Recent developments in the treatment of Duchenne muscular dystrophy and spinal muscular atrophy

Abstract

Pediatric neuromuscular disorders comprise a large variety of disorders that can be classified based on their neuroanatomical localization, patterns of weakness, and laboratory test results. Over the last decade, the field of translational research has been active with many ongoing clinical trials. This is particularly so in two common pediatric neuromuscular disorders: Duchenne muscular dystrophy and spinal muscular atrophy. Although no definitive therapy has yet been found, numerous active areas of research raise the potential for novel therapies in these two disorders, offering hope for improved quality of life and life expectancy for affected individuals.

Introduction

Pediatric neuromuscular disorders represent a heterogeneous group of clinical conditions that can be classified into four major categories, based on their neuroanatomical localization. These are (1) motor neuron diseases, (2) neuropathies, (3) disorders of the neuromuscular junction, and (4) myopathies. In addition to their effect on skeletal muscle strength, neuromuscular disorders are often multisystemic as they can affect other forms of muscle (cardiac and smooth muscle) and other organ systems, potentially leading to cardiovascular, pulmonary, nutritional, and orthopedic complications.

The urgent need and drive towards the development of novel therapeutic approaches in pediatric neuromuscular disorders is exemplified by two common conditions that are frequently encountered in the pediatric neuromuscular clinic: Duchenne muscular dystrophy and spinal muscular atrophy. These two conditions can lead to progressive muscle weakness and atrophy, resulting in severe limitations of motor abilities. There have been many advances in biomedical sciences over the last few decades resulting in improved care, quality of life, and life expectancy for both disorders, but the only pharmacologic treatment that has been demonstrated to alter the natural history of these diseases to date is corticosteroid therapy for Duchenne muscular dystrophy.

Duchenne muscular dystrophy

Overview

Duchenne muscular dystrophy is a severe form of childhood muscular dystrophy. This is a recessive X-linked inherited disorder primarily affecting skeletal and cardiac muscles. It affects one in 3600–6000 live male births [Bushby et al. 2010a]. Affected individuals commonly present with mildly delayed motor milestones during the toddler years, often with toe-walking, difficulty rising from the floor, and frequent falls. They have proximal muscle weakness, resulting in the classic Gowers maneuver when rising from the floor. By the age of 5 years, there is usually a marked discrepancy between their physical abilities and that of their peers. Affected boys lose ambulation by the second decade, occasionally earlier [Biggar et al. 2006; Bushby et al. 2010b]. Respiratory, cardiac and orthopedic complications also emerge around this time. Without significant interventions, the life expectancy in affected individuals is often less than two decades. Recent advances in the supportive and medical therapies to address the associated multisystemic complications have helped to improve the quality of life and lengthen the life expectancy in many Duchenne muscular dystrophy patients into the fourth decade, and occasionally even into the fifth decade.

Duchenne muscular dystrophy occurs as a result of mutations in the dystrophin gene (DMD, locus Xp21.2) [Koenig et al. 1987; Monaco et al. 1986]. The dystrophin gene is a large gene which encompasses 79 exons over 2.6 million genomic base pairs with a cDNA length of approximately 14 kilobases [Muntoni and Wood, 2011]. It is expressed in all muscle types (skeletal, smooth and cardiac muscle). The most common mutations in Duchenne muscular dystrophy involve out-of-frame deletions of one or more exons. The resultant effect is an absence of dystrophin, an important cytoskeletal muscle protein product [Hoffman et al. 1987], leading to the Duchenne muscular dystrophy phenotype. A better understanding of the pathophysiologic cascade resulting in myofiber degeneration and impaired regeneration is important to generate novel therapeutic approaches.

Consensus statements regarding standards of care have recently been published for Duchenne muscular dystrophy patients, recommending early commencement of steroid treatment, though the optimal age of initiation remains controversial, as does the optimal choice of steroid and dosing regimen [Bushby et al. 2010a, 2010b; Sejerson and Bushby, 2009; Wang et al. 2007].

