma. Despite having no expected impact on the MBNL binding capacity of the toxic mutant RNA, variant repeat DM1 alleles are associated with a delayed age at onset. Likewise, CAA variants in the polyglutamine encoding CAG/CAA array are associated with increased somatic stability and a later age at onset in HD. Further, after correcting for age at sampling and pure repeat length, blood DNA somatic expansion scores for both DM1 and HD patients are inversely correlated with age at onset and positively correlated with disease progression. These data all strongly suggest that somatic expansion drives disease pathology in both DM1 and HD. Therefore, genetic modifiers of somatic expansion should modify disease severity. Notably, we have established that variants in exon one of the MSH3 DNA repair gene are associated with decreased gene expression, and decreased somatic expansion and delayed age at onset in both HD and DM1, highlighting this gene as potential therapeutic target in both disorders.

Top

A new potential target for modulating somatic instability identified in a mouse model of the fragile X-related disorders

Karen Usdin, PhD, National Institutes of Health

Accumulating evidence supports a common expansion mechanism for all repeat-expansion diseases. We have previously shown that many genetic modifiers of age at onset and disease severity in these disorders identified by GWAS also significantly affect repeat expansion when ablated in a mouse model of the fragile X-related disorders, including MSH3 and FAN1. This lends support to the idea that this mouse model is suitable for understanding human expansions. Our more recent work in this mouse model suggests a process in which MutLγ generates a double-strand break in the repeat tract. This break is then repaired by a mechanism that is independent of the major pathways of double-strand break repair, including nonhomologous end joining, Polθ-mediated end joining, and homologous recombination. We have now also identified another key factor in this process that may be a suitable target for modulating somatic instability in humans. This, together with other recent findings, suggests an interesting model for repeat expansion.

Top

MutSβ: Targeting a complex molecular machine

Ravi R Iyer, PhD, CHDI

Human genome-wide association studies have implicated multiple genes in the DNA mismatch repair (MMR) pathway in modulating the age at onset as well as the progression of Huntington’s disease (HD). Emerging data from studies of HD patients and animal models of the disease suggest that these effects are mediated, at least in part, by the role of MMR proteins in promoting somatic HTT CAG repeat expansion. We postulate that the attenuation of somatic CAG-repeat expansion by pharmacological inhibition of MMR proteins or genetic inactivation of the MMR machinery has the potential to delay the age at onset of HD and slow its progression. To test this hypothesis, CHDI has initiated a program in conjunction with five CRO partners to develop inhibitors targeting the MMR pathway, with an initial focus on MutSβ, a heterodimeric protein complex composed of MSH2 and MSH3 that recognizes CAG extrahelical extrusions and initiates an aberrant DNA repair process resulting in CAG expansion. Here we provide an overview of the program, lay out the rationale and potential liabilities of targeting MutSβ, and discuss our strategies to address this challenging target with the primary objective of identifying CNS-active compounds. We also present early findings from our experiments to address key knowledge gaps.

Top

Targeting DNA damage response genes with oligonucleotides for therapeutic modulation of HTT somatic instability

Brian R Bettencourt, PhD, Triplet Therapeutics

DNA damage response (DDR) genes are potent modifiers of the onset and severity of Huntington’s disease (HD), and act via modulation of DNA repeat somatic expansion at HTT and other diverse disease loci (e.g. DMPK, ATXN1/2/3/7). We developed antisense oligonucleotides (ASOs) and small interfering RNAs to knock down specific DDR genes selected for association with somatic instability phenotypes and loss of function tolerance. This approach operates upstream of disease gene-targeting approaches and targets a fundamental driver of multiple DNA repeat disorders. We measured the impact of DDR gene knockdown on repeat expansion over time in both HD patient-derived cell lines and HD model mice. Single doses of DDR targeting ASOs administered via multiple routes to non-human primates were well tolerated and drove significant knockdown in relevant brain regions. Repeat intrathecal dosing was also safe and well tolerated; we further characterized target knockdown under this dosing paradigm. In parallel with preclinical development, we initiated SHIELD-HD, a multinational natural history study in HD gene-expansion carriers, including prodromal and early-manifest individuals, to assess somatic expansion, DDR gene-expression, and various outcomes of disease progression over time. These activities will be discussed in the context of a coordinated development program seeking to stop HD manifestation and progression via halting somatic CAG-repeat expansion.

