Spinal muscular atrophy (SMA) was once the leading genetic cause of infant death and is now treatable with three very different drug approaches: a one-time gene therapy, an infused antisense drug and a daily oral medicine. Novartis's Zolgensma sits at the center of the market's biggest debates, including its price of more than $2 million, its liver-safety warning and a 2019 data-manipulation episode that the FDA reviewed.
Core Patient Information
What is spinal muscular atrophy?
Spinal muscular atrophy (SMA) is a rare, progressive neuromuscular disease in which motor neurons in the spinal cord degenerate, causing muscle weakness and wasting. NICE describes it as a genetic mutation in the SMN1 gene on chromosome 5q and estimates that about 1 in 10,000 people are born with SMA. In the United States, roughly 9,000 people live with the condition, and roughly 450 to 500 babies are born with SMA each year.
Weakness affects the muscles closest to the trunk first, including those used for sitting, crawling, swallowing and breathing. Intelligence is typically unaffected. Severity ranges from infants who cannot sit or breathe unaided to adults with mild weakness. Until 2016 there were no approved disease-modifying drugs; today three therapy classes exist, which has made SMA one of the most closely watched rare-disease markets and a test case for newborn screening and one-time gene therapy.
What causes spinal muscular atrophy?
SMA is caused by loss or mutation of both copies of the SMN1 gene, which produces the survival motor neuron (SMN) protein that motor neurons need to survive. Most patients have a deletion of SMN1 exon 7. A near-identical backup gene, SMN2, makes a smaller amount of working SMN protein, and NICE notes that the number of SMN2 copies is inversely related to severity and can help predict the course of disease.
For example, babies with type 1 SMA typically have two SMN2 copies, whereas people with type 3 typically have three or four. Without enough SMN protein, motor neurons in the spinal cord die and the muscles they control weaken. Nothing a parent does during pregnancy causes SMA. The two main drug strategies reflect the biology: Spinraza and Evrysdi increase SMN protein by modifying SMN2 splicing, while Zolgensma and Itvisma deliver a working copy of the SMN1 gene.
Is spinal muscular atrophy inherited?
Yes. SMA is autosomal recessive, so a child must inherit a non-working SMN1 gene from each parent. Parents are usually healthy carriers with no symptoms, which is why many families have no history. When both parents are carriers, each pregnancy has a 25% chance of an affected child, a 50% chance of a carrier child and a 25% chance of an unaffected non-carrier. A pediatric physical therapy fact sheet gives the incidence as about 1 in 11,000 live births.
Carrier screening before or during pregnancy is recommended by professional groups, and a New York newborn screening study suggested the true incidence may be lower than older estimates, possibly because of carrier testing and prenatal diagnosis. A small share of cases arise from new mutations, and siblings of an affected child may be tested early because treatment works best before symptoms start. Genetic counseling helps families weigh reproductive options.
What are the early symptoms of spinal muscular atrophy?
In infants with the most severe form, parents may notice a "floppy" baby who has poor head control, a weak cry, little spontaneous movement and trouble feeding or breathing. Legs are often weaker than arms, and the chest may look bell-shaped because breathing muscles are weak. In later-onset forms, early signs include delayed motor milestones, difficulty sitting or standing, frequent falls, trouble climbing stairs and tremor in the fingers.
In type 3, weakness appears between 18 months and 10 years. Reflexes are typically reduced or absent. Because the disease progresses and lost motor neurons cannot be replaced, the window for treatment is narrow: therapies work best when started before or soon after symptoms begin. Any infant with weak muscle tone, unusual breathing or loss of previously gained skills should be evaluated urgently, and a genetic test can give a definitive answer quickly.
What are the different types of spinal muscular atrophy?
SMA is classified by the age of onset and best motor function achieved. NICE groups it into five types, 0 to 4. Type 0 is the rarest and most severe, affecting babies before birth. Type 1, the most common, appears in the first six months of life; infants cannot sit unsupported and typically face life-threatening respiratory failure without treatment. Type 2 begins between about six and 18 months, and children can sit but usually never walk independently.
Type 3 appears from 18 months to 10 years, with patients typically able to walk, although many lose this ability over time. Type 4 starts in adulthood with mild to moderate weakness. Because treatments are now started in presymptomatic babies identified by newborn screening, the traditional types are increasingly supplemented by genetic information, especially SMN2 copy number, which predicts the likely course of disease for a screened infant.
