Clinical perspectives: Treating spinal muscular atrophy

Abstract

Spinal muscular atrophy is a rare and progressive neuromuscular disease that, without treatment, leads to progressive weakness and often death. A plethora of studies have led to the approval of three high-cost and effective treatments since 2016. These treatments, nusinersen, onasemnogene abeparvovec, and risdiplam, have not been directly compared and have varying challenges in administration. In this review, we discuss the evidence supporting the use of these medications, the process of treatment selection, monitoring after treatment, the limited data comparing treatments, as well as future directions for investigation and therapy.

Graphical abstract

This clinical perspectives article discusses the path from clinical trials to treatment of spinal muscular atrophy (SMA), a rare and progressive neuromuscular disease that causes progressive weakness. McPheron and Felker discuss the three different treatments available for SMA that have not been directly compared and have varying challenges in administration.

Introduction

Spinal muscular atrophy (SMA) is an autosomal recessive genetic disorder caused by homozygous deletions or biallelic pathogenic variants in the gene SMN1, or “survival of motor neuron 1” (OMIM 600354).¹ The SMN protein is ubiquitously expressed, and plays a role in multiple cellular functions, including RNA metabolism, signal transduction, and cellular homeostasis.² Decreased expression of the SMN protein leads to the death of anterior horn cells in the spinal cord.³ SMA is a progressive neuromuscular disease: atrophy of skeletal muscles leads to muscle weakness, hypotonia, dysphagia, and eventual respiratory failure and premature death. The incidence of severe (type 1) SMA is about 1 in 10,000 live births, and it is more common in Caucasian and Asian populations than in individuals of African descent.⁴ Prior to the introduction of specific medications to treat SMA, care was mainly supportive such as maximizing pulmonary support, monitoring growth, and providing adequate nutrition, and managing physical and orthopedic complications that arose throughout the disease progression.⁵ Three treatments have been approved since 2016 for SMA, including gene replacement therapy in 2019.

Clinical features, diagnosis, and pathogenesis

There are multiple types of SMA, representing a spectrum of severity.⁵ The types of SMA were designated based on clinical features, prior to the genetic mechanisms being understood and treatments developed.⁶ SMA type 1 is the most severe. Symptoms of untreated SMA type 1 present within the first 6 months of life and include muscle weakness, loss of head control, dysphagia, and failure to thrive. Without disease-modifying medications, these infants make no significant progress in achieving motor skills and do not sit independently. Lifespan without permanent ventilation is expected to be less than 2 years. Untreated individuals with SMA type 2 have a slower course, with symptoms presenting between 6 and 18 months of age with delayed gross motor milestones. Untreated individuals with SMA type 2 are expected to achieve independent sitting, but are not able to stand or walk independently and also have a shortened lifespan. Some patients with SMA type 2 also develop dysphagia and hypoventilation. Untreated SMA type 3 presents at age >18 months, after individuals have learned to walk independently; they experience progressive muscle weakness and can lose the ability to walk.

While there are two other types, clinical trials have focused on the treatment of individuals with SMA type 1, 2, or 3. SMA type 4 presents in the second or third decade of life and is characterized by mild muscle weakness that generally does not have a significant impact on quality of life, so these individuals are not usually included in therapeutic drug trials. On the other end of the spectrum, SMA type 0 is a congenital disease which presents with decreased fetal movement in utero, severe weakness and respiratory distress at birth, and often have congenital heart disease.^7,8 CNS involvement is often reported with progressive atrophy.⁹ Autonomic dysfunction and skin necrosis have also been seen.¹⁰ Without treatment, it is universally fatal within the first few weeks of life, and individuals with this type have not been included in therapeutic studies since anterior horn cell injury and resultant muscle weakness has already occurred prenatally.¹¹ The phenotype for all types of SMA is typically improved with disease-modifying treatment.

SMA is an autosomal recessive condition caused by pathogenic variants or deletions of both copies of the SMN1 gene located at 5q13. The most common molecular cause of SMA is a homozygous deletion of exon 7 of the SMN1 gene, accounting for 95% of cases of SMA.¹² Other molecular causes include deletion of the entire SMN1 gene, or smaller genetic changes including missense, nonsense, or frameshift variants in SMN1. There does not appear to be any genotype-phenotype correlation between the type of SMN1 variant and the severity of disease, meaning that the disease is not any more or less severe whether an individual has the common exon deletion or another type of variant. Diagnostic testing for SMA includes deletion/duplication testing and sequencing of SMN1 as well as determining copy number of SMN2. Genome sequencing can also be an effective way of diagnosing infants or children who present with less specific symptoms of SMA.¹³

Disease severity of SMA is modified by copy-number variants of the gene SMN2, which is a homologous gene located near the SMN1 gene located on chromosome 5q13 (OMIM 601627). SMN2 produces only ∼10% of functional SMN protein due to a missense variant (280C>T), which leads to exon 7 skipping and mRNA degradation (Figure 1).^14,15 However, since the SMN2 gene does produce some protein, extra copies of this gene are protective against motor neuron death and thus slow disease progression.⁵ Individuals without SMA are expected to have two copies of SMN1 and one to two copies of SMN2.¹³ However, it is possible for copies of the SMN2 gene to be lost or duplicated. Eighty percent of individuals with SMA type 1 have either one or two copies of SMN2, 82% with SMA type 2 have three copies of SMN2, and 96% with SMA type 3 have three or four copies of SMN2.¹⁶ There is also a variant in exon 7 of SMN2, c.859G>C, which creates an exon splicing enhancer leading to increased production of the normal-length SMN protein and decreases disease severity.¹⁷ Most of the initial clinical drug trials excluded individuals with this variant to reduce the number of variables affecting the outcomes.

Figure 1.

SMN1 and SMN2 genes

Natural history of SMA

Prior to 2016, the treatment landscape for SMA was bleak. Treatment was strictly supportive, with death occurring before 2 years of age for most SMA type 1 patients, unless tracheostomy, ventilator, and g-tube were pursued.¹⁸ Patients with older-onset types of SMA were discovered only when they became weak, often having difficulty with ambulation and sometimes noted to have a fine postural tremor called polyminimyoclonus.

Supportive care included physical and occupational therapy to maintain range of motion and prevent contractures.⁵ Physical medicine and rehabilitation physicians and therapists were critical to help with supportive seating, wheelchairs, and other equipment needs. Many children with SMA only saw pulmonologists and other equipment and rehabilitation providers.