Significant progress has been made over the last few years towards developing therapies to target the pathology of Duchenne muscular dystrophy or improve muscle growth and regeneration (Table 1). These can be broadly classified into cell-based therapy, gene therapy, antisense oligonucleotide therapy, myostatin inhibition, stop codon read-through therapy as well as therapy targeting growth factor pathways. A recent study in zebrafish suggests that phosphodiesterase inhibitors may also show promise as therapeutic agents [Mazzone et al. 2010]. The optimal mode of delivery in targeting all of the muscles in the body is an issue that vexes researchers looking at various novel approaches [Muntoni and Wood, 2011].

Table 1.

Status of therapeutic development in Duchenne muscular dystrophy.

	Preclinical	Phase I	Phase II	Phase III
Cell-based therapy	√ [Blau, 2008]	√ [ClinicalTrials.gov identifier: NCT01610440]	√ [ClinicalTrials.gov identifier: NCT01610440]
Gene therapy	√ [Odom et al. 2008]	√ [ClinicalTrials.gov identifier: NCT00428935]
Antisense oligonucleotide • Targeting exon 51 • Targeting exon 44	√ [Muntoni and Wood, 2011] √ [Muntoni and Wood, 2011]	√ [ClinicalTrials.gov identifier: NCT01128855] √ [ClinicalTrials.gov identifier: NCT01037309]	√ [ClinicalTrials.gov identifier: NCT01153932] √ [ClinicalTrials.gov identifier: NCT01037309]	√ [ClinicalTrials.gov identifier: NCT01254019]
Stop codon read through • Gentamicin • PTC124	√ [Barton-Davis et al. 1999] √ [Welch et al. 2007]	√ [ClinicalTrials.gov identifier: NCT00451074] √ [Hirawat et al. 2007]	√ [ClinicalTrials.gov identifier: NCT00264888]
Growth factor pathways • Insulin-like growth factor • Myostatin • Transforming growth factor- β1	√ [Barton et al. 2002] √ [Wagner et al. 2002] √ [Cohn et al. 2007]	√ [ClinicalTrials.gov identifier: NCT01207908] √ [ClinicalTrials.gov identifier: NCT00104078]	√ [ClinicalTrials.gov identifier: NCT00104078]

The choice of animal model is also a vexing one, as a number of options exist, and it is not clear which one best recapitulates the human disease. The mdx mouse, which harbors a naturally occurring mutation in exon 23 of murine dystrophin [Bulfield et al. 1984; Ryder-Cook et al. 1988; Sicinski et al. 1989], is the most popular model for Duchenne muscular dystrophy. However, this mouse does not display weakness to the degree that is seen in the human disease. The reason for this is unclear, although recent data suggest that a discrepancy in telomere length between the species may be responsible [Sacco et al. 2010]. Other mouse models include the mdx^5cv, which has an artificially induced mutation in exon 10 [Chapman et al. 1989]; and mdx mice that are simultaneously deficient in utrophin [Deconinck et al. 1997; Grady et al. 1997]. Zebrafish [Berger et al. 2010; Guyon et al. 2009] and dog [Cooper et al. 1988; Sharp et al. 1992] models of muscular dystrophy also have some advantages for various studies. The dystophin-deficient dog has the advantage of reflecting many aspects of the human condition; however, variability of phenotypes, difficult statistical elaboration and limited genetic tractability make the dog model a less than perfect dystrophic animal model. Over the last decade, zebrafish models of muscular dystrophy have become increasingly popular due to the following attributes: compact size, proliferative capacity, genetic tractability, and transparent embryos that are easily manipulated [Hua et al. 2011]. Although there are obvious physiological and anatomical differences between zebrafish and humans, zebrafish offer many advantages in identifying disease genes and investigating potential therapies.

