In late May 2019, the U.S. Food and Drug Administration (FDA) approved Zolgensma (onasemnogene abeparvovec), gene therapy for spinal muscular atrophy (SMA) with a bi-allelic mutation in the survival of motor neuron 1 (SMN1) gene in children under the age of two. The drug is suitable for both symptomatic patients and pre-symptomatic patients identified by genetic testing.
Regulatory approval for Zolgensma was granted to AveXis, acquired by Novartis for $8.7 billion in April 2018.
In mid-May 2020, the European Medicines Agency (EMA) granted Zolgensma marketing authorization to treat patients with spinal muscular atrophy (with genetic defect in the 5q chromosome arm) with a bi-allelic mutation in the SMN1 gene and either a clinical diagnosis of SMA type 1 or the presence of up to 3 copies of the SMN2 gene.
The key advantage of Zolgensma over any other pharmacological approach to the therapy of spinal muscular atrophy lies in the possibility of a complete cure of the disease — after a single injection of the drug. In any case, one can safely count on this, if we proceed from the fundamental principles of gene therapy.
However, it should be clearly understood that, first, the beneficial effect of Zolgensma will appear gradually and, second, Zolgensma does not reverse the impairments already caused by the disease. In other words, some signs and symptoms of spinal muscular atrophy (such as difficulty swallowing and breathing, and muscle weakness) may persist.
Since spinal muscular atrophy results in progressive and irreversible damage to motor neurons, the benefit of using Zolgensma depends on the severity of the disease at the time of treatment. And therefore, it obviously makes sense to provide treatment for suitable patients with Zolgensma from the first days of their life. Although patients with symptomatic spinal muscular atrophy in the later stages are unlikely to achieve the same general motor development as healthy peers without the disease, they will still benefit clinically from gene therapy.
Price of Zolgensma for Spinal Muscular Atrophy
Novartis has priced Zolgensma at $2.125 million per dose. One injection is enough, and repeated ones are not recommended to avoid immune reactions. Such an impressive price tag made Zolgensma the most expensive drug in the world.
By the way, Novartis previously stated that a reasonable cost of the drug falls within the range of $4.6–5.4 million.
One would think that the Swiss pharma giant has engaged in a policy of incredible greed but things have been carefully calculated. The Basel-based company proceeded from a pharmacoeconomic model of cost-effectiveness accumulated over ten years of treatment, the return on which is measured in added quality-adjusted life years for such a serious, disabling, and deadly disease as spinal muscular atrophy.
Spinraza as a Starting Point for Zolgensma Pricing
Novartis, when pricing Zolgensma, was guided by the fact that the price of the new drug should be half the cost of a 10-year course of therapy for spinal muscular atrophy with Spinraza (nusinersen). Until now, it was the only drug by Biogen and Ionis Pharmaceuticals approved for this disease. Considering that the first year of nusinersen treatment costs $750,000 and each subsequent year costs $325,000, Spinraza would cost $4.125 million over ten years of therapy.
What Do Experts Say About Cost of Zolgensma?
According to estimates by the US Institute for Clinical and Economic Review (ICER), a reasonable price for Zolgensma ranges from $310,000 to $900,000 or from $710,000 to $1.5 million depending on which evaluation methodology to use — per quality-adjusted life-year (QALY) or per life-year gained (LYG).
Following approval of Zolgensma and publication of its effectiveness in pre-symptomatic patients and patients with spinal muscular atrophy type 2, ICER experts have changed their estimate of the reasonable price of gene therapy raising the bar to $1.1–1.9 million (QALY) and $1.2–2.1 million (LYG). Thus, the expert opinion on the upper limit agreed with Novartis’ price proposal.
So How to Pay for Zolgensma?
Novartis has implemented a program that will, first, allow paying an astronomical amount for Zolgensma in installments over 5 years ($425,000 annually), and, second, stop paying for the treatment if it suddenly stops working effectively for up to 5 years.
A pay-over-time option to help ease possible short-term budget constraints, especially for states, small payers, and self-insured employers. A 5-year outcomes-based agreement is actually a marketing ploy since the healing effect of gene therapy is unlikely to cease over this period of time.