Prior to the availability of effective treatments, natural history studies were performed to serve as historical control data for comparison with treated infants in future clinical trials. The Pediatric Neuromuscular Clinical Research (PNCR) study enrolled 34 infants with SMA type 1 between 2005 and 2007.¹⁸ The primary combined endpoint of the study was death or permanent ventilation, defined at >16 h a day receiving invasive or non-invasive ventilatory support. The mean age to reach this endpoint was 13.5 months in infants with SMA type 1. They also evaluated motor function with The Children’s Hospital of Philadelphia Infant Test for Neuromuscular Disorders (CHOP-INTEND) tool and found that scores steadily decreased over time in infants with SMA type 1 by a mean of 1.27 points per month.¹⁹ They also evaluated a cohort of infants with SMA type 2 and found that these infants had a slower decline in motor function and longer ventilator-free survival.

The NeuroNEXT study was initiated in 2012 and enrolled 26 infants with SMA, as well as 27 healthy infants as controls.²⁰ The primary combined endpoint was death or permanent invasive ventilation in this study, and they found that the mean age of event-free survival for infants with SMA and 2 copies of SMN2 was 8 months. In this study, they also measured motor function via validated scales including the Test of Infant Motor Performance Screening Items, CHOP-INTEND, and the Alberta Infant Motor Scale. Infants with SMA had lower baseline motor scores when enrolled, and these scores steadily declined during the study, indicating progressively worsening motor skills in infants with SMA. In contrast, all healthy infants in the control group who completed the 24-month-long study survived through the study, demonstrated increased motor scores, and weighed more on average than infants with SMA.²⁰

Treatment of SMA

Nusinersen

Initial treatment strategies for SMA were aimed at increasing the production of SMN protein from the SMN2 gene. Nusinersen (Spinraza) was the first drug that proved to be effective in improving outcomes for infants with SMA. Nusinersen is an antisense oligonucleotide that works by binding to SMN2 mRNA at the location of the variant, which causes exon 7 skipping during mRNA splicing. This allows the mRNA to be read through and completely translated into a full functional protein. Nusinersen is administered intrathecally via lumbar puncture.²¹

Clinical trials

Initial nusinersen trials evaluated infants with SMA type 1 and two copies of SMN2. The phase 1 trial data indicated that the medication appeared safe, and phase 2 data demonstrated improved survival and motor function compared with historical controls.^21,22 The phase 3 ENDEAR trial was a randomized, double-blind study comparing nusinersen administration with sham lumbar puncture procedure in 121 symptomatic infants, with 80 infants in the treatment group and 41 infants in the placebo group.²³ The first primary endpoint was a motor-milestone response, which used the Hammerstein Infant Neurological Examination Section 2 tool to determine whether infants were making improvement in at least one category of motor skills such as gaining head control, rolling over, sitting, crawling, or standing. As per the final analysis, 51% of infants in the treatment group achieved a motor-milestone response, as compared with 0% of infants in the control group who all had worsening of motor function and CHOP-INTEND scores over time. The second primary outcome was event-free survival, defined as infants being alive and without permanent ventilation. At the time the study was completed, 39% of infants in the treatment group had died or required permanent ventilation, as compared with 68% of infants in the control group. Thus, clinical trials proved that nusinersen improved survival and motor function in infants treated with this medication. They also found that infants who received treatment earlier after symptom onset appeared to have improved survival, suggesting that early treatment for SMA could improve outcomes.

Effectiveness of nusinersen was also studied in children with symptom onset at greater than age 6 months, consistent with SMA type 2 or type 3, in the phase 3 CHERISH trial.²⁴ This study enrolled 126 children who were randomly assigned to the treatment group versus sham procedure; 84 children received nusinersen and 42 were in the control group. They underwent assessment at 15 months post treatment via the Hammersmith Functional Motor Scale Expanded Score and the Revised Upper Limb Module Score, and a higher percentage of the children who were treated had an improvement in scores over time. Nusinersen was approved by the FDA based on interim data in December 2016.

Additional studies continued to show effectiveness of nusinersen. The NURTURE open-label extension study evaluated the effectiveness of nusinersen in pre-symptomatic infants, as compared with the historical controls, as it would be unethical to withhold effective treatment for the sake of a research study control group.²⁵ They treated 15 infants with 2 copies of SMN2, and 10 infants with 3 copies of SMN2. Interim data at a mean of 2.9 years of follow-up found that all 25 infants were alive without requiring permanent ventilation, all 25 achieved sitting without assistance, and 22 achieved walking independently.

Subsequent 5-year data analysis in the NURTURE study showed continued effectiveness of nusinersen, with 23/25 subjects now walking independently and continued improvements in their Hammersmith Functional Motor Scale—Expanded (HFMSE). A subgroup analysis divided the infants with two copies of SMN2 into those who had retained reflexes and ulnar compound muscle action potential amplitudes ≥ 2 mV and those without. Better scores were seen in the HFMSE for those children with retained reflexes. The baseline differences between the patient population in the presymptomatic nusinersen trial and presymptomatic gene replacement therapy trial, which prevent direct comparison of those two drug study groups, were highlighted.²⁶

Multiple regional or small cohort reports exist, providing real-world evidence on effectiveness of nusinersen, with earlier treatment leading to greater improvements in motor skills.^27,28,29,30 A large prospective observational study including adult patients from Germany, Switzerland, and Austria (SMArtCARE registry) showed significant improvements from baseline in the HFMSE, the Revised Upper Limb Module (RULM), and 6 min walk test after up to 38 months of treatment with nusinersen in 237 patients.³¹ Patients with milder symptoms showed clinically meaningful improvements in motor skills, whereas weaker and non-ambulatory patients showed more stabilization of their symptoms. Adverse drug reactions in this large cohort were common (64% with at least one adverse reaction) and included lumbar-puncture-related complications.

The SMArtCARE registry also looked at 143 treatment-naive children with early-onset SMA who were then treated with nusinersen for up to 38 months. The noted advancement in motor function for children who were started on treatment at less than age 2 years, and these motor improvements were most prominent in the first 14 months of treatment, with stabilization after that time. The subgroup that were older than age 2 years at start of treatment showed stabilization of CHOP-INTEND scores but no advancement of motor milestones. The authors noted that bulbar and respiratory function did not track with the motor advancements for these patients, with the percentage of patients needing tube feeding and ventilatory support increasing over time.³²

A Polish study of 130 patients with SMA on nusinersen, ages 5 to adulthood, showed increasing percentages of patients who improved in multiple scales of motor function when comparing 14 months and 30 months of treatment.³³ Seventy-one percent of these patients saw clinically meaningful improvement in HFMSE over a 30-month period, and the treatment was well tolerated.