Preclinical and clinical studies

Cell-based therapy

Stem cell therapy has been studied for the potential treatment of Duchenne muscular dystrophy since the 1990s. Early human studies involving myoblast transplantation showed some promise [Gussoni et al. 1992], but the outcomes of multiple human clinical trials were disappointing [Huard et al. 1992; Karpati et al. 1993; Mendell et al. 1995; Miller et al. 1997; Tremblay et al. 1993]. Ongoing research efforts seek to identify the ideal stem cell population that will facilitate efficient muscle regeneration in the treatment of Duchenne muscular dystrophy. In theory, stem cells used to treat Duchenne muscular dystrophy need to fulfill certain criteria. These include the ability to expand in vitro without losing myogenic potential. An effective and efficient mode of delivery is needed to enable proliferation and migration within the host muscle; stem cells have to replace or repair damaged muscle fibers. Finally, new muscle fibers must be capable of expressing dystrophin, leading to an improvement in muscle strength [Meng et al. 2011b]. Various pluripotent cell populations have been examined to find one with the greatest myogenic potential. Stem cell populations that have been studied are derived from different regions of the body at various stages of development. These include embryonic stem cells [Stillwell et al. 2009], satellite cells, muscle-derived stem cells [Meng et al. 2011a; Sacco et al. 2008], side population (SP) cells [Gussoni et al. 1999; Jackson et al. 1999; Lee et al. 2000; Luth et al. 2008], bone marrow-derived stem cells [Cossu and Molinaro, 1987; Ferrari et al. 1998], mesoangioblasts [Galvez et al. 2006], blood and muscle-derived stem CD133+ stem cells, adipose-derived stem cells [Goudenege et al. 2009], and pericytes [Blau, 2008; Fariniet al. 2009; Meng et al. 2011b; Muntoni and Wood, 2011]. Patient-specific pluripotent stem (iPS) cells have shown potential as sources for autologous cell transplantation therapy in light of their ability to proliferate in vitro and to differentiate into multiple cell lineages both in vitro and in vivo [Mazzone et al. 2010].

There have been two reports to date of humans with Duchenne muscular dystrophy who received stem cell transplants for immunodeficiency syndromes. Both transplantations were well tolerated and cured the immunodeficiency syndromes. However, they did not show any significant increase in dystrophin expression or muscle strength [Gussoni et al. 2002; Kang et al. 2010]. A phase I/II clinical trial to assess the safety and efficacy of human stem cells to treat Duchenne muscular dystrophy is being planned [ClinicalTrials.gov identifier: NCT01610440]. This trial aims to study the effects of human umbilical cord mesenchymal stem cell transplantation on muscle strength and motor function of boys aged 5–12 years with Duchenne muscular dystrophy.

Current obstacles include isolating a sufficient quantity of a specific stem cell population for the purposes of a therapeutic transplantation, finding an optimal mode of delivery, and maximizing engraftment of donor cells.

Gene therapy

Significant progress has been made in the field of gene therapy. Different classes of viral vectors have been researched extensively to find the ideal vector that would be safe, readily made and easily administered for gene therapy. One of the challenges faced by gene therapy researchers includes the large size of the dystrophin cDNA. Originally, adenoviruses were used as vectors for gene delivery, but immunogenicity was found to be a major problem, most notably in the Jesse Gelsinger case of 1999 [Marshall, 1999]. Since then, recombinant adeno-associated viral (rAAV) vectors have shown promise as gene transfer vehicles due to their lower immunogenicity, especially certain serotypes that have high tropism for skeletal muscle [Odom et al. 2008]. One challenge is the amount of DNA that such a viral vector is able to carry. The large size of the dystrophin gene makes it too large to fit into a rAAV vector. Smaller versions of dystrophin (mini- and micro-dystrophins) have been developed to address this problem [Scott et al. 2002].

A recent clinical trial assessed the efficiency of a functional dystrophin transgene using rAAV vector that was administered via intramuscular injection into six Duchenne muscular dystrophy patients. The study found that though there was dystrophin transgene expression observed within a month of treatment, there was difficulty in establishing long-term transgene expression. This led to an analysis of T-cell immune responses to the dystrophin transgene, and a major finding of that study was that T-cell immunity against rAAV was present, sometimes prior to treatment [Mendell et al. 2010a]. This could at least partially explain the low levels of dystrophin transgene expression that were observed after administration of the viral vector. The influence of cellular immunity on the outcome of gene therapy is not clearly defined and this may present a major obstacle to further experimental therapy designed to increase dystrophin expression in Duchenne muscular dystrophy.

Utrophin is an autosomal paralog of dystrophin that is usually expressed at the neuromuscular junctions within skeletal muscle fibers. Utrophin can functionally substitute for the lack of dystrophin in animal models [Tinsley et al. 1998]. A microutrophin cassette packaged within a rAAV vector demonstrated improvement of muscle contractile properties with increased lifespan in dystrophin^-/-/utrophin^-/- double-knockout mice. These findings were comparable to that of a structurally similar microdystrophin using a rAAV-mediated delivery model [Odom et al. 2008]. Given that utrophin is normally expressed in skeletal muscles, this approach may be a promising treatment option for Duchenne muscular dystrophy as it avoids the potential immune responses that are associated with exogenous dystrophin.