How is spinal muscular atrophy diagnosed?
Diagnosis is made by genetic testing. A blood test looks for a deletion of exon 7 in both copies of the SMN1 gene, which confirms the diagnosis in the great majority of cases; the fact sheet notes that a small remainder have other variants that require sequencing. Laboratories also report SMN2 copy number, which helps predict severity and guide treatment decisions. Muscle biopsy and nerve conduction studies, once standard, are rarely needed.
In symptomatic children, diagnosis is often made by a neurologist or geneticist after clinicians notice low muscle tone or delayed milestones. The biggest unmet need is speed: motor neurons lost before treatment are not recovered, so delays of weeks matter. That is why newborn screening has become the preferred route for diagnosis, and why families with an affected child are advised to test siblings and plan early treatment decisions with specialists.
What is newborn screening for spinal muscular atrophy?
Newborn screening tests a few drops of blood collected from a heel prick for the SMN1 deletion that causes most SMA. The federal advisory panel added SMA to the Recommended Uniform Screening Panel in July 2018, following FDA approval of the first effective therapy, and New York began screening on October 1, 2018. In that state's first full year, 8 of 225,093 infants screened had two missing SMN1 copies.
Some programs also measure SMN2 copy number to predict severity. A pediatric physical therapy fact sheet reported that 36 states offered screening by 2021. Screening lets treatment begin before symptoms, which matters because outcomes are better with earlier treatment. In England, by contrast, there is currently no newborn screening programme for SMA per NICE, so access depends on where a child is born.
What is the life expectancy for someone with spinal muscular atrophy?
It depends on type and treatment. Without treatment, type 1 SMA was typically fatal or required permanent ventilation in early childhood; a policy briefing describes SMA as a condition that can lead to death before the age of two. Children with type 2 often survive into adulthood with significant disability, and type 3 and 4 patients have longer survival, although quality of life varies.
The picture is changing: infants treated in the presymptomatic stage can reach milestones that were once impossible, and some have been followed for several years. Roche says Evrysdi has treated more than 21,000 patients. Long-term outcomes remain uncertain because the first therapy was approved only in 2016, and no one yet knows how durable a one-time gene therapy is across decades. Respiratory care, nutrition, orthopedic management and physical therapy remain essential in all types.
What is the difference between Zolgensma, Spinraza and Evrysdi?
All three raise SMN protein, but they differ in mechanism, route and schedule. Spinraza (nusinersen), approved in 2016, is an antisense drug injected into the spinal fluid; its high-dose regimen uses two 50 mg loading doses 14 days apart, then 28 mg every four months in treatment-naive patients, and the FDA approved it in March 2026. Evrysdi (risdiplam), approved in August 2020, is a daily oral SMN2 splicing modifier for patients aged two months and older.
Zolgensma, approved May 24, 2019, is a one-time intravenous gene replacement for children under two, and its intrathecal sibling Itvisma was approved in November 2025 for patients two and older. Spinraza and Evrysdi are chronic therapies that must be continued, while Zolgensma is a single dose carrying a boxed liver warning. Choice depends on age, severity, safety tolerance, convenience and insurance coverage.
Market and Controversy Information
Why does Zolgensma cost more than $2 million?
Novartis, which sells the therapy, priced Zolgensma at $2.125 million when it launched in 2019. Novartis had acquired AveXis for $8.7 billion in 2018, and it argued that a one-time cure-like therapy should be valued against lifetime costs of chronic drugs and care. ICER stated that the price fairly aligns with the benefits and falls within the upper bound of its value-based price benchmark range.
The newer intrathecal Itvisma lists at $2.59 million, compared with $2.1 million for Zolgensma. Novartis says that at this price it costs 35% to 46% less than chronically dosed treatments over 10 years. That comparison is the company's, and independent verification is limited. Payers and critics point to the small number of treated patients with long-term follow-up when questioning the price, and long-term durability remains the central open question.
What was the Zolgensma data-manipulation controversy?
In 2019 the FDA disclosed that Novartis's AveXis unit had submitted a license application containing manipulated data from an animal test. AveXis told the agency on June 28, 2019 that personnel had manipulated data from a mouse potency assay used as a product release test during early clinical development. The company said two senior executives allegedly altered or instructed others to alter a small amount of raw data, and that both had left.