Aspects of medical treatment

Nusinersen was the first high-expense treatment to arrive in the world of neuromuscular child neurology in 2016. The addition of nusinersen to the armamentarium against SMA led to a steep learning curve for teams negotiating prior authorizations as well as guarantee of payments.³⁴ Multiple discussions of cost-effectiveness and ethical considerations of using these high-cost drugs have occurred locally, nationally, and internationally.^35,36

Nusinersen is administered intrathecally four times in the first 2 months (loading doses), and then maintenance doses are given every 4 months. Nusinersen is challenging to administer intrathecally to patients with significant scoliosis, which is typically seen in older patients who have accrued significant weakness. These patients often have to undergo a lumbar puncture using the help of fluoroscopy or CT guidance, which requires collaboration with interventional radiologists and exposure to radiation, although low-dose radiation protocols have been developed.³⁷ Alternative puncture sites, such as cervical and transforaminal lumbar have also been used^38.39. Some patients have reservoirs implanted, which helps ease administration of medication, reduces exposure to radiation, but can be associated with side effects.⁴⁰

Monitoring for the side effects of thrombocytopenia, coagulation abnormalities, and kidney dysfunction requires blood tests and quantitative urine protein before every dose per FDA label, although abnormalities in these tests are generally mild or transient proteinuria.^41,42 The most common side effects of nusinersen are due to the route of administration, with post-lumbar puncture syndrome reported in up to 19% of patients with later-onset SMA.⁴³ Both bacterial and aseptic meningitis have been reported in case reports.⁴⁴ Spinal adhesive arachnoiditis has been described with repetitive administration in one patient.⁴⁵ Further trial-reported side effects included pyrexia, headache, lower respiratory infection, and constipation.^42,46

Adverse reactions less often reported include several cases of asymptomatic increased intracranial hypertension, with several publications suggesting that patients with SMA might be at higher risk for increased intracranial hypertension at baseline.⁴⁷ More recently, a case of symptomatic increased intracranial hypertension after nusinersen administration was reported in an adult with arachnoid cysts.⁴⁸ A recent review of adverse events reported to the FDA database also included transient deafness with nusinersen administration in four patients.⁴⁹

Dosing is the same for every patient, regardless of age or size, at 12 mg. A wearing-off effect has been reported in adult patients on nusinersen therapy, with proposals of shortening the dosing interval, using higher doses, or using alternative medications as potential solutions.⁵⁰ A three-part phase 2/3 trial studying high-dose nusinersen, known as the DEVOTE trial, is underway, with part A results showing that six participants tolerated 28-mg doses with a dosing schedule of three loading disease and then maintenance doses every 4 months.⁵¹

In clinical practice, nusinersen is well tolerated. Several groups have looked at patient’s adherence to nusinersen dosing prior to the approval of risdiplam using administrative databases. These studies were sponsored by competing drug companies, and found varied results that were felt to be due to methodological differences in dealing with large datasets.^52,53 Electronic health record database analysis prior to the availability of risdiplam supported good compliance with nusinersen treatment.⁵⁴

Monitoring SMA patients for treatment effectiveness is often difficult, as improvements are not uniformly picked up on formal scales, especially in older adults. For example, subjective reports including improved stamina and louder voice were not picked up by the HFMSE.^55,56 Ongoing discussions are had with payors to discuss that stabilization of progression is, in fact, an improvement from the typical course of this progressive disease.

Nusinersen administration to a large patient population can place a significant burden on procedure suites, sedation teams, and interventional radiology. The administration of the medication is also burdensome on patients and their families.⁵⁷ The original FDA label for nusinersen included older children and an adult population that was initially not well studied, which led to difficulties in insurance coverage for these treatments. In our practice, nusinersen is now generally readily approved by insurance companies in patients who have not received SMA gene therapy, onasemnogene abeparvovec (OA), due to the real-world evidence that came out after FDA approval.

Risdiplam

A second medication, risdiplam (Evrysdi), was also shown to be effective for treatment of SMA and was initially approved by the FDA in August 2020 for aged 2 months and older, and for infants under 2 months of age in May 2022. Risdiplam is a splicing modifier that allows incorporation of exon 7 into the final transcript of SMN2, increasing the percentage of full SMN2 protein made.⁵⁸ Unlike nusinersen, which is administered intrathecally, risdiplam is an oral medication.⁵⁹

Clinical trials

Safety and effectiveness was evaluated for SMA type 1 in the FIREFISH trial, and in types 2 and 3 in the SUNFISH trial.^60,61 The phase 3 trial for SMA type 1 enrolled 21 symptomatic infants, who were initially divided into low-dose and high-dose cohorts; they were all moved to the higher dose after the fifth infant treated with low dose had disease progression and died during the study. A total of 17 infants were alive at the end of the study, and none of them required permanent ventilation; 4 infants died from respiratory complications. Seven of them were also able to sit unassisted. Preclinical studies showed retinal toxicity in animals, but this was not observed in infants. The 36-month data of the open-label extension of FIREFISH revealed stable or improved motor milestone scores.⁶²

In the trial for symptomatic individuals with SMA type 2 or non-ambulatory individuals with SMA type 3, the final cohort consisted of 120 individuals who received risdiplam and 60 received a placebo. On average, individuals who received risdiplam had improvement in motor scores, whereas those in the placebo group had a decrease in scores. Motor function increases were sustained in the second and third years after treatment.⁶³ This medication also appeared to be well tolerated, with reported increased side effects in the treatment group including fevers, diarrhea, rashes, and arthralgia.

Risdiplam in asymptomatic infants with SMA is currently being evaluated through the RAINBOWFISH trial.⁶⁴ Interim results suggest that this is effective and safe: 18 infants have been enrolled, none have had serious adverse events, and 7 infants who have received risdiplam for at least 1 year’s duration are all making good progress in motor development and are orally feeding well.⁶⁵ Another aim of the study is to determine the correct dose for infants less than age 2 months.

