Genetically Targeted Therapies for Inherited Neuromuscular Disorders

It is an exciting era for inherited neuromuscular disorders (NMD), which, for a long time, have been well-defined phenotypically but have had limited treatments beyond supportive care. Inherited NMD are now becoming treatable conditions with breakthroughs in targeted genetic treatment. In this review, we discuss treatment of inherited NMD, including some promising potential therapies being researched.

Principles of Genetic Therapy

Genetic mutations can lead to downstream effects of increased, decreased, absent, or misfolded protein structure and function to cause clinical symptoms. Thus, when approaching therapy for genetic conditions, knowing the genotype and the downstream functional changes a pathogenic variant produces is crucial to applying optimal intervention.

Gene therapies can largely be broken down into 3 functional categories: 1) increasing gene expression; 2) decreasing gene expression; and 3) gene modification.¹ Disorders with a well-characterized, monogenic cause are best suited for genetic intervention. Disorders caused by a loss of function (eg, spinal muscle atrophy [SMA]), can be targeted by increasing gene expression. Disorders with a toxic gain of function, such as amyotrophic lateral sclerosis (ALS) caused by superoxide dismutase 1 (SOD1) mutations, can be targeted by decreasing mutant SOD1 gene expression.² Disorders with wide genetic and phenotypic heterogeneity, such as dystrophinopathies, can be targeted by genetic modification.

It is also important to distinguish gene therapy that affects gene expression from genome editing and somatic vs germline gene editing.¹ All commercially available genetic therapies at this time alter gene expression, affecting only the person treated and not their potential offspring or other family members. As with all genetic disorders, treatment should be paired with appropriate genetic counseling regarding family planning and risk for future generational inheritance, as well as impact on carrier relatives when relevant. The individual’s genome is (in theory) not changed and thus still harbors the pathogenic variant. Genome editing, either somatic or germline, is only done in the research setting at this time. Genome editing targets an individual’s DNA permanently by removing or correcting a pathogenic variant. Although genome editing is promising, as with all new technologies, there are important ethical considerations (eg, off-site targeting, uncertain long-term safety and durability, and medical equity).

The standard of care for most inherited NMD includes interdisciplinary screening and management of neurologic and other systemic comorbidities to maximize function and minimize symptoms. It is important to reinforce with patients and caregivers that follow-up must continue after receiving novel therapeutics. Reviews and guidelines often mention potential treatments on the horizon, but no guidelines that include targeted treatments are available at time of this review.^3-6

SMA

SMA is primarily a disease of the motor neuron in the anterior horn of the spine and brainstem, characterized by progressive flaccid weakness, atrophy, and diminished/absent reflexes. In the most severe forms, bulbo-respiratory insufficiency occurs. The most common form has infantile onset and a natural history in which the motor milestone of sitting is never achieved and death typically occurs by age 2 years unless feeding is supplemented and breathing is mechanically supported. Classic 5q SMA results from homozygous deletions or mutations of the survival motor neuron 1 (SMN1) gene on the long arm of chromosome 5. In humans, there are 2 homolog genes, SMN1 and SMN2, formed from an inverted duplication. Nucleotide changes that differentiate SMN2 from SMN1 create an exon splicing site (ESS) in SMN2 that suppresses transcription of exon 7 most of the time such that SMN2 creates only 10% to 20% as much functional SMN protein compared with SMN1. People carry from 0 (10%-15% of the population) to 4 or more SMN2 copies. When polymorphisms in SMN1 result in loss of functional protein from that gene, there is a dose-dependent genotype-phenotype correlation with increasing SMN2 copies resulting in a less severe disease course, although exceptions to this correlation do occur.⁷

Over the last 5 years, genetically targeted treatment of SMA has transformed the clinical picture of the disease. Both SMN1 and SMN2 expression have been targeted, and clinical discussions are now focused on disease stabilization and often improved function with treatment.

SMN2 Modulation

Nusinersen is an antisense oligonucleotide (ASO) that binds SMN2 pre-messenger ribonucleic acid (RNA),⁸ stabilizing transcription via increased inclusion of exon 7, which increases functional SMN protein production. Nusinersen is administered intrathecally as 4 loading doses over 2 months followed by maintenance dosing every 4 months throughout a person’s lifetime. Rare side effects include thrombocytopenia, coagulation abnormalities, and renal toxicity, all of which are monitored at each dose.⁸ Intrathecal access can be challenging in those with less severe SMA treated later in life if severe scoliosis has occurred, especially with prior spinal fusion. Lumbar puncture-related complications over time and repetitive exposure to sedation for those who require it affect tolerability.