Gene therapy approaches have also been used successfully in mouse models of other muscular dystrophies such as limb-girdle muscular dystrophy 2D (α-sarcoglycanopathy) [Fougerousse et al. 2007; Pacak et al. 2007]. Human clinical trials of gene therapy have been performed in limb-girdle muscular dystrophy type 2D [Mendell et al. 2009, 2010b]. Although initial trials are small, they have shown promising results, laying the foundation for future gene therapy research for this group of patients.

Antisense oligonucleotide therapy

The goal of antisense oligonucleotide therapy in Duchenne muscular dystrophy is to convert out-of-frame deletions into in-frame deletions by altering the splicing patterns of pre-mRNA in mutant dystrophin. This approach may restore the open reading frame via targeted exon skipping, in the hope of at least partially restoring the levels of functional, albeit truncated, dystrophin protein. Challenges include the selection of dystrophin mutations to target, the exon(s) to be skipped, the presence of other disease modifying factors, and the severity of the disease process during the time of treatment, as muscle tissue needs to be present for regeneration to occur on a large scale [Muntoni and Wood, 2011].

There are several human clinical trials that have been conducted using the antisense oligonucleotide approach. Two-thirds of Duchenne muscular dystrophy patients have one or more exon deletions, of which 70% lie between exon 45 and exon 55 [Aartsma-Rus et al. 2009]. Skipping of exon 51 could potentially restore the reading frame for 13% of all Duchenne muscular dystrophy patients [van Ommen et al. 2008]. Two types of antisense oligonucleotides that have been developed to induce skipping of exon 51 are 2’-O-methylphosphorothioate oligoribonucleotide (2OMePS) and phosphorodiamidate morpholino oligomer (PMO). In phase I studies, these oligonucleotides were well tolerated and were associated with a modest increase in dystrophin protein expression. A pilot study involving intramuscular injections of a 2OMePS that induces skipping of exon 51 demonstrated that the intervention was well tolerated and that expression of dystrophin was modestly increased [van Deutekom et al. 2007]. In a phase I/II study studying local restoration of dystrophin expression with intramuscular injection of a PMO that induces skipping of exon 51, the treatment was not associated with any significant adverse events. Specific dose-dependent exon skipping was seen in the treated muscles with expression of dystrophin seen in the group of patients who were in the high-dose cohort [Kinali et al. 2009]. These encouraging results were followed by studies involving systemic administration of the same compounds. A phase I/IIa safety and pharmacokinetics trial of the 2OMePS administered via subcutaneous injection demonstrated good tolerance, increased dystrophin expression, and a modest improvement in performance on the 6-minute walk test [Goemans et al. 2011]. A phase II open-label, dose escalation study using intravenously administered PMO demonstrated good tolerance, increased dystrophin expression, but no clear clinical improvement [Cirak et al. 2011]. Currently, multiple studies are underway to assess the clinical efficiency and safety of two 2OMe antisense oligonucleotides; a phase III randomized, double-blinded placebo study [ClinicalTrials.gov identifier: NCT01254019] and a phase IIb trial to assess different dosing regimens [ClinicalTrials.gov identifier: NCT01153932]. A phase I study to assess the pharmacokinetics, safety and tolerability of subcutaneous injections of 2OMe antisense oligonucleotides in nonambulant individuals with Duchenne muscular dystrophy was completed recently [ClinicalTrials.gov identifier: NCT01128855]. In addition, a multicenter phase I/IIa pilot study is being conducted to assess subcutaneous administration of 2OMe antisense oligonucleotide that targets the skipping of exon 44 in a subgroup of Duchenne muscular dystrophy patients [ClinicalTrials.gov identifier: NCT01037309].

A question often asked in these clinical trials regarding the use of antisense oligonucleotides in Duchenne muscular dystrophy patients is whether high doses that were used in animal studies would be required for long-term treatment or whether repeated administration of lower doses would be as effective.