While there are no head to head studies comparing the effectiveness of different SMA therapies, the JEWELFISH study is ongoing and aims to evaluate the safety and effectiveness of risdiplam in patients who were previously treated with other SMA therapies.⁶⁶ Individuals aged 1–60 are enrolled, encompassing SMA types 1–3, and both ambulatory and non-ambulatory people. Interval data analysis did not reveal any severe drug-related adverse events, and motor function has appeared stable on treatment.⁶⁷

There have been several single center or regionally based reports of real-world risdiplam use. Several expanded-access or compassionate-use programs revealed similar risdiplam safety profiles to the clinical trials. In Great Britain, 92 patients with SMA 1 and 2, aged 2–18, in an expanded-access program took risdiplam for 2–21 months. Sixty percent of this group were treatment naive. The most common adverse events were respiratory infections and gastrointestinal disturbances.⁶⁸ In Germany, 111 SMA 1 and 2 patients, aged 3–60 years, received risdiplam as part of a short-term compassionate use program and had similar adverse events to previous studies. Nearly 75% of patients with type 1 SMA and half of the patients with SMA type 2 in that study had previously been on nusinersen, but stopped it due to reported difficulty with access or loss of efficacy (30.6% and 12.2% of patients, respectively).⁶⁹ A US-based expanded access program reported on 149 patients who received risdiplam prior to FDA approval, and also found similar adverse events to the trials.⁷⁰

Risdiplam in adults is generally well tolerated with improvements in patient reports and some improvement in functional scales. Thirty-one adult patients with type 2 and type 3 SMA in Croatia reported improvement or stabilization of strength, improvements in quality of life, and 59% reported some improvement in bulbar function after being treated with risdiplam for up to 2.5 years.⁷¹ A single-center German study of 25 adult patients with reported bulbar dysfunction who largely had SMA2, revealed subjective improvements in bulbar function after 12 months of risdiplam, with a trend toward improvement on the b-ALSFRS-R, an ALS bulbar function score.⁷² In 6 adult patients with SMA type 2 in Northern Ireland, all patients reported skin sensitivity to light with risdiplam treatment, and reported improvements in sense of well-being, strength, and speech, and had improvements in quality-of life and Egen Klassifikation scales, but improvement on the RULM was not seen.⁷³

Real-world risdiplam efficacy data have also been collected in children. In a French study, 28 children with SMA 1–3, ages 4–20, were studied. Ninety-five percent of the patients were on ventilatory support. Twenty patients had been on nusinersen previously, and 2 switched back to nusinersen during the course of the study. After a mean follow-up of 11 months, 53.5% of patients had motor improvements.⁷⁴ A retrospective chart-review study of infants who start risdiplam under 2 months of age is being planned.⁷⁵

Aspects of medical treatment

Risdiplam is an oral medication that is taken once daily, preferably after a meal. Risdiplam should be stored in the refrigerator, but can be kept at room temperature for up to 5 days.⁷⁶ Dosing for infants under 2 months of age is 0.15 mg/kg/day. Dosing for children over 2 months to 2 years of age is 0.2 mg/kg/day. Patients over 2 years of age and under 20 kg should be dosed at 0.25 mg/kg/day, and those over 20 kg should be dosed at 5 mg daily.⁵⁹

In clinical practice, risdiplam is a well-tolerated drug, with some stomach upset with dose titration and declines in the first 4 weeks of treatment.⁷⁷ Other reported side effects include fever, diarrhea, rash, upper respiratory tract infection, pneumonia, constipation, and vomiting.⁵⁹ Monitoring with laboratory tests is not recommended by the FDA, although our group does monitor liver function periodically and a recent Brazilian review recommended monitoring urine protein, renal function, liver enzymes, and platelets every 6 months.⁷⁸ Risdiplam may bring advantages in systemic effects of SMA, such as metabolic dysfunction, although there are no studies at present supporting this.⁷⁹

Early studies showed damage to germ cells in the testes of rat and monkey models, and the risdiplam label suggests consideration of sperm preservation. Subsequent studies sponsored by the drug company have indicated that sperm development continues normally after cessation of the drug, which led to their hypothesis that any problems with sperm development are likely to be a temporary problem in humans.⁸⁰

In clinics at Indiana University, Washington University, and Norton’s Children’s Hospital, after a discussion of the possible side effects, risdiplam is recommended to be started as soon as possible after the initial visit, before copy-number testing has returned, which is often before 7–10 days of life.⁸¹ A small proportion of parents in our practice avoid risdiplam for fear of fertility issues. Some newly diagnosed patients stay on risdiplam monotherapy. Patients are then monitored closely with clinical and physical therapy evaluations in a multi-disciplinary neuromuscular clinic.

In our center, the majority of older patients transitioned from nusinersen to risdiplam if they were ineligible for OA. The most common reason for this transition was due to the mode of administration of nusinersen, as has been reported in other studies.⁶⁸

OA

SMA gene replacement therapy with OA (Zolgensma) was approved in May 2019. OA is approved for children with SMA under the age of 2 years, with the caveat on the FDA prescribing label that it has not been evaluated in patients with advanced SMA.⁸² OA is a self-complementary adeno-associated virus serotype 9 (scAAV9) vector, which contains DNA encoding the SMN1 gene, and is under the control of a chicken β-actin promoter to promote rapid production of SMN protein.⁸³

Preclinical

Early studies in mouse models showed promising results; mice who received a viral product with inclusion of a gene that could produce SMN protein had longer survival, reduced motor neuron disease, and a higher amount of SMN protein detected in the spinal cord compared with untreated SMA control mice.^84,85 As animal trials advanced, AAV9 was chosen as a vector due to its ability to pass through the blood-brain barrier and to be taken in by motor neuron cells.^86,87 The SMN1 gene was introduced under control of a strong promoter to increase gene expression once incorporated into the DNA. Initial studies of the SMN1-containing scAAV9 in mouse models showed promising results. Mean survival of mice with deletion of SMN1 and two copies of SMN2 is only 13 days. With administration of the scAAV9 product shortly after birth, mice were able to achieve a normal lifespan.⁸⁸

Clinical trials

The START trial was a phase 1 trial to determine safety of OA treatment as a primary outcome, and time until death or need for permanent ventilation as a secondary outcome.⁸³ Fifteen patients with SMA type 1 and 2 copies of SMN2 were enrolled in the study, all symptomatic and <8 months of age. The AAV product was given as a single intravenous (i.v.) infusion; 3 infants received a low dose of 6.7 × 10¹³ vg per kg (cohort 1) and 12 infants received a high dose of 2.0 × 10¹⁴ vg per kg (cohort 2). All of the infants were alive and without permanent ventilation at the study endpoint of 20 months, which is a significant improvement compared with historical cohorts from the natural history studies; data from the NeuroNEXT study estimates that only 8% of untreated infants with SMA type 1 would achieve this goal. The medication was also generally well tolerated. The first patient was noted to have an elevation in aminotransferases, so the rest of the patients were treated with prednisolone 1 mg/kg daily for ∼30 days or until liver enzymes normalized. Three infants in cohort 2 also developed elevated aminotransferases. Other adverse effects were generally mild and appeared to be unrelated to treatment, such as respiratory illnesses. Of note, infants in the study also showed improvement in CHOP-INTEND scores indicating improved motor function. Those in cohort 2, who received a higher treatment dose, had a more significant improvement in scores throughout the course of the study. Only a single infant in cohort 2 did not achieve independent sitting at the study endpoint; this infant was also the oldest at the time of enrollment at 7.9 months. The other 11 infants could all sit independently for at least 5 s, swallow, and speak. These results are striking compared with historical cohorts in which no untreated infants were able to achieve significant improvement in motor milestones over time. Based on this striking data, OA received FDA approval in May 2019.