Top

Huntingtin protein acts in trans to regulate somatic instability of long CAG tracts

Jeff Carroll, PhD, Western Washington University

Huntington’s disease (HD) and several forms of spinocerebellar ataxia (SCA) are caused by expanded glutamine-coding CAG tracts, which are unstable during aging in specific cell types. Recent human genetic studies reveal that genes known to regulate somatic instability (SI) are also modifiers of age of onset in CAG-expansion disorders, suggesting a potential causal link. We have been investigating the hypothesis that huntingtin’s toxic gain-of-function effects emerge from perturbation amongst its endogenous cellular networks, rather than an entirely neomorphic etiology. If this hypothesis is correct, and if SI is a key driver of HD pathogenesis, we predicted that reducing HTT levels would alter SI in vivo. To investigate, we lowered HTT in knock-in mouse models of the HD mutation with multiple genetic and pharmacological interventions. Consistent with our motivating hypothesis, we find that lowering wildtype and/or mutant HTT leads to reductions in SI at the HTT locus in targeted central and peripheral tissues. To definitively determine whether this effect is occurring in trans, we chronically lowered wildtype HTT in the liver of Atxn2100Q/+ mice using antisense oligonucleotides (ASOs) – these results will be discussed. The most parsimonious explanation for HTT’s modulatory effect on SI is via alteration of the levels and/or activity of known SI modulator proteins – we will describe our ongoing transcriptomic and proteomic efforts in this direction. Our results suggest that the HTT protein plays a role in regulating SI of long CAG tracts in vivo and, to the extent that SI is a key driver of pathogenesis in other repeat diseases, that HTT lowering could be beneficial for multiple indications.

Top

Defining the course of Huntington’s disease with deep clinical phenotyping through the Enroll-HD platform

Swati Sathe, MD, MS, CHDI

Defining earlier biologic events in Huntington’s disease (HD) will allow for the generation and testing of hypotheses relating to the interaction between the underlying pathologic processes and disease progression. To facilitate earlier intervention in HD, it is imperative to identify biomarkers that unequivocally indicate pathologic onset, reliably predict progression of pathology, and correlate with clinical manifestations. Enroll-HD plays a vital role in identifying these key biomarkers critical to advancing clinical research. Enroll-HD is a multicenter, longitudinal observational study that has enrolled over 20,000 participants across four continents and over 160 sites. Given the magnitude of the study, it is poised to answer questions related both to biomarkers and clinical progression of HD and the relationship thereof. This talk will focus on the contributions of Enroll-HD and allied studies in the development of the HD research framework. First, the clinical observational data from Enroll-HD has led to several statistical and machine-learning models that allow for accurate clinical staging and prediction of progression. Second, platform studies nested within Enroll-HD augment the drive to identify biomarkers. The HD-Clarity study performs annual lumbar punctures to collect CSF samples, and the forthcoming imageClarity study will also annually collect state-of-the-art MRI imaging sequences on these participants. Besides making the CSF samples available to all interested parties for unbiased investigation, CHDI plans to perform hypothesisbased CSF analysis for known and novel CSF biomarkers to establish their utility in research and create a database that is accessible to researchers. ImageClarity will serve not only to detect a radiologic biomarker of disease progression but also to detect correlation with other (CSF) biomarkers. Finally, PET tracer studies, iMarkHD and iMagemHTT are designed to study sensitivity and specificity of novel mutant HTT-specific PET ligands in various stages of HD. These initiatives aim to provide much-anticipated breakthroughs in HD biomarker research and provide the missing link for refined clinical studies.

Top

Looking backwards to move forwards – the HD Young Adult Study

Sarah Tabrizi, BSc, MBChB, PhD, FRCP, FMedSci, University College London

With several therapeutic approaches in development for HD, identifying the ideal time to treat is essential if we are to optimally slow or prevent subsequent clinical decline. Previous large multi-site observational studies have highlighted disease-related differences in brain imaging, biofluid markers and even subtle functional impairment up to 15 years before expected disease onset. The Young Adult Study (YAS) examines the earliest premanifest HD adult cohort to date to establish whether there is a point at which gene carriers deviate from healthy controls. 64 young adult premanifest HD gene carriers, approximately 24 years from predicted clinical onset, and 67 matched controls underwent state of the art imaging, blood and CSF collection and detailed functional assessments, including a novel cognitive battery. I will report on the results from this unique cohort, describing the very first pathological events in adulthood. Identifying the point of deviation from a healthy state across these multiple domains will provide valuable insights into the time to initiate treatments to prevent neurodegeneration and thereby preserve quality of life.