Stop codon read-through therapy

Approximately 7% of individuals with Duchenne muscular dystrophy harbor nonsense mutations [Muntoni and Wood, 2011] and observations from earlier studies have shown that some antibiotics such as aminoglycosides can suppress aberrant stop codons by causing misreading of the RNA and allowing alternative amino acids to be inserted at the site of the mutant stop codon [Barton-Davis et al. 1999; Howard et al. 2004]. A human clinical trial showed that 6 months of gentamicin treatment increased dystrophin levels with reduced creatine kinase levels, implying that this treatment successfully induced read-through of stop codons [Malik et al. 2010]. However, gentamicin toxicity remains a major concern.

Initial concerns regarding possible gentamicin toxicity with chronic, high-dose therapy prompted the development of PTC124. PTC124 is a synthetic compound that appeared to permit the selective read-through of nonsense mutations in mRNAs by binding to the 60S ribosomal subunit. In mdx mouse models, administration of PTC124 orally, intraperitoneally or both for 2–8 weeks led to suppression of nonsense mutations, increased dystrophin production, and improvement in strength [Welch et al. 2007]. The suppression of nonsense mutations was documented using a luciferase reporter. A subsequent study demonstrated that PTC124 directly inhibits the activity of luciferase, suggesting that this reporter is not well-suited for assessing the efficacy of PTC124 [Auld et al. 2009]. Phase I studies in healthy volunteers suggested safety and tolerability were reasonable [Hirawat et al. 2007]. However, a phase IIb multicenter, randomized, double-blind study reportedly showed disappointing results. Duchenne muscular dystrophy boys aged 5 years and older were randomized to a placebo, low-dose or higher-dose PTC 124, receiving the study drug thrice a day for 48 weeks [ClinicalTrials.gov identifier: NCT00592553].

Growth factors

Instead of targeting the primary defect, counteracting the downstream effects of dystrophin deficiency could be an alternative treatment in Duchenne muscular dystrophy patients. Endogenous growth factors such as insulin-like growth factor I (IGF-1) and members of the transforming growth factor-β (TGF-β) superfamily have been shown to play an important role in the regulation of muscle regeneration and development of fibrosis in dystrophic muscle [Barton et al. 2002; Li et al. 2004]. IGF-1 promotes the formation of new muscle fibers and repairs damaged ones. An elevation in IGF-1 expression in mdx mouse muscle tissue appears to be effective in increasing muscle strength and preventing muscle necrosis [Barton et al. 2002]. Using the above principles, a phase I randomized clinical trial is currently being planned to assess the safety and efficacy profile of IGF-1 administered by subcutaneous injection once daily in Duchenne muscular dystrophy boys 5 years and older [ClinicalTrials.gov identifier: NCT01207908].

Myostatin, also known as growth differentiation factor 8 (GDF-8) belongs to the transforming growth factor-β superfamily of proteins, and inhibits muscle differentiation and growth. Myostatin is expressed primarily in skeletal muscle fibers. Blocking myostatin activity in the mdx mouse model resulted in increased muscle regeneration, decreased muscle fibrosis and increased strength [Wagner, 2008; Wagner et al. 2002]. Over the last decade, there has been increased interest in myostatin inhibition after a boy with a myostatin mutation was reported to be unusually strong and had dramatic muscle hypertrophy [Schuelke et al. 2004]. A phase I/II human clinical study demonstrated safety but no beneficial effect [Wagner et al. 2008], although there were a few limitations in this study. The trial was a safety trial with small sample sizes; hence, statistically significant changes between the treatment and placebo arms may not have been detected. The follow-up duration of this trial may also have been too short to detect any disease progression or improvement in muscle strength. Further animal and human studies are ongoing.

TGF-β1 belongs to a family of cytokines that have been shown to be involved in fibrosis formation in muscular dystrophy [Li et al. 2004; Wagner, 2008]. Losartan, a medication that is used in Duchenne muscular dystrophy patients to slow the progression of cardiomyopathy, is an angiotensin II type 1 receptor antagonist which also has antifibrotic properties via inhibition of the TGF-β signaling cascade. mdx mice treated with losartan demonstrated improved muscle regeneration and decreased fibrosis after 18 days, with attenuation of disease progression seen after 6–9 months of treatment [Cohn et al. 2007]. Human patients have undergone unofficial individual therapeutic trials of losartan by anecdotal reports, but the potential efficacy of this compound for muscular dystrophy has not been studied rigorously in a human clinical trial to date.

