Spinal Muscular Atrophy

Spinal muscular atrophy (SMA) is an autosomal recessive neuromuscular disorder of infancy, childhood, and young adulthood. Incidence of SMA is estimated to be 1 in 11,000 live births.¹ The cause of SMA is a biallelic homozygous deletion (95%) or other mutation in the survival motor neuron1 (SMN1) gene located on chromosome 5q, which results in degeneration of the anterior horn motor neurons. The SMN2 gene has a single nucleotide difference (C to T transition), generating a truncated protein without exon 7 (delta 7-SMN). The severity of SMA is ameliorated by the number of SMN2 copies an individual has. People with less severe SMA have more copies of SMN2. The SMN2 gene produces only a small amount of full-length SMN mRNA.²

The development and commercialization of targeted gene-based therapies have significantly altered prognosis for people with SMA. Until 2016, no treatment was available, and management consisted of supportive measures including respiratory and nutritional support and preventing complications of weakness. There are now 2 therapies approved by the Food and Drug Agency (FDA) for SMA and more novel therapies are being developed. Despite new therapy advancements, symptomatic management remains a vital part of caring for people with SMA.

Available Treatments

Nusinersen

Clinical Use. Nusinersen is given as an initial loading dose of 12 mg of nusinersen every 2 weeks for 6 weeks and again 1 month later. After the loading period, 12 mg of nusinersen is given intrathecally every 4 months. Complete blood count (CBC), prothrombin time (PT), partial thromboplastin time (PTT) and urine protein are monitored before each dose. The most common adverse reactions in infantile-onset SMA have been lower respiratory infection and constipation. The most common adverse reactions in the later-onset SMA have been pyrexia, headache, vomiting and back pain. More rare and severe side effects can occur like thrombocytopenia and elevation of protein in urine.

Mechanism of Action. Nusinersen increases the production of full-length SMN protein by promoting the inclusion of exon 7 in the SMN2 mRNA transcript. Nusinersen is a modified antisense oligonucleotide (ASO), designed to target a heterogeneous nuclear ribonucleoprotein (hnRNP)-A1 / A2–dependent splicing silencer, ISS-N1, in intron 7 of the SMN pre-mRNA. By displacing the silencer, accurate splicing of SMN2 transcripts is facilitated such that more transcripts containing exon 7 are produced and translated into full-length SMN protein.³

Efficacy in Clinical Trials. Nusinersen has been studied in presymptomatic infants, symptomatic infants less than age 6 months, symptomatic patients age 6 months to 2 years, and older individuals with symptomatic SMA type 2 and 3.There are 2 ongoing trials of nusinersen to treat presymptomatic individuals and for long-term follow up of individuals who participated in earlier trials of nusinersen.

The pivotal study of nusinersen was a phase 3 multicenter randomized double-blind sham-procedure control in 121 infants who were less than age 6 months and had symptomatic infantile-onset SMA (most likely SMA type 1). Participants were randomly assigned in a 2:1 ratio to have intrathecal administration of nusinersen (n=81) or a sham procedure (n=41). Infants who received nusinersen had intrathecal injections of 12 mg of nusinersen on days 7, 15, 29, 64, 183, and 302.

The primary endpoints were motor-milestone response, defined by the Hammersmith Infant Neurological Examination (HINE) and survival without use of permanent assisted ventilation. Secondary endpoints included overall survival, subgroup analysis of event-free survival according to disease duration, the Children’s Hospital of Philadelphia Infant Test of Neuromuscular Disorders (CHOP INTEND), and the Compound Motor Action Potentials (CMAP).

Section 2 of the HINE (HINE-2) measures the achievement of motor milestones.^4,5 Participants were categorized as responders if they: 1) had an improvement in at least 1 category on the HINE, which includes the categories head control, rolling, sitting, crawling, standing or walking, or had a 2-point or more increase in the score for kicking or achieved the maximal score for kicking, and 2) had improvement in more categories than they had worsen.⁶

Interim analysis showed that 41% (21 / 51) of infants treated with nusinersen responded to treatment compared with none of the infants who were sham treated responding. On the basis of this efficacy and ethical concerns of nontreatment for the sham-treated children, the trial was terminated and all children participating were treated with nusinersen. Final analysis showed lower rates of death and tracheostomy or more than 16 hours / day of ventilation in those given nusinersen vs sham treatment (39% [28 / 73] vs 68% [25 / 37]). At the end of the trial, 12 of 73 infants (16%) treated with nusinersen died compared with 14 of 37 (39%) of those given sham treatment.

