Antisense oligonucleotides for neurodegeneration

See allHide authors and affiliations

Science  27 Mar 2020:
Vol. 367, Issue 6485, pp. 1428-1429
DOI: 10.1126/science.aba4624

Antisense oligonucleotides (ASOs) have the potential to reduce, restore, or modify RNA and protein expression. Thus, they can target disease pathogenesis by altering the expression of mutant proteins (1). The recent regulatory approval of ASOs for the pediatric motor neuron disease spinal muscular atrophy has provided a regulatory pathway for additional ASO therapies in other central nervous system (CNS) diseases. Developments in ASO chemistry and advances in CNS delivery methods have enabled ASOs to enter clinical trials to treat Huntington's disease (HD). There are currently no available treatments that slow or prevent progression of HD, but two ongoing ASO-based clinical programs have shown promising results. Additionally, clinical trials of ASOs to treat amyotrophic lateral sclerosis (ALS), Parkinson's disease, and Alzheimer's disease are under way, with more in development for other neurodegenerative diseases. It is hoped that ASO-based approaches will provide effective diseasemodifying therapies for HD and similar neurodegenerative diseases soon.

ASOs are synthetic single-stranded DNA analogs, generally 16 to 22 bases long, that selectively bind to specific complementary RNA targets. A limitation of the original ASOs developed for clinical use was susceptibility to rapid degradation by cellular nucleases, but chemical modifications have since been introduced to improve their therapeutic utility. For example, substitution of sulfur for oxygen and modification of the 2′-position of the sugar to generate 2′-O-methoxyethyl (MOE)–modified ASOs with a phosphorothioate backbone resulted in improved nuclease resistance, potency, and better tolerability in patients. Further modifications of the ribose sugar moiety have led to improved efficacy by improving binding to target RNAs (2).

ASOs can modulate target gene expression through numerous pathways (2). One such pathway is through ribonuclease H1 (RNase H1) recruitment (see the figure). Following selective binding of the ASO to its target RNA, an RNA-DNA hybrid is formed, which induces messenger RNA (mRNA) degradation by RNase H1. Other pathways depend on the specific location of ASO binding to target RNA (2). For example, ASOs can target intron-exon junctions in precursor mRNA (pre-mRNA) to modulate RNA splicing. ASOmediated target suppression can be achieved by modulating splicing to introduce an outof- frame deletion, which results in reduced protein expression by nonsense-mediated decay of the corresponding transcript. ASOs targeting translation start sites in RNA can block the binding of ribosomes, leading to complete translational inhibition of target protein synthesis.

There are several advantages of ASOs over related RNA interference approaches. Unlike interfering RNAs, ASOs are readily taken up by neurons and have clear dosedependent and reversible effects. ASOs also have the advantage that they will not saturate endogenous microRNA pathways, a potential cause of toxicity in short inhibitory RNA (siRNA)–based approaches. ASOs are generally highly selective and can target both introns and exons because they bind to pre-mRNA rather than mature mRNAs, allowing selection of specific target sequences for ASOs that do not appear anywhere else in the genome. However, unlike viral-mediated siRNA approaches, repeated administration of ASOs is required to maintain therapeutic effects. This may also be an advantage: If an unwanted outcome occurs from suppression of target RNA (or perhaps off-target RNA), ASOs have an off-switch because their effects are fully reversible.

HD is an inherited autosomal dominant neurodegenerative disorder characterized by a triad of motor, cognitive, and psychiatric features. HD typically arises in midlife, with inexorable progression of disability over 10 to 15 years. HD is caused by an abnormally expanded CAG repeat in one allele of the huntingtin (HTT) gene, which is expressed as a long polyglutamine tract in the mutant protein (mHTT) that confers a toxic gain of function (3). Proposed pathological mechanisms caused by this alteration include early transcriptional dysregulation, synaptic dysfunction, altered axonal vesicular trafficking, impaired proteostasis, mHTT aggregation, defective nuclear pore complex and nuclear-cytoplasmic transport, oxidative damage, mitochondrial dysfunction, and extrasynaptic excitotoxicity (3). ASOs provide a direct approach to reduce mHTT expression by targeting its RNA for destruction, thus preventing translation of mHTT and proximally targeting the primary cause of disease (1). Two ASO-based therapeutic programs have recently entered clinical testing for the treatment of HD.

The HTT-targeting ASO RG6042 acts through RNase H1 to target both wild-type and mutant HTT pre-mRNA and results in HTT lowering. Preclinical studies of similar HTT-targeting ASOs in transgenic HD mouse models demonstrated decreased mHTT concentrations in brain tissue, correction of striatal gene transcriptional dysregulation, and phenotypic improvement (4). mHTT lowering was prolonged following even a single cerebral spinal fluid (CSF) injection of ASO in these mice, suggesting that the effects of these ASOs will be prolonged in HD patients. Lumbar intrathecal infusion of a similar ASO into nonhuman primates was also shown to effectively lower HTT in many brain regions relevant for HD pathology (4).

In the initial phase 1/2a trial of intrathecal RG6042, treatment of 46 patients resulted in a significant dose-dependent reduction in CSF concentrations of mHTT by 40 to 60% (5). The CSF concentrations of mHTT continued to decline during this short study, suggesting that maximal reduction was not reached. The amount of CSF mHTT reduction observed in this study is consistent with the reductions in mHTT required for significant phenotypic improvement in transgenic mouse models of HD (4). Intrathecal delivery of RG6042 ASO was safe and well tolerated in HD patients, and its potential effects on disease modification and clinical outcomes are being assessed as part of GENERATION HD1, a large phase 3 trial involving more than 800 early-stage HD patients (NCT03761849).