A double-blinded, placebo-controlled, phase 3 study showed nusinersen improved event-free survival of children with SMA type 1 with events defined as death or permanent assisted ventilation.⁸ The initial trial data, now supplemented by information from longer open-label studies and studies in individuals with milder and later-onset forms of SMA, show nusinersen provides significant functional improvement compared with placebo or natural history.⁹ Some who received early treatment have had near-normal development.¹⁰ Real-world data also shows promise in groups unlike those studied in clinical trials.¹¹

Risdiplam, administered orally or via gastrostomy tube, is a selective SMN2 splicing modulator that increases production of functional SMN protein and is approved in the US for treatment of all forms of SMA in individuals age 2 months or more. Risdiplam is taken up by nervous system and systemic tissues alike and whether SMN repletion in peripheral tissues is of clinical significance is a subject of ongoing research. Risdiplam increased SMN protein levels in participants with SMA¹²; children with SMA type 1 showed improved event-free survival after 12 months of treatment compared with natural history.¹³ Post hoc analysis suggested improved motor milestone development,¹³ and data from additional trials in participants with phenotypically milder or presymptomatic SMA are accumulating via several research protocols. This small molecule therapy has been well tolerated in clinical trials. In preclinical animal studies, effects on male fertility and teratogenicity are observed; there is no data yet available in humans. Effects of nonadherence to daily dosing in the context of interrupted supply, noncompliance, or poor oral tolerance (eg, during viral illnesses) are also unclear at this time.

SMN Gene Replacement

SMA is the first disorder to have an approved systemic in vivo gene therapy in the US. Available since 2019, onasemnogene is administered via a 1-time adeno-associated virus (AAV)-9 vector,¹⁴ and is indicated as a 1-time intravenous dose in children age 2 years or less with genetically confirmed biallelic mutations in SMN1 (in Europe, it is approved for those who weigh 21 kg or less). Side effects include thrombocytopenia, thrombotic microangiopathy, elevated troponin-I, and elevated aminotransferases.¹⁵ To help mitigate some of the patient’s immune-based response to the viral load, concomitant oral daily steroids are given for at least a month,followed by a 4-week taper if safe to do so. Patients are required to have a baseline antiAAV9 antibody titer of at most 1:50 before administration.

In 15 patients aged 0.9 to 7.9 months administered a single dose of onasemnogene, longer survival, superior motor milestone achievement, and better motor function occurred compared with historical cohorts.¹⁴ In an open-label study, 13 of 22 children with infantile-onset SMA who had 2 copies of SMN2 and were treated with onasemnogene were independently sitting at age 18 months and had improvements in other markers of motor, respiratory, and bulbar function.¹⁶ Ongoing trials are evaluating if there is benefit of dosing after age 24 months with intrathecal gene transfer delivery; however, preclinical concerns for dorsal root ganglion toxicity led to a hold on 1 study.

AAVs have come to the forefront of gene therapy delivery systems for many reasons including that these are known not to cause disease in humans and varying strains have different tissue tropism.¹⁵ A limitation to the use of AAVs in the treatment of NMD is the small packaging size of approximately 5.0 kb, which includes the 2 T-shaped inverted terminal repeats.¹⁵ Although this does not affect SMA or other diseases in which the causative gene is less than 5 kb, for diseases caused by larger genes (eg, Duchenne muscular dystrophy [DMD]), the AAV vector cannot deliver the full-sized gene.

Future Directions for SMA

Real-world data on short- and long-term effects of all 3 approved SMA treatments continue to be collected. Combination therapy is also a reality, with research trials gathering data about risdiplam after prior gene transfer or nusinersen (JEWELFISH^a) and nusinersen after gene transfer (RESPOND^b). Such data will be critical for evaluating efficacy and tolerability of combination treatments and informing ethical and economic considerations of access to these high-cost treatments across the spectrum of SMA severity.

It is likely that a combination approach will eventually be tailored to the individual with SMA with optimized timing of intervention based on genotype and age at diagnosis as well as symptom burden. Ideally, a combination approach would be informed by novel biomarkers of disease severity and treatment response. Prenatal therapies are also under study. Considering that prenatal carrier screening and newborn screening both focus on SMN1 deletions, silent carriers of SMA and the 5% of people with SMA due to other genetic changes (eg, point mutations in SMN1) will be missed by screening and continue to present with symptoms that will require urgent diagnosis and referral for treatment. It also is important to consider and include the large population of individuals living with symptomatic SMA in treatment approaches.