In the civilized countries of the capitalist world, where the national healthcare system really takes care of the well-being of its citizens and where insurance medicine is of the highest level, patients don’t shell out the entire cost of the drugs, only an insignificant amount within everyone’s reach. In other words, if you are insured, you have access to the entire armamentarium of the latest drugs, the effectiveness and safety of which is slowly but still growing every year. Of course, sometimes insurers refuse to include in their formularies those drugs that they think cost too much considering the benefits they bring.
In countries with developing economies like Russia, where enemies are seen all around and defense and national security spending exceeds public healthcare funding many times over, patients have to rely on the mercy of the government to both negotiate discounts with Novartis and allocate the budget for the purchase of Zolgensma. It is unlikely that a Swiss pharmaceutical corporation, being a purely commercial enterprise with investors behind its business, would go crazy to engage in charity.
How to Get Zolgensma for Free?
Novartis began distributing Zolgensma for free in January 2020. In order to receive Zolgensma for free, the physician overseeing the eligible patient must apply for it. Every two weeks, there will be a random selection process: whoever is lucky enough to receive the drug at no cost. The lottery is open to patients from any country in the world where Zolgensma has not yet been officially approved.
In 2021 the lottery with free access to Zolgensma gene therapy continues, and a maximum of one hundred patients can win a course of treatment.
Novartis Gives Away Zolgensma Free of Charge to Patients With Spinal Muscular Atrophy
At the beginning of 2020, Novartis launched an initiative to distribute Zolgensma free of charge, which a maximum of one hundred people will be able to receive at no cost. This limitation is due, according to the Swiss pharmaceutical giant, to production capacity. The drug is produced at one site, which must both meet the growing demand for the treatment in view of the imminent appearance of Zolgensma in other countries and ensure its uninterrupted supply for clinical trials. Novartis, which plans to launch two more manufacturing sites, promises to follow the philanthropic program over time gradually increasing the number of free doses of Zolgensma it distributes.
- Novartis has notified that as of mid-January 2021, 170 children have received a free infusion of Zolgensma. And that’s a decent proportion, considering that more than 700 patients have been treated with the drug so far. Despite the COVID-19 coronavirus pandemic, gene therapy has been successfully carried out in all areas of the world, including such countries as Belarus, Georgia, Chile, India, Mexico, Russia, Tunisia, Ukraine, United Arab Emirates, Vietnam, etc.
The Novartis initiative promptly sparked a furious outcry from patient communities that it was inappropriate and reprehensible to distribute a life-saving treatment, especially among children, through a lottery. Novartis allegedly acts inappropriately actually forcing patients to compete with each other. The rich corporation organizes “The Hunger Games” by pulling children’s names out of a hat, and every two weeks a lucky child from a poor country gets a sure shot at life. Real mockery!
Supposedly it would be more correct to make an unambiguous list in which patients or countries would be sorted according to the priority of need for treatment. For example, low priority should be given to those countries where Spinraza, the world’s first drug to treat spinal muscular atrophy, is officially approved because there is no reliable evidence yet that Zolgensma is more effective than Spinraza.
Novartis explained that it chose the lottery approach to giving away Zolgensma for free after discussions with bioethics experts. Any other options would have involved creating a complex pattern of criteria that a particular patient applying for gene therapy had to meet which would necessarily have led to unfair discrimination.
Indeed, there would be incredibly problematic questions, such as: Should Zolgensma be offered first to patients who have no other treatment options? Should Zolgensma be prescribed primarily based on age, since the youngest children would theoretically benefit the most? Is it wise to form a first-come, first-served waiting list?
The Novartis charity program, based on three principles — fairness, clinical need, and global accessibility — gives clinicians the ability to determine for themselves whether a given patient needs Zolgensma gene therapy. In doing so, everything is done to exclude any points that would favor children from more advanced healthcare systems for priority access to Zolgensma.
Clearly, the question of the feasibility of implementing a global program of free access to Zolgensma in the face of Novartis’ limited production capacity is not closed. And yet, even this kind of assistance to patients suffering from spinal muscular atrophy is the right solution.
Yes, the drug lottery is something decidedly new. And it reflects the sad picture with the world healthcare system and the ongoing struggle for free access to drugs.
In some countries, pharma companies are allowed to give away free unapproved drugs as part of compassionate use experimental therapy programs among patients with serious or life-threatening illnesses who no longer have any other treatment options. Pharma companies consider each such request on a case-by-case basis, according to the potential ratio of medical benefit to harm. Pharma companies also tend to direct patients to participate in clinical trials.