A 5-year extension study enrolled 13 of the infants from the phase 3 trial, and all 10 patients who received the higher dose were alive and did not require permanent ventilation at the time of data analysis.⁸⁹ All 10 of these patients also made progress in motor development, 2 of whom achieved independent standing. Spinal cord samples from 2 infants who passed away from their SMA and received OA showed that the drug was distributed well throughout the central nervous system, providing some reassurance that intravenous therapy was effective in delivering medication to essential areas for uptake.⁹⁰

Following this initial success, phase 3 trials continued to show similarly promising results. There were two main arms of this study: the STR1VE-US study which enrolled 22 patients in the US, and the STR1VE-EU study which treated 32 patients in Europe.^91,92 Both studies had inclusion criteria of symptomatic infants less than 6 months of age, with either 1 or 2 copies of SMN2. OA at a dose of 1.1 × 10¹⁴ vg per kg was administered i.v. to each infant, as well as prednisolone for at least 30 days to temper liver inflammation. In the STR1VE-US study, the co-primary outcomes were achieving independent sitting for at least 30 s at age 18 months, and survival without permanent ventilation at 14 months. Twenty of the 22 infants were alive without permanent ventilation at 14 months, as compared with 26% in a historical cohort as per the PNCR study results. Thirteen achieved the ability to sit unassisted for at least 30 s, a milestone essentially unheard of prior to therapies for SMA type 1. A secondary endpoint of maintaining the ability to thrive, defined as the ability to swallow thin liquids and maintain weight above the third percentile without supplemental tube feeds, was achieved by 9 of the 22 infants. There were 3 drug-related serious adverse events: 2 infants with significantly elevated aminotransferase levels, and 1 infant with hydrocephalus who required VP shunt placement. Of note, hydrocephalus was found to be 4 times more common in infants with SMA as compared with healthy controls, as per a case-control study using data from untreated infants.⁴⁷ In addition, 1 infant died at 7.8 months of age from a respiratory illness, and 1 infant required permanent ventilation at age 11 months. The STR1VE-EU study also had similarly optimistic results. Their primary endpoint was defined as the ability to sit independently for over 10 s by 18 months of age, and 14 of the 32 infants achieved this. They also noted that 31 of the 32 infants were alive without permanent ventilation at age 14 months. One child died from hypoxic-ischemic brain injury related to a respiratory infection during the course of the trial. Serious drug-related adverse events included fevers, increased aminotransferases, and gastroenteritis. These studies again proved that OA therapy is generally well tolerated and that outcomes for children with SMA type 1 are far superior compared with a historically untreated cohort.

It was also hypothesized that pre-symptomatic infants identified through prenatal testing or newborn screening (NBS) could benefit significantly from gene replacement therapy and remain asymptomatic if they are treated early, before damage to the lower motor neurons begins to occur.⁹³ The phase 3 SPR1NT trial investigated this theory.⁹⁴ There were 2 separate arms of the study, one treating infants with 2 copies of SMN2 predicted to have SMA type 1, and another arm enrolling infants with 3 copies of SMN2 and thus likely to have a less-severe phenotype. There were 14 asymptomatic infants with 2 copies of SMN2 who were each treated with OA prior to age 6 weeks. All of these infants achieved the primary endpoint of sitting independently for at least 30 s at age 18 months, and all were alive without need for permanent ventilation at the conclusion of the study. Some of the secondary endpoints of the study also revealed promising results: 13 of these infants maintained weight above the third percentile throughout the study with exclusively oral feeding, and 11 of them were even able to stand independently. In addition, 15 asymptomatic patients with 3 copies of SMN2 were also treated. The primary endpoint was independent standing for at least 3 s by age 24 months, and all achieved this. For comparison, only 24% of patients in the historical PNRC cohort achieved this. Fourteen of the 15 children were actually taking at least 5 independent steps by age 24 months . All of the children were alive and none required any type of ventilatory support by the end of the trial. A post-hoc analysis of the SPR1NT data also revealed that all 29 patients were fully orally feeding and had normal swallow.⁴³ These results speak to not only the effectiveness of gene replacement therapy, but also the benefits of pre-symptomatic diagnosis and treatment of SMA.

The RESTORE registry is a “prospective, multicenter, multinational, non interventional, treatment-neutral registry of patients with SMA, which was designed to integrate with existing real-world data (NCT04174157)”.⁹⁵ As of a May 23, 2022 data cut, 168 patients who received OA monotherapy are described out of a pool of 385.⁹⁶ Survival without permanent ventilation was achieved in 90% of SMA patients with 2 copies SMN2 at 2 years of life, and in 100% of patients with 3 copies of SMN2. Infants diagnosed via NBS had a 100% rate of event-free survival, regardless of SMN2 copy number, an overall lower age of attaining their first motor milestones, supporting faster motor development, higher CHOP-INTEND scores than those diagnosed symptomatically, and had overall higher scores than infants who were diagnosed clinically. The majority of patients fed by mouth (96.6%), with 13% reporting some swallowing/feeding difficulty and 4.7% with pneumonia or aspiration pneumonia. Other groups are presenting data on patients treated for SMA at earlier ages than in the trials. Of 65 patients with SMA treated at less than age 6 weeks with OA or nusinersen, none required permanent ventilation or total nutrition via tube feeds.⁹⁷ Patients with 2 copies of SMN2 were more likely to have gross motor delays.

There is an ongoing 15-year follow-up study for infants who received OA in the phase 1 START trial, and 13 of the 15 infants are enrolled.⁹⁸ As of May 2022, they were all still alive and none had severe treatment-related adverse events. Notably, 3 patients gained the new milestone of standing with assistance. Nine patients were receiving add-on therapy, either nusinersen or risdiplam.