Outcome measures

Therapeutic drug developments have highlighted the need for reliable, validated and sensitive outcome measures that can act as surrogate endpoints to assess possible changes following treatment. Markers of muscle function such as manual and quantitative muscle strength exist currently. One such measure that has established its feasibility and reliability is the 6-minute walk test that is increasingly used in patients with Duchenne muscular dystrophy to characterize ambulation over time [McDonald et al. 2010]. Other functional scales that have been used include the Gross Motor Function Measure, the Motor Function Measure, Hammersmith Motor Ability Scale for Duchenne Muscular Dystrophy Boys or the North Star Ambulatory Assessment [Mazzone et al. 2010]. More recently, the use of quantitative and qualitative analysis of muscle by different imaging modalities has been explored and may become an important component of clinical trials in the future.

Spinal muscular atrophy

Overview

Spinal muscular atrophy is another common inherited neuromuscular disease presenting in childhood, characterized by degeneration of lower motor neurons, leading to progressive muscle weakness and atrophy. The estimated incidence is 1 in 6000 to 1 in 10,000 live births; with a carrier frequency of 1 in 40 to 1 in 50 in most populations regardless of sex or ethnicity [Darras and Kang, 2007; Pearn, 1978]. The vast majority of cases of spinal muscular atrophy result from homozygous deletions in exon 7 of the survival of motor neuron 1 (SMN1) gene at locus 5q13 [Fougerousse et al. 2007; Lefebvre et al. 1995]. Approximately 5% of spinal muscular atrophy patients have point mutations in SMN1 or other genes [Darras, 2011]. The SMN gene is present in two isoforms on chromosome 5 that have very similar sequences: SMN1 and SMN2. The protein product of SMN2 often lacks exon 7 and is unstable, whereas the full-length SMN2 product is only translated in small amounts. Most spinal muscular atrophy patients have homozygous deletions of the SMN1 gene but have at least one copy of the SMN2 gene. There are rough inverse correlations between the SMN2 copy number and the disease severity, suggesting that SMN2 may help compensate in part for SMN1 deficiency [Lefebvre et al. 1997; Mailman et al. 2002; Markowitz et al. 2012; Swoboda et al. 2005]. Although the disease heterogeneity is partly due to the copy number of SMN2, the phenotypic variability within the patients carrying the same number of SMN2 copies [Montes et al. 2009] suggests that other genetic factors may influence the clinical course for specific affected individuals [Darras and Kang, 2007].

Spinal muscular atrophy has been divided into different clinical groups defined by their maximum function achieved: type 1 patients never sit, type 2 patients are able to sit at some point during their course but never walk, and type 3 patients are able to walk independently at some point during their course [Markowitz et al. 2012; Wang et al. 2007]. Some clinicians have also used the term spinal muscular atrophy type 0 to describe a subset of affected infants with prenatal onset, who were born with hypotonia, severe weakness, joint contractures (arthrogryposis), and respiratory insufficiency. At the contrasting end of the spectrum, there are affected individuals who have onset of symptoms in adulthood (type 4) [Darras and Kang, 2007; Markowitz et al. 2012; Pearn, 1980]. Common clinical features shared by spinal muscular atrophy patients are hypotonia, proximal weakness involving the trunk and the extremities, tongue fasciculations and later, joint contractures and orthopedic issues such as scoliosis. The weakness is typically symmetrical, but some cases of spinal muscular atrophy type 3 associated with asymmetric weakness have been described [Kang et al. 2006].

Consensus statements and treatment guidelines recommending multidisciplinary supportive measures for patients with spinal muscular atrophy indicate that with proper care, many affected individuals will have improved quality of life and increased life expectancy [Wang et al. 2007]. To date, there is no definitive treatment for spinal muscular atrophy, though there are major efforts exploring various pharmacological methods that can potentially upregulate the expression of SMN2 or produce more functional SMN protein (Table 2).

Table 2.

Status of therapeutic development in spinal muscular atrophy.