Keep in mind that any pharmaceutical company, no matter how humane it may claim to be, is just a for-profit business. The fact that Zolgensma costs a fabulous amount of money in the United States can be explained by the complete absence of legal regulations that would have any effect on the egregious pricing. But the fact that patient communities are hysterical cannot be explained by anything. Novartis’ charitable initiative is solely its whim, will, or fancy, and demanding something from a capital-starved Big Pharma player is akin to looking a gift horse in the mouth.
It is possible that Novartis’ charity program is driven by competition. First, Spinraza is making steady money by enrolling more and more patients. Second, Roche has already received regulatory approval for Evrysdi (risdiplam), a strong drug against spinal muscular atrophy that is made in an oral formulation. Novartis economists could reason: even if we lose part of our profits from Zolgensma sales, our competitors won’t be able to make money on patients we have already treated.
Anyway, it’s a sin to complain. Where everything rests on a voluntary basis, there is no room for orders.
In Which Countries Is Zolgensma Officially Approved?
Gene therapy Zolgensma is officially approved for use in the treatment of spinal muscular atrophy in the following countries:
- United States (May 24, 2019)
- Japan (March 19, 2020)
- European Union as well as the United Kingdom, Iceland, Liechtenstein, and Norway (May 19, 2020)
- Brazil (August 17, 2020)
- Canada (December 16, 2020)
Zolgensma should receive regulatory approval in Switzerland, Australia, Argentina, and South Korea in the first half of 2021.
How Does Zolgensma Work?
Spinal muscular atrophy is a group of rare (9.1 cases per 100,000) genetic diseases characterized by progressive degeneration of spinal cord and brainstem motor neurons and manifested by hypotonia, skeletal muscle atrophy, and general weakness. The pathology is caused by a loss of function or dysfunction (deletion, rearrangement, or mutation) of survival of motor neuron 1 (SMN1) gene. Almost all patients (95–98%) are homozygous for the defective SMN1 gene meaning that both parents are carriers of the recessive genetic disorder.
Onasemnogene abeparvovec (AVXS-101) is a gene therapy that provides a single intravenous dose to replace the missing or defective SMN1 gene with its functional copy. The result is normal production of survival motor neuron (SMN) protein and a corresponding cure for spinal muscular atrophy.
The gene therapy with onasemnogene abeparvovec implies adenoviral delivery to the body of the SMN transgene encoding a fully functional SMN protein and embedding in the motor neuron nuclei. Due to the ability of the onasemnogene abeparvovec to cross the blood-brain barrier with subsequent transduction of target cells, the expression of SMN in motoneurons in all parts of the brain and spinal cord has been confirmed. The use of cytomegalovirus enhancer and chicken beta-actin as a hybrid promoter determines the rapid and sustained expression of SMN.
Evaluation of the biodistribution of transgenic DNA and mRNA and SMN protein by Droplet Digital PCR (ddPCRTM), reverse transcription polymerase chain reaction (RT-PCR), and immunohistochemical staining revealed that Zolgensma vector genomes, RNA transcripts, and SMN proteins, respectively, were found in all spinal cord regions studied, including the cervical, thoracic, lumbar, and sacral regions. SMN protein expression in motor neurons of the spinal cord was established at a level similar to that of tissues not affected by type 1 spinal muscular atrophy. SMN protein expression was also detected in cortical and subcortical regions of the motor cortex and medulla.
Analysis of choline acetyltransferase (ChAT), a marker of motor neurons, demonstrated an abundance of the latter with normal size and shape.
In general, a single intravenous injection of Zolgensma was shown to be able to restore SMN expression in motor neurons deprived of the functional SMN1 gene. Onasemnogen abeparvovec was developed with the goal that spinal muscular atrophy is a monogenic disease, i.e. it is enough to deliver the correct copy of the problem gene to the body to stop the progression of the pathology.
Clinical Efficacy of Zolgensma
The efficacy of Zolgensma in pediatric patients less than two years of age has been studied in two clinical trials (nonrandomized, open-label, multicenter), the ongoing STR1VE-US (NCT03306277) phase 3 and the completed START (NCT02122952) phase 1/2a.
