How far away is spinal muscular atrophy gene therapy?

Abstract

Spinal muscular atrophy (SMA) is a devastating motor neuron disease primarily affecting children, for which there is currently no known disease-modifying therapy or cure. The identification of the disease gene, survival motor neuron, led to an expansion in SMA research and allowed the creation of numerous animal and cellular models. This led to a significant increase in our understanding of the pathophysiology of SMA, culminating in the development of multiple SMN-dependent and -independent therapies. Among the most exciting options, viral gene therapy has emerged as one leading candidate. A growing body of pre-clinical evidence suggests that administration of scAAV9 carrying an SMN transgene can be both efficacious and translationally viable. In this article, we briefly review the progress which has been made in the field, and provide a commentary on some of the challenges which remain.

Keywords:

• AAV9
• delivery route
• gene therapy
• SMN
• spinal muscular atrophy
• therapeutic time window

Spinal muscular atrophy (SMA) is a motor neuron disease affecting primarily children, in which motor neurons connecting the spinal cord and brainstem to skeletal muscle are lost, resulting in progressive paralysis. Clinically, SMA is classified into three main types (I–III), which range in their clinical severity. The most severe form of this disease (SMA type I) was originally described in the 19th century by clinicians Guido Werdnig and Johann Hoffman. However, it was not revealed until 1995 that 95% of SMA patients carry mutations and deletions within a single gene, which was understandably named survival motor neuron (SMN1) [1]. Since then the research field has rapidly expanded, and over the past 20 years there has been immense progress in the development of animal and cellular models, as well as in our understanding of the function of SMN protein and the pathogenesis of SMA. From this growing body of knowledge, a number of therapeutic strategies have emerged. Thus, it is easy to see why many regard the SMA research field to be at its most exciting stage, at least since the identification of the gene. However, this optimism must be balanced with considered caution. SMA remains a devastating motor neuron disease with no disease-modifying therapy available. Treatment currently revolves around palliative care and SMA is still the most common genetic cause of infant mortality. This highlights the importance of our continued efforts to understand the pathogenesis of disease and move therapeutics forward in a careful and efficient manner.

A number of therapeutic strategies have been explored for SMA. Many of these target a second partially functional copy of the SMN gene, termed SMN2. These include antisense oligonucleotides, aimed at correcting the splicing defect of SMN2, and small molecules aimed at increasing expression of the SMN2 gene, or increasing the stability of the small amounts of SMN protein present [2]. There are also a number of SMN-independent therapies under investigation. However, one of the most promising strategies proposed for SMA is replacement of the SMN1 gene by viral gene therapy. This approach relies on using a viral vector to deliver a functional copy of the SMN1 gene to all patient cells that have a defective copy, resulting in a permanent restoration of gene expression. As an essentially monogenetic disorder, SMA is an ideal candidate for such a therapy. Furthermore, due to the presence of the SMN2 gene, patients retain very low SMN protein levels, meaning expression of SMN from the new construct is unlikely to induce an immune response. The potential for this strategy is therefore clear. For the remainder of this article, we will discuss the advances which have been made in the development of gene therapy for SMA and discuss some of the outstanding hurdles which have yet to be faced.

Development of gene therapy for SMA

The ability to use viruses to drive exogenous expression of SMN in patient cells was first demonstrated back in 2003 [3]. The next major breakthrough occurred in 2010, when Foust et al. demonstrated that a single intravenous injection at P1 of a self-complementary adeno-associated virus, serotype 9 (scAAV9), carrying the SMN1 gene could increase the survival of an SMA mouse model from 14 to over 250 days [4]. The success of this approach appears to lie in part with the use of the AAV9 construct, which can cross the blood–brain barrier and appears to show remarkable affinity for motor neurons, and the use of the self-complementary subtype, which allowed rapid expression of the transgene. Contemporary studies confirmed the benefits of intravenous scAAV9-SMN1 delivery in P1 SMA mice, and also demonstrated remarkable protection following CNS and intramuscular delivery [5–7]. Importantly, the therapeutic relevance of this approach has been demonstrated by successful systemic or intrathecal delivery of viral vectors carrying a green fluorescent protein into juvenile pigs and primates, resulting in robust expression [4,8,9]. Indeed, intrathecal delivery of scAAV9 carrying human SMN1 conferred significant benefit to a pig model of SMA [10]. The clinical relevance of AAV-derived therapies has also been demonstrated by recent reports of using AAV to delivery follistatin to muscle of patients with Becker muscular dsystrophy [11]. The remarkable rescue of SMA animal models, and proof of translational potential, has provided the impetus required to approve clinical trials for the use of scAAV9-SMN1 in SMA patients. A Phase I dose-escalation study for the systemic delivery of scAAV9-SMN1 into type 1 SMA patients, led by Jerry Mendell, began recruiting in April 2014. The primary objective of this study is to establish the safety of this treatment over a 2-year period.

Outstanding questions

Should the hopes of the field be realized, and the results observed in animal models translate into human patients, a number of crucial challenges are likely to still come to the fore. These include the successful delivery of therapy at the critical time-points in disease progression where SMN levels need to be restored, and in all tissues that need to be targeted, in order to see robust therapeutic benefits.