	Preclinical	Phase I	Phase II	Phase III
Pharmacological methods
• Histone deacetylase inhibitors	√ [Andreassi et al. 2004]	√ [ClinicalTrials.gov identifier: NCT00374075]	√ [ClinicalTrials.gov identifier: NCT00528268]
• β-adrenergic agonist	√ [Angelozzi et al. 2008]	√ [Tiziano et al. 2010]
Antisense oligonucleotide	√ [Williams et al. 2009]	√ [ClinicalTrials.gov identifier: NCT0149701]
Gene therapy	√ [Foust et al. 2010]
Cell-based therapy	√ [Ebert et al. 2009]

One such example is albuterol, a β-adrenergic agonist that was shown to promote the inclusion of exon 7, increasing full length SMN2 messenger RNA and SMN2 protein on an in vitro level [Angelozzi et al. 2008]. This was translated into clinical trials involving spinal muscular atrophy types 2 and 3 patients which demonstrated an increase in full length transcript levels of SMN2 and improvement in muscle strength and Hammersmith Functional Motor Scale score without significant adverse effects [Angelozzi et al. 2008; Kinali et al. 2002; Pane et al. 2008; Tiziano et al. 2010]. Despite these promising results, this treatment has yet to be evaluated in a placebo-controlled, randomized blinded trial to confirm its efficacy.

Outcome measures

There are many challenges being faced by clinicians involved in clinical trials surrounding spinal muscular atrophy patients. One of these includes the natural history of the disease, especially in types 2 and 3. These patients experience early decline in their strength followed by a prolonged plateau period [Markowitz et al. 2012]. Reliable and sensitive outcome measures are important not only to assess severity of disease and disease progression, but also for defining the impact and effect of therapeutic agents on the rate of progression of disease over a period of time. New outcome measures have emerged over the years and while these appear to be reliable in spinal muscular atrophy patients, a consensus needs to be reached on the type of outcome measures to be used in clinical trials. Categories of outcome measures used include clinical scales, electrophysiologic measures, ultrasound data, and biomarkers. More recent natural history studies document the use of such outcome measures in greater detail [Kaufmann et al. 2011, 2012]. A cross-sectional, multicenter clinical trial evaluating SMN protein, transcript, and copy number, with motor function assessed using the Modified Hammersmith Functional Motor Scale, carried out across a range of clinical severity of patients with spinal muscular atrophy, demonstrated that these biomarkers correlated with spinal muscular atrophy subtype [Crawford et al. 2012]. More recently, a multicenter clinical trial which aims to study the natural history and biomarkers in infants with spinal muscular atrophy type 1 has been launched by the National Institute for Neurological Disorders and Stroke (NINDS) through its clinical trial network, Network for Excellence in Neuroscience Clinical Trials (NeuroNext).

Clinical outcome measures

The Children’s Hospital of Philadelphia (CHOP) Test of Strength in spinal muscular atrophy and Infant Test for neuromuscular disease were developed to assess the motor skills of children with neuromuscular disease [Montes et al. 2009]. These include observational and clinical assessments of neck, trunk, proximal, and distal limb strengths. Other outcome measures used in spinal muscular atrophy patients include the Gross Motor Function measure [Nelson et al. 2006], the Hammersmith Functional Motor Scale [Main et al. 2003] and the Modified or Expanded Hammersmith Functional Motor Scale [Krosschell et al. 2006; Nelson et al. 2006; O’Hagen et al. 2007]. The 6-minute walk test is an objective measurement of the distance a patient can walk in 6 minutes, and has been shown to be a reliable outcome measure in ambulatory patients with spinal muscular atrophy [Montes et al. 2010]. It is easy to administer and can be used to quantify one’s functional mobility.

Respiratory assessments such as pulmonary function tests can be used to monitor the respiratory status of spinal muscular atrophy patients. However, this test relies on patient cooperation and children generally need to be at least 5 years or older in order to achieve reliable results.

Electrophysiological outcome measures

Motor unit number estimation (MUNE) is an electrophysiological measure that was originally developed for the study of amyotrophic lateral sclerosis (ALS), but was subsequently adapted for spinal muscular atrophy based on the similarities in motor neuron pathology between the two diseases. The MUNE is a noninvasive representation of the number of functional motor units that exist in a specific myotome [Bromberg and Swoboda, 2002; McComas, 1991; Wu et al. 2010]. Both the MUNE and maximum compound motor action potential (CMAP) amplitude have been used to provide a measure of the overall health of the motor neurons and their innervated muscle groups in research studies of spinal muscular atrophy [Lewelt et al. 2010; Swoboda et al. 2005]. There appears to be a statistically significant correlation between CMAP amplitude during initial assessment and the functional outcomes of these patients [Swoboda et al. 2005].