DMD

DMD is a severe X-linked neurodegenerative disorder that is the most common muscular dystrophy of childhood, affecting about 1 in 3,500 male births.¹⁷ Natural history of DMD includes losing the ability to walk by age 13, progressive scoliosis with school-age onset, detectable and progressive cardiomyopathy in the preteen years, a need for pulmonary interventions during adolescence, and death in a person’s late teens or early 20s.¹⁷ Cognitive symptoms from anxiety and attention deficit disorder to more severe cognitive impairment and autism spectrum disorder are often present, although cognitive decline does not occur. There are also milder forms of dystrophinopathy, such as the Becker form defined by ambulation beyond age 16 years, a cardiomyopathy form without skeletal muscle weakness, and asymptomatic elevated creatine kinase levels.¹⁷ Before 2016, therapeutic options included optimized interdisciplinary supportive care and corticosteroids, which together may increase life expectancy into the fourth or, rarely, fifth decade of life for a larger proportion of individuals. The typical effect of steroids is an approximate 1 to 3 years’ slower progression of disease milestones, but this comes with a host of iatrogenic comorbidities, most prominently a higher risk of behavioral issues, linear growth suppression, obesity in younger persons, and eventually poorer bone health with risk for cataracts.¹⁷ Options include prednisone/prednisolone or deflazacort. The ideal steroid type and dosing schedule is still not firmly established. Individualized dose reductions related to side effects or family preferences, alternative dosing schedules (eg daily, weekend-only, or 10 days on and 10 days off), and heterogeneity in disease phenotype limit evenly matched cohort comparisons.¹⁸ Considering the side effect burden of corticosteroids and the modest decrease in time of disease course and life-limiting functional loss, patients and their families as well as provider teams have long hoped together for more therapeutic options.

Stop Codon Readthrough

Ataluren is a targeted therapy for the dystrophin (DMD) stop codon that accounts for approximately 10% of DMD. Ataluren is available in several countries, including many in the European Union, South Korea, and Israel, but is not approved in the US.

Exon Skipping

Eteplirsen is an ASO conditionally approved in the US in 2016. Eteplirsen targets DMD exon 51, causing it to be skipped (not translated), which is beneficial for the 14% of people with a DMD mutation for which this skipping restores the reading frame.¹⁹ Approval was based primarily on data from a phase 2b clinical trial in 12 boys who were treated for 24 weeks followed by an open-label extension. Notably, 2 participants lost ambulation soon after enrollment, suggesting they had a more severe phenotype.²⁰ Longer-term follow-up provided further evidence for efficacy compared with an historical cohort.²¹ The approval of eteplirsen remains controversial because it hinged on surrogate biomarker data of slightly increased dystrophin expression in muscle biopsies. The quantity and quality of dystrophin expression needed for clinically meaningful benefit remains ambiguous and is continuing to be studied.^22,23 Tolerability of eteplirsen has been similar in real-world use as in the trial with minimal side effect burden. In the clinical trial, there were rare side effects of balance disorder, vomiting, and contact dermatitis.²⁰ Studies are ongoing with sparse additional data presented in peer-reviewed publications, although an effect on slowing pulmonary decline has been shown.²⁴ Despite a broad indication on the prescribing information, eteplirsen has often been inaccessible to persons who were eligible.²⁵

Other exon-skipping therapies for DMD mutations are golodirsen, conditionally approved for persons with mutations in DMD exon 53 (8% of DMD),¹⁹ also based primarily on DMD expression data from muscle biopsies in a clinical trial. Viltolarsen, another exon 53-skipping treatment, and casimersen, an exon 45-skipping treatment, have also been approved. There are now clinically available exon-skipping treatments for approximately 25% to 30% of people with DMD mutations.

Intravenous administration of exon-skipping therapies weekly requires a central line (port) insertion in people with challenging peripheral access, which is common in people living with weakness—especially those who are nonambulatory. It is also challenging to determine when a therapy may no longer be beneficial later in the disease course. Because muscle fiber destruction begins prenatally in DMD, eventually muscle reserve is depleted to the point of replacement by fat and connective tissue. There may be a timepoint when some interventions’ efficacy is no longer detectable; the risk may also be higher with more comorbidities later in the disease. Defining a treatment response or failure is a challenge in an individual because there can be an initial “honeymoon phase” of a few months of functional improvement after initiation of steroids, but after that, the decline phase continues, although with a slightly less negative slope.¹⁷ With exon-skipping treatments, a slower decline is expected, but this is challenging to prove in an individual because their course without the intervention is unknown. No matter how well matched cohorts are based on genotype and phenotype, every person living with DMD has their own experience of the disorder. There are also confounders that can arise (eg, fractures or critical illness), with disuse atrophy and a new baseline after recovery that affects attempts to track clinical aspects of the disease or biomarkers of severity.