Timing of intervention

It is clear from several genetic and pre-clinical studies that for greatest benefit, SMN levels must be restored as early as possible. Indeed, while intravenous administration of scAAV9-SMN1 into pre-symptomatic (P2) SMA mice resulted in an increase in mean survival from 14 to 250 days, equivalent treatment at P5 resulted in just 15 days extension in mean survival [4]. Furthermore, following ICV administration of scAAV9-SMN1 in SMA mice, there is a clear correlation between survival, weight and motor performance compared with time of intervention [12]. From a scientific standpoint, this raises a number of intriguing questions. What happens during these very early stages in disease that limit this therapeutic time window? Does this time correlate with irreversible motor neuron damage, beyond which time even restoring SMN levels is not sufficient for their survival? Or does this time point reflect developmental requirement for SMN within the neuromuscular system? In addition, is there a developmental component to severe types of SMA which may limit the efficacy of SMN replacement therapies? Understanding what limits this therapeutic time window may be the key to developing efficient combinational therapeutics which can boost the efficacy of SMN-dependent therapies. However, as an aside from these scientific curiosities, it is clear that for greatest therapeutic efficacy, treatment should be given before symptoms emerge. For that reason, current efforts to establish a newborn screening program for SMA will become critically important [13]. A fundamental impedance to the development of such a program is the absence of an effective treatment, and the ethical issues which therefore surround it. Perhaps the growing number of clinical trials will facilitate progress.

Tissue specificity

Another key issue facing the SMA research community is defining the cells types where SMN levels must be restored. Delivery of SMN viral vectors is typically performed via systemic (intravenous) or CNS (i.e., intracerebroventricular) methods. The CNS-specific modes of delivery have clear benefits. They require a much lower dose, and are thereby cheaper and faster to produce. However, this dosage route may result in suboptimal gene delivery to peripheral tissues. The question of which cells types are affected in SMA has long been controversial. While motor neurons are clearly the primary pathological target in SMA, in mouse models of SMA defects in other tissues, including brain, muscle, heart and pancreas have been described [14,15]. Indeed, as quality of care improves and children with severe types of SMA live longer, we are hearing anecdotal evidence of similar defects in patients. We must consider whether a CNS-specific restoration of SMN in patients may therefore unmask secondary defects in peripheral tissues.

Numerous studies have attempted to define which tissues require SMN restoration to rescue an SMA-like phenotype. These studies have generated mixed results, from which it can be difficult to draw clear conclusions. While it is true that CNS delivery of viral vectors carrying SMN transgenes can ameliorate pathology in peripheral tissues [12,16], systemic delivery of SMN-dependent therapies often results in a greater extension in lifespan [17]. Perhaps the most parsimonious conclusion at present is that restoring SMN specifically to the CNS can have significant benefit upon the phenotype, but for maximum rescue SMN must be restored to all tissues.

Ultimately, the answers to these questions may only be revealed by long-term studies of patients receiving either CNS or systemic administration of SMN gene therapy. In the interim, it is important to continue efforts to understand the role of SMN in both motor neurons and other cell types, and develop therapeutics that can also target peripheral tissues.

Future perspectives

The future of gene therapy as a treatment for SMA is likely to be highly dependent upon the results of current clinical trials. The occurrence of significant side effects during this treatment would undoubtedly have important implications for all future therapeutic development. All pre-clinical studies have suggested that this treatment should be well tolerated. However, until a significant number of patients have been safely treated, the future of this therapy remains uncertain. In the absence of any significant adverse events, we would anticipate the number of gene therapy-based trials for SMA will increase in size and scale. Should the hurdle of safety be passed, attention will then be focused on treatment efficacy. This will inevitably involve looking carefully at timing and routes of intervention. It may also involve specific stratification of trials aimed at treating different forms of SMA, allowing specific investigation into whether existing patients with type II and type III SMA could benefit from this type of treatment, even during advanced stages of disease. This is undoubtedly a critical time within the SMA field, and we await the results of the current clinical trials with great interest and hope. Regardless, multiple therapeutic options for SMA continue to advance toward clinical application. Therefore, we must harbor cautious optimism that the next 5 years of research in the SMA field will continue to move us closer to a meaningful treatment for this disorder.

Key issues

• Spinal muscular atrophy (SMA) is a monogenetic childhood motor neuron disease for which there is no know treatment or cure.

• Recent work has revealed that intravenous delivery of scAAV9 carrying a transgene for SMN can dramatically rescue severe mouse models of SMA.

• Work in pig and primates has proven scAAV9 can cross the blood–brain barrier following intravenous injection, and can express an exogenous protein in their motor neurons.

• Phase I clinical trials using scAAV9 in SMA patients are currently underway.

• It is still unclear which tissues we need to target for SMN replacement therapy.

• It is likely SMN replacement therapies will have to be given prior to symptom onset for maximum benefit.

• The results of the current clinical trials will be directive to future trials using viral vectors in SMA patients.

Financial & competing interests disclosure

TH Gillingwater is supported by grants from the SMA Trust, Muscular Dystrophy UK, and the AxonomiX Network. LM Murray is supported by grants from the Muscular Dystrophy Association, Fight SMA, Cure SMA and the Gwendolyn Strong Foundation. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.