Quantitative muscle ultrasound

Quantitative muscle ultrasound is a technique that has been shown to be useful in assessing progression of neuromuscular diseases in both children and adults. This approach was used in an observational study to assess the value of muscle ultrasound in patients with spinal muscular atrophy types 2 and 3 by correlating strength with luminosity ratios [Wu et al. 2010]. Luminosity ratios were calculated by dividing muscle luminosity by subcutaneous fat luminosity. The results demonstrated a positive correlation with more severely affected spinal muscular atrophy individuals having higher luminosity ratios.

Others

Serologic biomarkers measuring blood SMN messenger RNA and protein levels have also been studied but these have not been shown to correlate reliably with disease severity [Sumner et al. 2006].

Therapeutic approaches

Histone deacetylase (HDAC) inhibitors are a class of drugs that have been extensively investigated as potential therapeutic drugs for spinal muscular atrophy patients. While these agents were shown to increase full length SMN2 transcript levels in cell lines, clinical trials with different agents including phenylbutyrate [Mercuri et al. 2007], valproic acid [Sumner et al. 2003; Swoboda et al. 2009], and hydroxyurea [Chen et al. 2010] have been disappointing. A possible reason that some of these clinical trials fail to demonstrate a statistically significant positive outcome may be due to the small sample size leading to an underpowered study. Another HDAC inhibitor, LBH589 (hydroxamic acid) is a promising candidate for the treatment of spinal muscular atrophy. It not only demonstrated an increase in SMN protein levels by 10-fold in human spinal muscular atrophy fibroblasts, but also appears to induce SMN2 expression in spinal muscular atrophy fibroblasts that were unresponsive to valproic acid [Garbes et al. 2009].

More recently, in vitro and mouse models have suggested that antisense oligonucleotides that prevent the skipping of exon 7 can promote an increase of production of full length SMN2 mRNAs [Hua et al. 2010; Kinali et al. 2009; Williams et al. 2009]. This increase may be sufficient to compensate for the lack of SMN1 in these animal models. The challenges facing this approach include finding an efficient delivery method that ensures penetration of the blood–brain barrier into the central nervous system. In mouse models of spinal muscular atrophy, intracerebroventricular injections of antisense oligonucleotides appear to be more efficacious than systemic delivery with respect to increasing full-length SMN2 expression in motor neurons [Dickson et al. 2008; Hua et al. 2010; Passini et al. 2011; Porensky et al. 2012; Williams et al. 2009], although some data suggest that systemic delivery may also be beneficial [Hua et al. 2011]. In a study assessing periodic intracerebroventricular delivery of antisense oligonucleotide in a mouse model, this delivery method appears to demonstrate a higher level of SMN expression in the central nervous system, providing improvement in the phenotype with enhanced motor function [Williams et al. 2009]. A multicenter phase I trial began enrollment in December 2011 to assess the safety and tolerability of an antisense oligonucleotide administered as a single dose intrathecal injection [ClinicalTrials.gov identifier: NCT0149701].

Recent preclinical studies involving stem cells and gene therapy also show promise. Pluripotent stem cells have been derived from skin fibroblasts of a child with spinal muscular atrophy [Ebert et al. 2009]. Self-complementary adeno-associated virus 9 (scAAV9) vector containing SMN crossed the blood–brain barrier, rescued motor function, and prolonged life expectancy in mice with spinal muscular atrophy when administered during the neonatal period [Foust et al. 2010]. In another study, an adeno-associated virus vector containing human SMN was injected into the central nervous system of newborn mice with spinal muscular atrophy, resulting in clinical improvements in muscle strength and coordination, as well as increased median life span [Passini et al. 2010]. Stem cell therapy and gene therapy are still in the preliminary stages of development, but have immense potential for yielding transformative therapies in the future. A common obstacle facing all these novel therapeutic approaches is the difficulty of penetration or delivery into the central nervous system.

Conclusions

Over the last decade, major advances have been made in the development of potential therapies for pediatric neuromuscular diseases, including Duchenne muscular dystrophy, spinal muscular atrophy, and limb girdle muscular dystrophies. These advances in research have provided families with children affected with these severe and often, fatal diseases, a glimpse of hope that continued research in these areas will eventually lead to more effective treatments.