Top

Integrated Huntington’s disease progression model

Jianying Hu, PhD, IBM Research

Huntington’s disease (HD) evolves over a long period of time with nonlinear progression and complex and multifaceted manifestation, making the task of modeling the progression of HD based on observational studies extremely challenging. We developed a framework to build probabilistic disease progression models that are based on continuous-time hidden Markov models (CTHMM) using longitudinal observational medical data. The framework consists of two steps. The first step empirically determines the number of disease states. The second step builds a probabilistic disease progression model with the determined number of states. The model discovers typical states along the trajectory of the target disease, learns the characteristics of these states in the form of multi-dimensional probability distributions, as well as transition probabilities between the states. We applied the framework to an integrated observational HD dataset curated from four
recent observational HD studies containing over 16,000 HD gene-expansion carriers and 2,000 controls. The resulting HD progression model, called the Integrated HD Progression Model (IHDPM), identified nine disease states along with the corresponding disease characterizing probability distributions and state transition probabilities. Compared to the existing HD staging system, the IHDPM 1) covers a wider range of HD progression, including prodromal, transitional and manifest phases; 2) is able to quantitatively describe complex multi-dimensional patterns of progression; 3) discovers multiple potential HD progression pathways; and 4) reveals the expected time durations of the identified states. Besides providing population level insights, the IHDPM can also be applied to individuals to provide a nuanced view of their progression history and better assessment of their current disease status. The IHDPM could be used to improve patient outcomes
in two ways: First, it could help improve the management and care of patients through better assessment of the patient’s current disease state and likely future trajectory, and provide insights into the most critical clinical measures at each stage of disease progression. Second, it supports more effective drug discovery by enabling objective baselines to assess the effectiveness of interventions, select the most appropriate cohort for a trial (trial enrichment), develop clinical trial simulators for trial design optimization, and identify biomarkers that could lead to new targets.

Top

Toward a biologic definition of Huntington’s disease: Insights from highly parallel single-cell analysis

Steven A McCarroll, PhD, Harvard Medical School/Broad Institute of MIT and Harvard

In Huntington’s disease (HD), as in many brain illnesses, we do not yet know the pathophysiological process – the series of molecular and cellular events by which CAG-repeat expansions in the huntingtin gene lead to disease. A critical need is to develop a biologic definition of HD that can enable preventative therapeutic interventions and biomarkers that accelerate clinical trials.

To address this unmet need in HD and other brain illnesses, my lab set out several years ago to develop a technology by which the molecular states of thousands of individual brain cells could be characterized simultaneously – at single-cell resolution, and in a systematic manner, for all genes and cell types at once. Our goal is to produce informative, systematic, quantitative data sets on the biological states of all of the cell types in a tissue. The technology we developed – droplet-based single-cell RNA-seq (Drop-seq) – now enables such analyses. This approach has already led us to surprising discoveries about the striatum, which is strongly affected in HD: we discovered an abundant type of spiny projection neuron (“eccentric” SPNs) that had eluded earlier research, and we found a novel type of interneuron that is abundant in the human striatum but absent from the mouse brain.

We and others recently adapted this technology to study human archival (fresh frozen) brain tissue, and we have begun to investigate what such analyses can teach us about cell-biological changes in HD. We have so far analyzed RNA expression in more than 400,000 individual nuclei sampled from the caudate nucleus of 17 HD gene-expansion carriers and 6 controls. We find that the loss of SPNs is already substantial in premanifest carriers of expanded HTT alleles; is almost total by late disease stages; and differentially affects distinct SPN subtypes early in HD. The data distinguish between cell-type-specific changes that are shared by all patients and changes that are present in only some patients despite their substantial SPN loss. I will also share ways in which such data can be used to nominate and characterize biomarkers of cell-typespecific biological changes, which we hope could inform clinical trials.