Selective lowering of mHTT is theoretically an attractive approach to HD therapy because it would overcome concerns about the potential loss of wild-type HTT function. Selective mHTT-lowering ASOs that target specific single- nucleotide polymorphisms (SNPs) linked to the CAG expansion show promise in preclinical models of HD (6). A chemistry platform has been developed that allows the chirality of the phosphorothioate modification to be controlled during ASO synthesis. It is reported that “stereopure” ASOs have improved activity, stability, and specificity compared with stereoisomer mixtures (7). Two clinical trials were initiated in 2017 to assess two different stereopure ASOs targeting specific HDassociated SNPs. These studies require precise genotyping to ensure that the targeted SNP is accurately phased to the HTT allele with the CAG expansion. Not all individuals with HD can be treated with this SNP-based approach, and it is estimated that the two ASOs being tested are applicable to ∼65% of HD patients from North America or Europe. Each ASO is given as a monthly intrathecal bolus over three consecutive months, and it was recently reported that at the highest dose tested, one of the ASOs caused a modest 12.4% reduction in CSF concentration of mHTT. An additional higher-dose cohort will now be added to this trial (NCT03225833).

Reducing pathological protein expression

Antisense oligonucleotides (ASOs) are small, single-stranded DNAs that can bind specific RNA sequences on precursor messenger RNAs (pre-mRNAs) and mRNAs. The resulting RNA-DNA hybrid can induce ribonuclease H1 (RNase H1) degradation of the targeted RNA, modulation of splicing, or blockade of translation.

It is not known whether HTT lowering or mHTT-selective lowering will be most effective. Both approaches have distinct strengths and limitations, and this question will ultimately be answered empirically in clinical efficacy trials. Preclinical studies in a humanized transgenic HD mouse model found that the benefits of ASO-mediated lowering of total HTT concentrations by 75% were similar to those of ASOs that selectively reduced mHTT. These and other studies have suggested that the degree of mHTT lowering is the most critical parameter for preclinical efficacy (8). A substantial advantage of HTTlowering approaches compared to SNP-based selective mHTT targeting is the potential to develop a single therapeutic agent for the entire HD population.

There is optimism that HTT-targeting ASOs may lead to a viable disease-modifying therapy for HD, as well as the development of ASOs for other neurodegenerative diseases associated with aberrant protein production. Following a substantial preclinical development program, an ASO targeting superoxide dismutase 1 (SOD1) was found to be safe and well tolerated after lumbar intrathecal infusion in a phase 1 trial for the treatment of SOD1 mutated ALS (9). A more potent MOE-modified ASO (ISIS-SOD1Rx) is being evaluated in a phase 1/2a clinical trial (NCT02623699). Promising preclinical data have also been generated using ASO-based approaches for Parkinson's disease, targeting leucine-rich repeat kinase 2 (LRRK2, NCT03976349) and α-synuclein (10); for Alzheimer's disease by targeting Tau protein (11), which is currently in a phase 1 clinical trial (NCT03186989); and for prion diseases by targeting the prion protein PRP (12).

ASO-based therapies are also of interest in diseases whose etiology is similar to that of HD, such as the polyglutamine protein–related forms of spinocerebellar ataxia (13) that are caused by polyglutamine inclusions, and in frontotemporal dementia (FTD) with Tau or TAR DNA-binding protein 43 (TDP- 43) pathologic inclusions. Indeed, ASOs that lower ataxin 2 expression have shown benefit in mouse models of both spinocerebellar ataxia 2 and TDP-43–related FTD (14). The most common genetic cause of ALS and FTD is a GGGGCC repeat expansion in the C9ORF72 gene that induces RNA-mediated neurotoxicity. ASOs that selectively target these repeat-containing RNAs may be a useful therapeutic approach to this class of diseases (15); clinical trials are in development.

ASOs have already changed the landscape of therapeutic development for neurodegenerative diseases. Their advancement in the clinic will require continued development and research, including optimization of target sequence selection, improving biological activity, testing new delivery technologies, and maintaining an appropriate safety profile. Improving the delivery of ASOs to target cells is an important area of future development, including intrathecal pumps and the use of lipid-based and polymer-based nanocarriers. Such delivery systems will potentially improve the controlled release of ASOs and cell and tissue specificity, and may provide additional protection from nuclease degradation. Beyond potency and specificity, another crucial feature of a good candidate molecule is the ability to reach its intracellular target at sufficient concentration. Given the substantially increased potency of MOEmodified ASOs, there are safety concerns regarding excessive on-target lowering of proteins (such as wild-type HTT) and potential off-target effects. It is critical that extensive preclinical assessment of both potency and off-target effects is performed in the development of new ASO therapies.

The ASO RG6042 is the result of more than a decade of extensive preclinical assessment in multiple model systems and is a testament to effective academic and industry collaboration in drug development. With a growing number of ASO therapeutics being tested in clinical trials, this exciting technology holds the potential to change the therapeutic landscape for many neurological and non-neurological conditions (including cancer, and cardiovascular, infectious, and pulmonary diseases) in the near future.

References and Notes

Acknowledgments: Both authors have undertaken paid consultancy regarding ASOs for Ionis Pharmaceuticals, Takeda Pharmaceuticals, and F. Hoffman-La Roche. B.R.L. is cofounder of Incisive Genetics Inc. and has associated patents pending.
View Abstract

Stay Connected to Science

Editor's Blog

Navigate This Article