Clinical symptoms of spinal muscular atrophy type 1 participants manifested before the age of six months. Genetic testing of the subjects showed that all had bi-allelic deletions of the SMN1 gene, two copies of the SMN2 gene, and no modification of c.859G>C in exon 7 of the SMN2 gene (the presence of this indicates a mild disease phenotype). The initial antibody titer against AAV9 was ≤ 1:50. All subjects received a single intravenous infusion of onasemnogen abeparvovec.
The efficacy of gene therapy Zolgensma for spinal muscular atrophy, on the basis of which the regulator gave a positive verdict, was assessed by two measures. First, patient survival as the time from birth to either death or the need for permanent ventilation. The latter implies an invasive procedure (tracheostomy) or ventilation for ≥ 16 hours of respiratory assistance per day for ≥ 14 consecutive days in the absence of an acute reversible illness, excluding perioperative ventilation. Permanent ventilation is considered a surrogate for death. An acute reversible illness is defined as any condition other than spinal muscular atrophy that results in increased medical intervention.
Second, improvements in motor skills, according to the Children’s Hospital of Philadelphia’s Infant Test of Neuromuscular Disorders (CHOP INTEND) scale, whose overall score ranges from 0 to 64 (more is better) and which assesses the ability to control head movements, roll from back to sides, sit, crawl, stand, etc.
The STR1VE-US (NCT03306277) clinical trial included 22 patients who did not require non-invasive ventilation and who could be fed orally (parenteral nutrition was not required). The baseline CHOP INTEND score was an average of 31.0 (18–47). The mean age of the patients was 3.9 months (0.5–5.9). Participants received a single dose of Zolgensma at 1.1×1014 vector genomes per kilogram (vg/kg).
As of early March 2019, clinical results were as follows.
91% of patients (n=20/22) remained alive but one died (at 7.8 months of age) due to respiratory failure unrelated to gene therapy Zolgensma and another left the experiment (at 11.9 months of age) and required permanent ventilation at the time of study abandonment.
70% of respondents (n=14/20) reached 14 months of age without the need for permanent ventilation.
- For reference, according to the natural history of spinal muscular atrophy type 1, only a quarter of patients cross the 14-month survival threshold without the need for permanent ventilation.
Gene therapy Zolgensma testified to a significant improvement in motor function from the first month after administration of the drug.
CHOP INTEND scores increased by an average of 6.9, 11.7, and 14.3 points at one, three, and five months after Zolgensma administration, respectively.
95% of patients (n=21/22) reached a clinically meaningful score of 40 or higher on the CHOP INTEND scale, 50% (n=11/22) had a minimum score of 50, and 9% (n=2/22) achieved 60 or higher.
- For reference, according to the natural history of spinal muscular atrophy type 1, children 6 months of age or older never have or maintain a CHOP INTEND score of 40 points or higher. Moreover, between 6 and 12 months of age, a decrease of an average of 10.7 points is recorded.
Improvements in motor skills at the mean age of 11.9 months (9.2–16.9) were as follows:
- independent head control for at least 3 seconds: 80% of patients (n=16/20) [two participants were able to do this before gene therapy so were excluded from analysis]
- ability to roll from back to sides: 41% (n=9/22)
- ability to sit without assistance for at least 30 seconds: 50% (n=11/22)
- ability to stand with assistance: 5% (n=1/22)
- ability to crawl: 5% (n=1/22)
- ability to stand up independently and then stand without outside support: 5% (n=1/22).
Further observations indicated that at the end of May 2019, 83% of patients (n=5/6), who were 18 months old, were able to sit independently for at least 30 seconds, and one patient had learned to stand and walk with support.
The START (NCT02122952) clinical trial involved 15 patients who received either a low dose of the gene therapy drug (n=3) or a high dose (n=12). The low dose was approximately one-third of the high dose, which, in turn, could not be accurately assessed due to, as subsequently discovered, the incorrect methodology used to measure Zolgensma concentration as well as its decline over storage time. Presumably, the low dose was in the range of 4.3–4.6×1013 vg/kg and the high dose was in the range of 1.1–1.4×1014 vg/kg. In the low-dose cohort, the mean age of participants was 6.3 months (5.9–7.2) and in the high-dose cohort it was 3.4 months (0.9–7.9).
After 24 months of follow-up, the data for the high-dose cohort are as follows.
All patients survived without the need for permanent ventilation.
For reference, according to the natural history of spinal muscular atrophy type 1, only 8% of patients survive to the age of two years without the need for permanent ventilation.