Top

Catalyzing translational innovation

Christopher P Austin, MD, National Center for Advancing Translational Sciences/National Institutes of Health

The process by which observations in the laboratory or the clinic are transformed into demonstrably useful interventions that tangibly improve human health is frequently termed “translation.” This multistage and multifaceted process is poorly understood scientifically, and the current research ecosystem is operationally not well suited to the distinct needs of translation. As a result, biomedical science is in an era of unprecedented accomplishment without a concomitant improvement in meaningful health outcomes, and this is creating pressures that extend from the scientific to the societal and political. To meet the opportunities and needs in translational science, NCATS was created as NIH’s newest component in December 2011, via a concatenation of extant NIH programs previously resident in other components of NIH. NCATS is scientifically and organizationally different from other NIH Institutes and Centers. It focuses on what is common to diseases and the translational process, and acts as a catalyst to bring together the collaborative teams necessary to develop new technologies and paradigms to improve the efficiency and effectiveness of the translational process, from target validation through intervention development to demonstration of public health impact. This talk will provide an overview of NCATS’ mission, programs, and deliverables, using Huntington’s disease as an exemplar and with a view toward future developments in drug discovery and translational medicine.

Top

HTT-lowering strategies – timing, biomarkers and key questions

Ignacio Munoz-Sanjuan, PhD, CHDI

HTT-lowering therapeutic agents are being explored clinically for the treatment of Huntington’s disease (HD). The HTT-lowering team at CHDI is enabling the development of effective therapies targeting HTT expression using various therapeutic modalities, as well as developing translational biomarkers to assess therapeutic effects in the adult brain. The presentation will address important questions relating to the timing of intervention, agent biodistribution, the cell-autonomous role of mHTT in driving specific dysfunctions in the striatum, and exploration of a number of translational biomarkers using PET in rodent models of HD, including the optimization of mHTT-selective PET-imaging agents. We have used the recently published zinc-finger repressor (ZFP) proteins developed by Sangamo Biosciences as our therapeutic agents in experimental model systems, as well as inducible HD mouse models. Questions about HTT loss-of-function and the impact of HTT-lowering strategies directed at HTT mRNA levels vs protein degradation will be raised and areas for future investigation highlighted.

Top

The subthalamic nucleus as a target for mHTT lowering in Huntington’s disease

Mark D Bevan, PhD, Northwestern University Feinberg School of Medicine

Although multiple, deleterious effects of mutant huntingtin (mHTT) species on neuronal function have been identified, the mechanisms that underlie the progressive dysregulation and degeneration of cortico-basal ganglia circuitry and expression of symptoms in Huntington’s disease (HD) remain poorly understood. To further address these issues, we applied state-of-the-art optical and electrophysiological interrogation approaches in symptomatic Q175-KI mice, focusing on neurons in the indirect pathway of the basal ganglia. Consistent with the susceptibility of this pathway in HD, we found that the in vivo activities of its component nuclei were highly dysregulated. GABAergic D2 receptor-expressing striatal projection neurons (D2-SPNs), GABAergic external globus pallidus (GPe) neurons, and glutamatergic subthalamic nucleus (STN) neurons were hypoactive, hyperactive, and hypoactive, respectively. However, the aberrant activities of GPe and STN
neurons in vivo were not simply the result of abnormal upstream D2-SPN activity because their intrinsic autonomous firing properties were persistently upregulated and downregulated, respectively. Furthermore, optogenetic inhibition of hyperactive GPe neurons did not fully restore STN activity in vivo. In the absence of evidence that current HTT-lowering therapeutics distribute uniformly throughout the human CNS, knowing which nodes of the cortico-basal ganglia circuit to target for clinical efficacy is paramount. As a first step towards this goal, we virally expressed zinc finger protein transcription factors targeting the pathogenic CAG repeat in STN neurons, to selectively repress the expression of mHTT. This treatment alleviated mitochondrial oxidant stress in STN neurons and restored their intrinsic autonomous firing to normal levels. Lowering STN mHTT also rescued the basal and movement-related activities of STN neurons in awake, behaving Q175-KI mice and reduced the locomotion deficits that these mice commonly exhibit. Together, these data demonstrate that despite widespread dysregulation of the indirect pathway, lowering mHTT in STN neurons is sufficient to ameliorate abnormal STN activity in vivo and ex vivo and improve motor function. Thus, our work argues that the cell-autonomous effects of mHTT in the STN contribute significantly to HD pathophysiology and, given its small size and key location within the cortico-basal ganglia circuit, the STN may represent a tractable and potent target for mHTT-lowering therapies in HD.