There is also a long-term follow-up study enrolling patients who have been involved in the STR1VE, SPR1NT, and STRONG studies.⁹⁹ As of May 2022, 81 patients were enrolled in the study including 38 symptomatic and 25 pre-symptomatic individuals with SMA type 1 who received i.v. gene replacement therapy, and 18 with SMA type 2 and non-ambulatory who received intrathecal administration. Only 1 patient with SMA type 1 progressed to require permanent ventilation defined as BiPAP >16 h per day, and no unexpected adverse effects have come to light in this study. All patients maintained previous motor milestones and 27 achieved new milestones.

Ongoing studies continue to evaluate the effectiveness and safety of OA. The STRONG study is a trial of intrathecal administration in non-ambulatory patients with 3 copies of SMN2.¹⁰⁰ OA is weight based, so the concern is that older and thus larger patients would be at increased risk for adverse effects with systemic therapy given the higher viral load. Intrathecal therapy allows the medication to be administered at lower doses since it is more effectively transmitted to the spinal cord this way. Adverse events in the phase 1 trial included hepatotoxicity, thrombocytopenia, and mild cardiac symptoms. Long-term follow-up studies support durable efficacy, with half of the patients in the intrathecal administration group achieving clinically significant improvements on their HFMSE testing, and none of the patients from that cohort requiring feeding support.¹⁰¹

In the United Kingdom, OA is approved for patients with genetically confirmed SMA type 1, or infants up to age 12 months identified pre-symptomatically with less than 3 copies of SMN2.^102,103 This allowed for study of patients receiving OA who ranged in age from 0.6 to 89 months, and patients weighing between 3.2 and 20.2 kg. Patients treated at a younger age were more likely to have a significant increase in their CHOP-INTEND scores. Heavier patients (≥13.5 kg) were more likely to have higher transaminases and require higher and longer steroid treatments. Asymptomatic troponin-I elevations and thrombocytopenia were seen in 33.7% and 71.7% of patients, respectively.

The SMART trial looked at patients aged 1.5–9.1 years and weighing from 8.5 to 21 kg. Over 83% of patients experienced asymptomatic hepatotoxicity, but this was similar across the weight ranges. Prednisolone was continued at higher doses than used in past trials and also for longer durations (median 175 days) for these patients. These older patients maintained their motor function over a 52-week period or showed slight improvements.¹⁰⁴

A meta-analysis looked at the overall safety and efficacy of OA in a total of 250 patients who were mostly SMA type 1, treated at a range of 0.5–59 months. A total of 86.9% of patients achieved a CHOP-INTEND score of ≥40 points at long-term follow-up, which is significantly improved compared with the natural history studies, during which a score of ≥40 points was rarely achieved and not maintained. In patients treated at older than age 8 months, there were higher rates of thrombocytopenia and elevation of transaminases.¹⁰⁵

Further real-world evidence for OA safety and efficacy is plentiful. In a review of the first 21 children treated in the state of Ohio, 89% of patients with repeated motor assessments showed improvement.¹⁰⁶ In Israel, 25 patients with SMA types 1 and 2 showed gradual improvements, with CHOP-INTEND scores increasing a median 13 points. Children who were treated at a pre-symptomatic stage showed more improvements in motor function.¹⁰⁷ In an Australian experience, 21 children with 2 and 3 copies of SMN2 showed improvements in motor milestones (76%) as well as in bulbar function (95.2%).¹⁰⁸ Eleven patients in Qatar with SMA type 1 or 2 received OA. Those with SMA type 1 were treated first with nusinersen. Type 1 patients receiving non-invasive ventilation were able to be weaned off respiratory support after a mean of 10.75 months after OA treatment. Hospitalizations were also decreased after OA therapy.¹⁰⁹

Aspects of medical treatment

In recent years, discussion about treatment with OA most often occurs with an infant identified via NBS or prenatally. Treatment with OA requires an extensive consent discussion with families, as well as discussion about steroid treatment and potential side effects. The steps in obtaining a prior authorization as well as generally obtaining a guarantee of payment for the drug require days to weeks, and must be clearly explained to the families.

In our center’s experience, prenatal identification leads to the benefit of allowing treatment discussion prior to the birth of the child. This allows families to feel comfortable with their treatment choice, as well as to arrange delivery at their local treatment center, which allows immediate confirmatory testing to be sent as well as an immediate introduction to the neuromuscular team.

Before administration of OA, baseline labs, including liver function tests, platelets, and troponin-I are obtained to assure no underlying abnormalities. On occasion, elevated transaminases are seen at baseline but may be reflective of the mildly high CK that can be seen in patients with SMA.¹¹⁰ AAV9 antibody titers must be checked prior to ordering of the drug. Two laboratories are used for this testing, Athena Diagnostics and Cellular Technology Limited (CTL). Seronegativity per Athena’s test is ≤ 1:25, while CTL’s seronegativity cut-off is < 1:50. On occasion, babies will have elevated antibody titers, felt to be a transplacental transfer from mother.¹¹¹ Elevated antibodies in the newborn period nearly universally decline to treatment-ready levels over time, with recommended timing of repeat testing at the discretion of the ordering physician.

Prednisone/prednisolone is started the day prior to the OA infusion at 1 mg/kg/day, and continued for at least 2 months. Prednisone dosing is adjusted in response to abnormal laboratory values. It is encouraged to involve hepatology or gastroenterology if transaminases reach 2× ULN.⁸²

OA is given as a one-time infusion, typically in an outpatient infusion center. OA is administered as a single dose of 1.1 × 10¹⁴ vg per kg of body weight, given i.v.⁸² Coordination must be arranged with pharmacy, as the medication is typically prepared only once an i.v. is successfully placed. OA is dosed based on weight ranges, and a final weight check should be obtained just prior to final order of the medication. Infusion center nursing, accustomed to weight adjusting medications on day of infusion, must be educated that the dose cannot be adjusted on the day of infusion. The patient must be healthy without signs of infection before administration. The infusion is given over 1 h and is generally well tolerated. Patients are typically observed for some hours after the infusion.