Gene Transfer With Microdystrophin or Minidystrophin

Because the size of the DMD gene precludes gene transfer with existing viral vector technology, several programs are underway that use tailor-made DMD regions felt to be the most critical for dystrophin function within the muscle fiber. Several potential concerns remain, such as cellular immunity issues. It is plausible that truncated dystrophin, including cellular forms, primes the cellular immune system in some DMD phenotypes, and tolerability concerns related to immune activation have arisen in multiple clinical development programs. Individuals with immunity to AAV are also not candidates for gene therapy with current technology; some investigators have raised questions about plasmapheresis and other measures to attenuate this issue, which may allow for repeat dosing if optimized. The truncated dystrophins are not fully functional. Thus, disease expression over time is expected, and penetrance and efficacy may vary across tissues. Durability is also a concern considering that the AAV genome generally does not integrate into the host genome, the micro/mini-DMD transgene would persist in host cells as an episome and not be replicated during mitosis. Thus, in any tissue with cell division or turnover (eg, mildly dystrophic muscle), the transgene could eventually become diluted or lost, although satellite cells might harbor some reserve. Alternatively, some have raised the possibility of the rare occurrence of DNA integration; if this happens at any significant frequency, it could raise the concern of altering the expression of endogenous, chromosomal genes.²⁶

Gene Repair or Editing

Clustered regularly interspaced short palindromic repeats-associated protein 9 (CRISPR/Cas9)-mediated genome editing a mouse model of DMD has been shown to partially restore dystrophin expression in cardiac and skeletal muscle. Clinical development programs are anticipated and short- and long-term safety, including potential for off-target effects will be closely monitored.

Challenges and Future Directions

Recent advances in SMA care are a trailblazing example, not just for NMD but for genetically targeted medicine in general. The exponential rise in genetically targeted treatment trials has been obvious over the past decades and continues to hold promise for rare and more heterogeneous genotypes and phenotypes. Genetically targeted treatments are actively being explored in clinical trials with many more on the horizon for several other NMDs (Table), including genetic ALS, limb-girdle muscular dystrophies, glycogen storage disease type II (Pompe disease), giant axonal and Charcot-Marie-Tooth (CMT) inherited neuropathies, and congenital myasthenic syndromes. In 2020, there was a sobering announcement of multiple deaths with higher doses in a gene transfer trial for myotubular myopathy for which highly encouraging efficacy outcomes had been seen; these deaths were postulated to be related to liver morbidity unique to myotubular myopathies.

Early diagnosis and referral for treatment is key in neurodegenerative disorders, and newborn screening, now in place for SMA in the majority of the US and several other countries, makes this more possible. The pace of clinical decline can be highly variable across the phenotypic spectrum, however, leading to questions of the optimal timing of intervention based on available biomarkers (eg, when to treat people with SMA who have >4 SMN2 copies).²⁷

Target tissue penetrance is an important consideration. For example, a potential ASO treatment of myotonic dystrophy failed in a clinical trial, and muscle penetrance was a concern.² Cardiac effects are yet unproven with exon-skipping treatments for DMD, and these are a significant contributor to morbidity. Central nervous system effects are also minimal with approaches used for DMD and only a minority of people with DMD have no cognitive concerns. Next-generation exon-skipping agents are under study for subpopulations of people with DMD. The balance between target tissue expression, including effect and durability of protein expressed vs toxicity of medications and potential off-target effects of the genetic material or the protein expressed, requires extensive additional research.

Efforts continue to support other aspects of NMDs with nongenetically targeted therapeutics. For example, there are active trials of myostatin inhibition for SMA and antifibrotics and mitochondrial function enhancers for DMD.

The quest to optimize genetic treatments will not be complete until a treated individual has no detectable signs or symptoms because of successful early identification and customized treatment. At present, we can affect some NMDs significantly, but our treated individuals continue to require long-term follow-up by specialized interdisciplinary teams for both health optimization and long-term monitoring for potential complications of treatment. We have entered an age of targeted therapies in the neuromuscular clinic; there are many unresolved clinical, ethical and economic questions that require extensive further research.