Top

HTT: The Outer Limits

Marcy E MacDonald, PhD, Massachusetts General Hospital

The unstable expanded CAG triplet repeat that is the root cause of Huntington’s disease is in exon 1 of the 67-exon huntingtin (HTT) gene, which spans ~185 Kb of DNA near the telomere of the chromosome 4 short arm. HTT specifies mRNAs of ~9.5 Kb and ~13.5 Kb, which both encode the same ~3,144 amino acid huntingtin (HTT) protein. Next generation sequence analysis of thousands of individuals from the general population reveals that HTT is highly constrained, with many fewer chromosomes than expected harboring deleterious loss of function mutations. Yet those rare individuals with one normal HTT chromosome and one chromosome with a null or a deleterious protein-altering mutation do not exhibit any abnormal phenotype. In the absence of any normal chromosome, the inheritance of two deleterious HTT loss of function mutations is associated with severe developmental abnormalities. Across many hundreds of control individuals, HTT mRNA levels vary widely in any given tissue, including regions of the brain. However, HTT eQTL SNPs tagging regulation of HTT expression levels are not associated with genome-wide significant modification of the age at onset of Huntington’s disease symptoms. Moreover, Huntington’s disease individuals who inherit two expanded CAG repeat HTT chromosomes exhibit timing of onset and disease-interval transit that are not significantly different from those Huntington’s disease individuals who have a single disease chromosome. These observations confirm the true dominance of the CAG expansion mutation and implicate upper and lower limits on HTT expression from conception that are compatible with life, whereas the effects of interfering with HTT in different tissues and at different stages of life may emerge from trials of HTTinterfering molecules tested as potential therapeutic agents.

Top

Challenges of gene-targeted therapeutics: Emerging insights from SMA

Charlotte J Sumner, MD, Johns Hopkins University School of Medicine

The splice-switching antisense oligonucleotide (ASO) nusinersen was FDA approved for use in all spinal muscular atrophy (SMA) patients in 2016. Clinical trials and commercial experience reveal substantial variation in treatment efficacy, ranging from normal attainment of early motor milestones, including sitting and walking, to no change in motor function. Discrepancies of response of specific spinal segments have also been observed. Although the molecular and cellular mechanisms underlying these variations are poorly defined, the timing of drug initiation is critical as drug administered neonatally at a presymptomatic stage can be markedly more effective than postsymptomatic delivery. To begin to understand the molecular and cellular factors governing drug efficacy, we have analyzed survival motor neuron (SMN) protein levels and SMA pathology in tissues from patients and age-matched controls isolated during expedited autopsies, and also in SMA mice over developmental time. These studies highlight declines of SMN expression during the perinatal period and the caudal to rostral gradient of nusinersen and SMN levels after lumbar intrathecal delivery, which may limit drug efficacy. Our studies also demonstrate severe impairments of SMA motor axon development that begin prenatally and are followed by fulminant neonatal degeneration. Optimal therapeutic reversal requires prenatal drug treatment in SMA mice. Together these studies highlight some of the challenges that may be associated with treatment of neurodegenerative diseases, including HD, with intrathecally-delivered ASOs.

Top

RNAi-based modulation of huntingtin expression in rodents and NHPs

Anastasia Khvorova, PhD, University of Massachusetts Medical School

The talk will cover the discovery and development of a novel di-valent siRNA scaffold, which achieves potent (95%), sustained (> 6 months), and highly specific (RNAseq) HTT silencing throughout rodent and NHP brains, including the deep brain regions putamen and striatum that are essential in the clinical manifestation of Huntington’s disease (Alterman et al, Nature Biotechnology, 2019). Chemical engineering (backbone constraining (Yamada et al, under review, 2020) in combination with chemically defined mismatches) allow allele-specific modulation of huntingtin in vivo in rodent HD models (Conroy et al, in preparation, 2020). At least 150x in vitro selectivity is necessary for in vivo selective modulation. We will discuss the potential safety implications of close-to-complete selective and non-selective HTT lowering in clinical translation as a novel HD therapy.