Gene therapy Zolgensma demonstrated a significant improvement in motor skills starting from the first month after administration of the drug. An important observation: the dynamics of improvement correlated with age — the earlier treatment was carried out, the more confident and faster-restored motor functions, despite the initially low CHOP INTEND score.
The baseline mean CHOP INTEND score of 28.2 points increased to 49.6, 53.6, and 55.5 points at 6, 12, and 24 months post-treatment, respectively. 92% of patients (n=11/12) reached a clinically meaningful CHOP INTEND score ≥ 50.
Improvements in motor skills are as follows:
- independent head control for at least 3 seconds: 92% of patients (n=11/12)
- ability to roll from back to sides: 75% (n=9/12)
- ability to sit without assistance for at least 5 seconds: 92% (n=11/12)
- ability to sit without assistance for at least 10 seconds: 83% (n=10/12)
- ability to sit without assistance for at least 30 seconds: 75% (n=9/12)
- ability to stand with assistance: 17% (n=2/12)
- ability to stand independently: 17% (n=2/12)
- ability to walk with assistance: 17% (n=2/12)
- ability to walk independently: 17% (n=2/12).
60% of patients (n=6/10), who did not initially require regular maintenance bilevel positive airway pressure (BiPAP) ventilation, maintained this status.
86% of patients (n=6/7), who did not initially receive parenteral nutrition, maintained this status. Only one patient had to be temporarily tube fed due to wound healing during the difficult recovery period after scoliosis surgery. 80% of patients (n=4/5) on parenteral nutrition were able to switch to standard oral feeding.
Administration of Zolgensma provided sustained support for bulbar function: 92% of patients (n=11/12) were able to speak, swallow, and eat orally.
All participants experienced common childhood respiratory illnesses which, however, in the case of spinal muscular atrophy type 1 usually ended in tracheostomy or death. Meanwhile, all patients successfully underwent appropriate hospitalization without tracheostomy or the need for permanent ventilation.
At the beginning of March 2019, after nearly four years of follow-up of agreeing participants (n=10) in the high-dose cohort, all (n=10/12) remained alive and did not require continuous ventilation or nutritional support. The mean time since Zolgensma administration was 3.7 years (3.3–4.3) and the mean age of patients was 3.9 years (3.4–4.8). All patients retained their previously acquired motor skills. Two of the four patients, who had previously required maintenance BiPAP, refused it.
At the end of May 2019, with the mean time since gene therapy Zolgensma of 3.9 years (3.5–4.6) and the mean patient age of 4.2 years (3.7–5.0), all patients in the observed high-dose cohort (n=10/10) remained alive continuing to demonstrate improved physical skills. Two patients, who had never previously received nusinersen, were able to stand with assistance. 70% of patients (n=7/10) were no longer prescribed additional nusinersen treatment. 60% of patients (n=6/10) do not require maintenance BiPAP.
At the end of December 2019, after the mean of 4.5 years (4.1–5.2) following Zolgensma administration, when the mean age of patients in the high-dose cohort was 4.8 years (4.3–5.6), all participants (n=10/10), who agreed to participate in the long-term clinical follow-up (LT-001), remained alive and did not require permanent ventilation.
The motor skills acquired at the completion of START were preserved in all patients, and 20% (n=2/10) of them acquired the ability to stand with support (none of them received additional Spinraza).
Among other achievements: 60% (n=6/10) did not receive other disease-modifying therapy; the same number of participants did not turn to regular daily oxygen support.
The STR1VE-EU (NCT03461289) phase 3 clinical trial (nonrandomized, open-label, multicenter) invited patients (n=33) with spinal muscular atrophy type 1 whose genome was characterized by a bi-allelic deletion or point mutation of the SMN1 gene and the presence of one to two copies of the SMN2 gene. The age of the participants should not have exceeded 6 months by the time of initiation of gene therapy Zolgensma.
A feature of the STR1VE-EU enrollment was that the disease severity of some participants was significantly more severe when compared with patients in the START (NCT02122952) and STR1VE-US (NCT03306277) clinical trials. Thus, the baseline CHOP INTEND score was an average of 27.6 (14–55), over a quarter of subjects (27%, n=9/33) were on ventilation, and nearly one-third (30%, n=10/33) required parenteral nutrition.
The mean age of subjects at the time of Zolgensma administration was 4.1 months (1.8–6.0), the mean age of symptom manifestation was 1.6 months, and the number of SMN2 gene copies was 2.