The monitoring after OA administration is intensive. Weekly labs and frequent visits are needed for 2 months, and then labs every 2 weeks for at least a month after that. Laboratory monitoring includes monitoring of hepatic function, platelets, coagulation factors, and troponin-I. Transduction of the anterior horn cells does not occur immediately and may not be complete. Clinically, patients are watched for progressive feeding dysfunction, respiratory distress, or progression in weakness that can occur while the OA becomes effective. Families are instructed to follow social isolation procedures.¹⁰⁶

OA vector DNA is shed in saliva, urine and feces after OA administration, with duration of feces the longest at 30 days.¹¹² Families are instructed to follow good hand hygiene practices and seal diapers in disposable trash bags for 30 days post-infusion to avoid developing their own immunity to the AAV vector in the future.⁸² Parents who cared for children who received Duchenne muscular dystrophy gene replacement therapy using rAAVrh74 had a 3-fold greater risk of seropositivity than control.¹¹³

While OA is generally well tolerated, there are ongoing studies for surveillance of additional adverse effects over time. The most common significant adverse drug-related event appears to be liver toxicity, ranging from asymptomatic elevation of liver enzymes to fatal hepatotoxicty.⁸² It is recommended to administer prednisolone for at least 30 days to mitigate this reaction per clinical trial data. Post-marketing studies and registries found that, while aminotransferase elevations were common, they were often mild, and commonly occurred during the time that systemic steroids were tapered. In one study, two infants met criteria for acute liver failure, and both responded well to increased prednisolone and completely recovered.¹¹⁴

Several cases of thrombotic microangiopathy (TMA) have been reported, including fatal cases.^115,116 TMA typically occurs about 1 week after treatment with OA, and can occur despite prednisolone treatment, and presenting signs include vomiting, edema, decreased urine output, and hypertension. Platelet counts should be monitored after gene replacement therapy and thrombocytopenia may also cause suspicion for TMA in providers.

Studies have also shown dorsal root ganglia toxicity in non-human primates when administered intrathecally and in both non-human primates and piglets when administered via i.v.^117,118 A mouse-model of SMN overexpression showed a delayed motor dysfunction secondary to neurodegeneration.¹¹⁹ These findings led to a temporary hold on intrathecal studies in humans from 2019 to 2021. A phase 1 trial revealed the safety of intrathecal OA in 32 patients, and a further study is ongoing.^100,120

Although AAV is generally a virus that is not felt to integrate into host DNA, creating instead an episome in the cell, there is theoretical risk of integration of AAV9 into the host DNA, as has been seen in mouse models. A case of an epithelioid spinal cord tumor in a 16-month-old with SMA who received OA at age 2 months has been recently reported, with OA nucleic acid signal seen in many of the tumor cells, but without clear evidence of AAV insertion into the DNA.¹²¹ Due to the risk of insertional oncogenesis, an Australian center monitors patients with annual abdominal ultrasound and α-fetoprotein levels.¹⁰⁸

Choice of therapy

In the absence of a double-blind, randomized controlled trial comparing the three treatments, the choice of treatments depends largely on parental preference after an in-depth discussion of risks and benefits of all the medications (Table 1), any accompanying medical problems, including the status of the patient’s spine, the patient’s level of concern for risk of fertility problems, and level of concern about frequent lab work, as well as the child’s severity and insurance coverage. The primary goal is to start treatment as soon as possible.¹²² The preferred treatment by families has varied over the past 5 years, depending on the medications available.

Table 1.

SMA treatment considerations

	Nusinersen	Onasemnogene abeparvovec	Risdiplam
Cost	$375,000/year	$2.125 million × 1	$340,000/year
Time to initiation of therapy	days–weeks	days–weeks	days
Administration	intrathecal	i.v.	oral
Frequency of administration	loading doses × 4, then every 4 months maintenance	one-time	daily
Age range	any	less than 2 years (US)	any
Risks	•chronic lumbar punctures •thrombocytopenia •coagulation abnormalities •renal toxicity •possible increased risk of hydrocephalus •possible repeated sedations •possible radiation exposure (fluoroscopy)	•hepatotoxicity, severe liver failure •thrombotic microangiopathy •systemic immune response •elevated troponin-I •duration of therapy unknown	•fever, diarrhea, rash •cannot take multidrug and toxin extrusion (MATE) transporters •fertility problems in animal models •compliance risk
Advantages	•no need for daily medication	•one-time treatment	•ease of administration •potential systemic treatment effects

As other drugs have become available, patients have decreased their use of nusinersen with switched to risdiplam due to mode of delivery. Sparse data are available in this area, but published papers report general satisfaction with a switch from nusinsersen to risdiplam as well as non-inferiority in 17 adult and pediatric patients.^123,124 Pharmaceutical company-sponsored indirect treatment comparisons of nusinersen and risdiplam have been published, and were generally inconclusive.^{125,126,127,128}

A typical first visit in our center for an SMA NBS-positive infant generally occurs in the first week of life. The first visit includes a discussion about the disease SMA, the types of SMA, and the historical untreated prognosis of the various types. Discussion of the treatment options are initiated, but often minimal due to the overwhelming nature of the diagnosis. AAV9 titers are drawn, in the case that the family chooses gene replacement therapy, as well as confirmatory SMA testing and baseline labs to rule out underlying medical conditions.

At our center, a second appointment is arranged within a week if the family is unsure of medication choices. At this point, SMN2 copy number and AAV9 titers have results available. Medication options can be reviewed. Choice of medication also involves discussion of current payor activities in the region for initial treatment and discussion of payment difficulties with potential add-on therapies.

With infants newly diagnosed with SMA, most families prefer not to use nusinersen due to the intrathecal administration as well as frequent blood and urine testing. The decision between risdiplam monotherapy and obtaining gene replacement therapy is more complicated. Many families choose gene replacement therapy in order to attempt to avoid the need for an add-on therapy, such as risdiplam. Concerns expressed by families about risdiplam include fear that the medicine would become unavailable, stop being made, or could have long-term side effects that are unknown. Discussion about the side effects of gene replacement therapy can be extensive due to the novelty of the mechanism of action for families as well as the risks involved (Table 1). At our institution, once the family decides to pursue gene replacement therapy, the financial group for high-cost drugs is informed and begins work on prior authorizations and guarantee of payment for the drug.