Top

Translatable biomarkers in gene therapy for Huntington disease: Innovative approaches and learnings from pre-clinic to the clinic

Astrid Vallès-Sánchez, PhD, uniQure

Huntingtin (HTT)-lowering therapies hold great promise to slow-down or halt neurodegeneration in Huntington disease (HD). From the different approaches under development, gene therapies using adenoassociated viral vectors (AAVs) are typically administered locally into the brain region of interest, after which they expand to interconnected regions though different mechanisms. A single administration should be sufficient to ensure long-term persistence of the therapeutic transgene (i.e. lowering agent), especially in non-dividing cells such as neurons. We have developed AAV5-miHTT, a recombinant AAV-based gene therapy expressing an engineered microRNA that specifically binds to HTT exon1, resulting in lowering of both full-length and exon1 HTT mRNA expression. Now on its way to clinical development, AAV5-miHTT has demonstrated safety and efficacy in small and large animal studies, with remarkable brain-wide spread and long-term persistence, through retrograde and anterograde transport of the AAV followed by extracellular spread of the therapeutic miRNA, and without off-target effects. Adequate translational measures to evaluate the safety, efficacy and durability of HTT lowering mediated by AAV5-miHTT in patients are much needed. To this end, we assessed the response of candidate biofluid and imaging biomarkers in preclinical models (from rodents to minipigs and non-human primates) to AAV5-miHTT administration. In cerebrospinal fluid (CSF), assessments included pharmacokinetic (miHTT expression) and pharmacodynamic (HTT protein, NFL) measures. Imaging biomarkers included volumetric magnetic resonance imaging (vMRI) and magnetic resonance spectroscopy (MRS). Many of these measures have the potential to follow-up safety and efficacy of HTT-lowering therapies in general. However, because of the unique properties of gene therapy approaches in contrast to other HTT-lowering therapies, a tailored biomarker panel for HTT-lowering gene therapies in HD patients may be needed. Data supporting this customized selection will be presented, in light of the known mechanism of action of AAV5-miHTT.

Top

Preliminary results from a 15-month open-label extension study investigating tominersen (RG6042) huntingtin protein antisense oligonucleotide in adults with manifest Huntington’s disease

Scott A Schobel, PhD, F. Hoffman-La Roche Ltd

Long-term safety, tolerability and biomarker effects of the HTT-lowering ASO RG6042 were evaluated in a 15-month open-label extension (OLE) study (NCT03342053) involving all 46 patients from the Phase I/IIa study (NCT02519036). Patients in the OLE were randomised to receive 120 mg RG6042 monthly (Q4W) or every 2 months (Q8W). At time of assessment, 43 of 46 participants had reached the 15-month visit time point. Over 500 doses of RG6042 have been administered, with all patients having received ≥8 doses Q4W or ≥6 doses Q8W. RG6042 is well tolerated over >1 year of treatment. There were fewer adverse events (AEs), serious AEs and potentially drug-related AEs in the Q8W arm versus the Q4W arm.

At 15-months, mean (standard deviation) lowering of trough (i.e. pre-dose) cerebrospinal fluid (CSF) mutant HTT (mHTT) levels were 69.9% (19.8%) in the Q4W arm and 44.2% (23.5%) in the Q8W arm, based on preliminary findings from an exploratory assay. These CSF reductions meet or exceed the levels associated with reductions of mHTT in brain tissue in HD mouse models that were associated with phenotypic benefit. CSF neurofilament light protein increased in both arms early in the study then decreased on continued treatment, with values in the Q8W arm returning to expected levels by 15 months.

Preliminary 15-month data from the OLE study suggest that Q8W dosing achieves pharmacologically relevant mHTT lowering and is suitable for chronic dosing in patients with HD. These data support the current development regimens of 120 mg RG6042 Q8W or Q16W (every four months), which are being tested in the randomised, double-blind, placebo-controlled GENERATION HD1 study in manifest HD.

This study was funded by Ionis Pharmaceuticals Inc and F. Hoffmann-La Roche Ltd. Medical writing was provided by MediTech Media UK and funded by F. Hoffmann-La Roche Ltd. genetic study of such symptoms is therefore warranted, with more refined definitions.

Top