Of those treated with nusinersen, 51% (37 / 73) had responded with motor improvements vs 0% of the 37 infants given sham treatment. Of infants treated with nusinersen group, 22% (16 / 73) attained full head control, 10% (7 / 73) were able to roll, 8% (6 / 73) sat independently, and 1% (1 / 73) was able to stand. None of the children treated with the sham procedure achieved these milestones. More infants treated with nusinersen vs sham (71% [52 / 73] vs 3% [1 / 37]) had a CHOP INTEND response. Adverse events were similar in the two groups. The most common adverse event were fever, constipation, and respiratory infections.⁶

A phase 2 open-label study of nusinersin in presymptomatic infants genetically diagnosed with SMA who have 2 (n=15) or 3 (n=10) copies of SMN2 and were given a first dose at age less than 6 weeks is ongoing.⁷ Results of an interim analysis in February 2020 were presented at the virtual Cure SMA conference in June 2020, no participants had died or required ventilatory support. The median age of participants is 3.8 years, past the expected age of symptom-onset, ventilatory support, and death for children with untreated SMA type 1 and 2. All participants are able to sit without support, 96% (24 / 25) achieved walking with assistance, and 88% (22 / 25) are able to walk independently. The HINE-2 motor milestone total scores have increased over time for all participants regardless of SMN2 copy number. The CHOP INTEND scores have increased and then remained stable over time. At the interim analysis, the maximum CHOP INTEND score of 64 was achieved by 67% (10 / 15) of participants with 2 SMN2 copies and 100% (10 / 10) of those with 3 SMN2 copies.⁸

Onasemnogene

Clinical Use. Onasemnogene is approved for children less than age 2 years with genetically proven SMA as a 1-time intravenous infusion over 1 hour. Pretreatment with systemic steroids (1 mg / kg / day) is started the day before the infusion and continues for 1 month after infusion. If liver function and clinical examination are normal 1 month after treatment, the child is weaned off steroids over 28 days. Some children may require a higher dose of steroids
(2 mg / kg / day) and weaning over a longer period. The most common adverse reactions to onasemnogene are elevation of aminotransferases and vomiting.

Mechanism of Action. Onasemnogeneis a 1-time gene-replacement therapy (GRT) designed to deliver a fully functional copy of the human SMN1 gene administered intravenously. The SMN1 gene is delivered on-replicating recombinant adeno-associated virus serotype 9 (AAV9) vector.

Efficacy in Clinical Trials. The pivotal study of onasemnogene was a phase 1 / 2 open-label single-center study in infants with SMA type 1 less than age 6 months. Participants received either 6.7x10^-13 vg / kg (n=3) or 2.0x10^-14 vg / kg (n=12) as a single intravenous dose. Prednisone 1 mg / kg was given the day before the infusion and at least 30 days after the therapy administration. All 15 participants have survived without needed ventilatory support compared with only 8% survival expected in untreated children. The CHOP INTEND scores improved from baseline by mean 9.8-point increase at 1 month and 15.4-point increase at 3 months. Of those treated with the higher dose, 92% (11 / 12) sat without support and could feed orally; 9 could roll over.⁹ At the June 2020 virtual Cure SMA conference, it was announced that as of December 2019, 2 participants could stand without assistance.¹⁰ Adverse events included increased aminotransferase levels in 4 participants, controlled with oral prednisone.

In an open-label single-arm multicenter phase 3 study of onasemnogene, 91% (20 / 22) children with symptomatic SMA type 1 who were treated at age less than 6 months survived and 82% were free of ventilatory support at age 18 months. Most (59%,13 / 22) were sitting independently for 30 seconds or more at age 18 months.¹¹

In an open-label single-arm multicenter phase 3 study of presymptomatic infants with SMA who had 2 (n=14) or 3 (n=15) copies of SMN2 and were treated with onasemnogene at age less than 6 weeks, 100% have survived without a need for ventilatory support. Children with 2 copies of SMN2 are reaching age-appropriate motor milestones, including 57% (8 / 14) sitting independently for 30 seconds or more and 29% (4 / 14) walking independently. The participants with 3 copies of SMN2 are similarly reaching age-appropriate motor milestones, including 27% (4 / 15) standing independently and 20% (3 / 15) walking independently. No participant has required a feeding tube, and most remained within the normal weight range.¹²

In a phase 1 / 2 trial of onasemnogene for children age 6 months to 5 years who could sit independently for more than 10 seconds (but not walk or stand independently) and had 3 copies of SMN2, participants received 1 of 3 possible doses intrathecally instead of intravenously. Interim data analysis, done in December 2019, was presented at the virtual CureSMA conference in June 2020. All participants were alive with no new clinical safety signals detected. There was a statistically significant and clinically meaningful early and sustained increase in Hammersmith Functional Motor Scale Expanded (HFMSE) compared with natural history in patients age 2 to 5 years who received for 1 of the 3 doses. However, this study is currently on hold as dorsal root ganglia toxicity was seen in animal studies.

Investigational Treatments

Other therapies, currently in clinical trials, include oral small molecules to modify SMN2 RNA splicing (risdiplam), monoclonal antibodies to promyostatin, and medications that act in the neuromuscular junction (amifampridine) (Table). Of note, risdiplam was accepted for priority review by the FDA in 2019. The decision is expected in August 2020.

Conclusion

The approval of effective therapies for SMA has completely changed the landscape for SMA, converting a usually terminal early-life diagnosis into a treatable condition. With available treatment and prolonged lives, new phenotypes are being defined and have entirely new complications and comorbidities. Standard of care is still essential for the success of these therapies. It is likely that a combination of SMN directed and SMN-independent treatments will be the future of SMA treatment.