The FDA inspected AveXis's San Diego laboratory and issued a Form 483, and its biologics chief Peter Marks said the agency was weighing civil or criminal penalties. Novartis said the data were a small portion of the submission and related to an older process no longer in use. The FDA maintained that Zolgensma was safe and effective. It later ended its review without penalties, classifying the matter as Voluntary Action Indicated.
When did Novartis and AveXis learn about problems with the Zolgensma data?
According to an FDA memo, AveXis appears to have become aware of the data manipulation as early as March 14, 2019, about ten weeks before the FDA approved Zolgensma on May 24, 2019. The company did not tell the FDA until late June. The agency's Wilson Bryan wrote that if the issue had been disclosed beforehand, approval would probably have been delayed beyond the May 31 target date.
Novartis chief executive Vas Narasimhan has said the company learned of the issues in mid-March and completed a preliminary investigation in early May. Novartis said it shared findings with the FDA once it had interim conclusions from its investigation. It later committed to notify the FDA within five business days of any credible data-integrity allegation affecting a pending application. The FDA said it had no reason to doubt the product's safety or effectiveness, and no financial penalty was imposed.
What are the most serious safety risks associated with Zolgensma?
The most serious risk is liver injury. The U.S. label carries a boxed warning for serious liver injury and acute liver failure, noting that cases with fatal outcomes have been reported. Novartis disclosed in 2022 that two patients died of acute liver failure in Russia and Kazakhstan during the corticosteroid taper, which it described as the first fatal cases of that kind, and the label was updated.
Patients receive corticosteroids before and after infusion, and liver function is monitored for at least three months. Other labeled risks include systemic immune responses, low platelet counts and thrombotic microangiopathy, a blood-clotting disorder that has been reported within the first weeks after infusion. Patients with preexisting liver impairment may be at higher risk. In 2019 the FDA also placed a partial clinical hold on an intrathecal trial after dorsal root ganglion inflammation findings.
Why do insurers restrict or deny coverage for some spinal muscular atrophy treatments?
Cost is the main driver, along with limited long-term evidence. Within a month of Zolgensma's approval, Bernstein analysts found that insurers' policies were stricter than the FDA label: one plan would not cover patients previously treated with Spinraza, and nine would not allow concurrent use. Current policies still reflect that caution. A Centene policy states that Zolgensma or Itvisma must not be prescribed concurrently with Spinraza or Evrysdi and excludes advanced SMA, such as complete paralysis or permanent ventilator dependence, and re-treatment.
A 2022 pulse check found payers covering almost all commercial lives require prior authorization for every SMA drug, and about half of lives covered did not pay for presymptomatic patients identified by newborn screening. Denials are often appealed, and outcomes vary by plan, so families usually work with a specialist center's prior-authorization staff.
Is Zolgensma more cost-effective than lifelong treatment with Spinraza or Evrysdi?
The evidence is incomplete. At launch, ICER judged the Zolgensma price consistent with its value benchmark, and Novartis argues that Itvisma costs 35% to 46% less than chronic therapy over 10 years. The comparison depends on assumptions about durability, which are unproven: one policy summary notes that the original FDA approval cited efficacy data in a total of 36 treated individuals. Chronic therapies add cumulative cost over decades but have also been tested for longer, and high-dose Spinraza and tablet Evrysdi are improving convenience.
How big is the SMA market, who are the key players and what is the outlook?
Three products shared about $4.9 billion in 2025 sales by my calculation: Spinraza at $1.55 billion, Evrysdi at 1.76 billion Swiss francs ($2.11 billion) and Zolgensma plus Itvisma at $1.23 billion. In the second quarter of 2026, Novartis reported $365 million for Zolgensma and Itvisma, up from $297 million, with growth driven by Itvisma. Biogen's U.S. Spinraza revenue rose 36.8% in the quarter, driven by demand and stocking for the high-dose regimen.
Muscle-targeted candidates such as Scholar Rock's apitegromab and Biohaven's taldefgrobep alfa are pipeline competitors. Regenxbio reported that licensed U.S. patents covering Zolgensma expired in January 2026, reducing its royalties. Biogen expects Spinraza revenue to be relatively flat in 2026. The outlook is shaped by newborn screening, switching between therapies and Itvisma's expansion into older patients, which Novartis expects to produce multibillion-dollar sales.