OA is approved for children with SMA and two to three copies of SMN2 by the majority of payors with ease. Children with a SMN2 copy number outside of this range can have difficulty getting approval. Infants with type 0 have been treated with nusinersen, risdiplam, and gene replacement therapy with small and mostly motor improvements,^129,130 but recent RESTORE data reveal two one-copy SMN2 babies treated with OA who pulled to stand.¹³¹ Currently there is no consensus on the appropriate treatment plan for infants with type 0 and, to complicate matters, they are often excluded from payor’s plans as they often need invasive ventilatory support.^129,130

Babies with four copies or more of SMN2 present similar challenges. These patients are also excluded from some payor’s plans due to having too high of a copy number, although they often may be treated once they have symptoms. In one large case review, 31% of patients with SMA type 3, which has onset after 18 months of life, had four copies of SMN2.¹³² A 2023 Italian study also showed a variability of types of SMA in those with four copies of SMN2, ranging from type 2 to type 4, and these patients showed a decline over time.¹³³ Although initial expert opinion reported gene replacement therapy treatment for babies with four copies of SMN2 was unnecessary, this was revised.^134,135 Extreme stress for these four-copy families, leading to being lost to follow-up, has been reported.¹³⁶ Wait-and-see strategies that were initially recommended have been rejected, due to reports of patients not being able to regain deficits, even with prompt initiation of treatment after appearance of signs or symptoms.¹³⁷

Combination and switching therapies

Not all cells are transduced in animal models of SMA gene replacement therapy, so improvement of symptoms with an SMN2-modifying drug after OA treatment could hypothetically be seen, and are often reported by patients.^138,139 A classification for describing combination therapies for SMA disease-modifying therapies has been proposed and validated using the RESTORE registry.¹⁴⁰ Combination and add-on therapies have been found to be generally well tolerated, but larger prospective reviews and randomized controlled trials are needed.^141,142,143

Combination therapies are being studied in the RESTORE registry⁹⁵ and other registries such as the MDA neuroMuscular ObserVational Research database.¹⁴⁴ The RESPOND study is examining the effectiveness of nusinersen treatment as an add-on to OA treatment, and an interim analysis reports improvements in motor function and typical adverse events.¹⁴⁵ Improvements in bulbar function were reported as improved by families, but not seen by investigators. In a long-term follow-up study of OA with 81 patients who received i.v. or intrathecal OA, 24 patients had been on an add-on therapy, and half of those 24 patients did not achieve new motor milestones.¹⁰¹ In another report, which included 10 patients who received a therapeutic dose of OA in the original trial, 6 patients used an add-on therapy, including 2 patients on risdiplam, three switching from nusinersen to risdiplam, and one stopping nusinersen.⁹⁸ Studies looking at the safety and effectiveness of risdiplam after OA therapy are pending.^146,147

A small study compared cohorts of children receiving nusinersen monotherapy versus children receiving nusinersen who switched to OA, finding no significant difference in these groups after 12 months of therapy onset.¹⁴⁸ The SmArtCARE registry reported on 76 children with SMA type 1 and 2 who received OA. In a subgroup of the patients who received nusinersen prior to OA treatment, 21 patients had motor evaluations performed before OA, at the time of OA, and 6 months after. In this group, the increase in mean score increased by 6.8 points in the 6 months before OA and 6.6 points in the 6 months after OA. Four children who switched from nusinersen to OA had a 10-point mean increase in HFMSE scores, but a combination effect of nusinersen and OA could not be ruled out due to the long half-life of nusinersen.¹⁴⁹

NBS

SMA was added to the Recommended Uniform Screening Panel in July of 2018, and, as of January 2, 2024, after a great deal of advocacy work by patients, families, disease groups, and physicians, all US states have an SMA NBS program.¹⁵⁰ Now, patients are identified either by NBS or by prenatal testing. NBS has been pivotal in diagnosing and treating infants with SMA pre-symptomatically or early in symptoms.¹³⁴ In a review of clinical trials of infants identified by NBS up to January 2023, 34/35 patients with three copies of SMN2 achieved independent ambulation, and of 34/57 patients with 2 copies of SMN2 achieved walking.¹⁵¹

The American College of Obstetricians and Gynecologists recommends carrier screening for SMA for all women who are pregnant or are considering becoming pregnant.¹⁵² In our Indiana experience, many families report that testing is not presented to them as needed, resulting in few families getting carrier testing. Inpatient pathways have been developed for newborns found to have SMA prenatally or on NBS.¹⁵³ In our center, prenatal genetic counseling and a prenatal meeting with a neuromuscular neurologist to discuss treatment options is offered.

Future directions and conclusions

Both individuals with SMA and parents of children with SMA in the US have expressed interest in fetal gene replacement therapy trials in SMA.¹⁵⁴ SMA is thought to be an excellent candidate for in utero gene replacement therapy, but the safety of these drugs in pregnant people is not yet clear.¹⁵⁵ In one animal study, fetal mice were given an in utero intracerebroventricular injection of SMN-containing AAV9 vector; the mice lived longer and were noted to have an increased number of motor neurons on pathology.¹⁵⁶ While studies of fetal treatments are especially complex given that effects on both the mother and fetus are unknown, it would appear that there is sufficient interest and support within the SMA community to consider this as a possibility.

The best timing of treatment in preterm infants remains a discussion.¹⁵⁷ Gene replacement therapy has been delivered successfully to twins with one copy of SMN2 at 33 weeks gestation with a report of normal neurologic development.¹⁵⁸ There is a balance of early delivery risk with improved outcomes with earlier treatment and less negative immune response to gene replacement therapy with risks of end-organ injury, including developmental concerns, especially due to the steroid treatment that accompanies gene replacement therapy.¹⁵⁹

Exciting work is ongoing to improve SMA treatment, including research into focused delivery of treatment to the CNS to avoid side effects in peripheral organs that may occur with i.v. administration. This is particularly important for older patients who will require a significantly higher amount of OA, as dosing is weight based.¹⁰⁰ Alternative vectors, capsids, and promotors, as well as engineered virus-like particles are being studied to reduce toxicity and improve transduction efficiency.^{160,161,162,163}

Inequities in access and infrastructure difficulties remain for patients in the US and abroad.^164,165 Some database analyses indicate that large numbers of children under the age of 2 years with SMA have not received OA.¹⁶⁶ It is proposed that numerous challenges may contribute to this discrepancy including socioeconomic status of the patient, differences in NBS process, and availability of specialists varying by location, and the complexity of getting this medication approved, although of course additional analysis is needed to evaluate these issues in detail and try to close these gaps so that all patients are getting equitable treatment.¹⁶⁶

It is also important to recognize that there is a completely novel population of children with severe forms of SMA post-treatment who will require ongoing care. Gene replacement therapy use in humans itself is new, and it is unknown whether the effects are lifelong, and whether additional drug-related events could occur over time. In addition, there could be other systemic effects of SMA that present later in life and were previously unrecognized, as this will be the first group of children with severe SMA who are living a long lifespan.¹⁶⁷ While it is exciting and a reason to celebrate, it is also a reason to continue regular medical follow-up for individuals who have received OA, and underscores the importance of having continued post-market studies and registries.

There have been many exciting developments in the care of patients with SMA over the past decade. Early treatment is key. Studies are underway to evaluate the best treatment course for SMA using the approved medications, as well as to evaluate the effectiveness of several SMN-independent treatment molecules.