As of late December 2019, when the mean of 10.6 months (1.8-15.4) had elapsed after Zolgensma administration and the mean age of patients was 14.6 months (6.9-18.6), nearly all (97%, n=31/32) remained in event-free status, including 94% (n=30/32) who reached 10.5 months of age and 56% (n=18/32) who reached 13.6 months. An event was defined here as the need for either tracheostomy or respiratory support (via noninvasive ventilation) for 16 or more hours daily for 14 or more consecutive days in the absence of acute reversible complications, excluding perioperative ventilation.
- For reference, in the natural history of spinal muscular atrophy type 1 left untreated, event-free survival by age 10.5 and 13.6 months is fair for only 50% and 25% of infants.
The vast majority (92%) of patients, who did not initially require oxygen support, either maintained this status or received BiPAP ventilation for prophylactic purposes during acute complications. Two-thirds (67%) received oral food without needing nutritional support, indicating stabilization (or halting) of the progression of spinal muscular atrophy.
Two-thirds (66%) of the subjects demonstrated motor skills that are not fixed if the disease is left untreated. Thus, 25% (n=8/32) were able to sit independently for at least 10 seconds (primary endpoint), 63% (n=20/32) gained the ability to control head, 25% (n=8/32) were able to roll from back to sides, and one patient was able to stand, crawl, and walk with assistance.
The mean increase in CHOP INTEND score was: 5.9 points (1 month after Zolgensma administration, n=31), 10.1 points (3 months, n=29), 13.3 points (6 months, n=27).
In 66% of patients (n=21/32) the increase in CHOP INTEND score came out to 40 or more points, in 38% (n=12/32) it came out to 50 or more points.
- For reference, in untreated spinal muscular atrophy type 1, patients almost never reach a CHOP INTEND score of 40 or more points.
Regarding the safety profile of Zolgensma, one participant (aged 6.9 months) had to withdraw from the clinical trial due to a serious adverse event of hypoxic-ischemic brain injury and acute respiratory distress syndrome (ARDS), which resulted in death. The autopsy showed that gene therapy had nothing to do with it.
Literally, all subjects (97%, n=32/33) experienced at least one adverse event, with gene therapy causing serious adverse reactions in six participants. Almost all patients (88%, n=29/33) experienced an increase in hepatic transaminases which was stopped by the corticosteroid prednisolone. In four patients, the platelet count dropped below 75,000/μL.
Zolgensma Safety Issues
The prescribing information for Zolgensma has a black box warning about the risks of acute and severe liver damage and elevated aminotransferases levels, with this possibility increasing with existing hepatic impairment.
In addition, there is no unequivocal data on how the patient’s body will behave in the case of repeated infusions of Zolgensma. It is believed that the high titer of antibodies against the viral vector AAV9 as the base of the drug, noted after the infusion, will be a serious obstacle to another dose of gene therapy.
Administration of Zolgensma to patients with advanced forms of spinal muscular atrophy (e.g., complete limb paralysis, constant dependence on ventilation) has not been verified.
Among the most common adverse events to gene therapy Zolgensma were an increase in aminotransferases ALT and/or AST (27.3% of patients) and vomiting (6.8%).
Novartis Is Not Afraid of Falsification With Zolgensma
In August 2019, the FDA released a statement regarding an investigation into the manipulation of data included in Zolgensma’s New Drug Application (NDA).
In late June 2019, two months after Zolgensma was approved, AveXis self-reported data manipulation, a problem affecting the accuracy of certain animal drug test results, to the regulator. At issue is data collected during the quantification of the active ingredient in vivo experiments on mice, which is critical to characterizing the comparability of versions of the drug studied in phase 1 and 3 clinical trials. There is a view that misrepresentation may affect the interpretation of the results of phase 1 clinical trials and some preclinical studies included in the NDA. However, as stated, there is no reason to believe that other information was tampered with in passing and there is no evidence that the clinical data of patients could have been deliberately altered.
If the regulator had been aware of the falsification in advance (because AveXis had been known about it since at least mid-March), Zolgensma would have received approval later. But, as the FDA pointed out, there is no reason to recall the drug from sale. Because of the impressive amount of data, concerns about its veracity are limited to a small portion used to fine-tune the manufacturing process. Otherwise, the efficacy and safety of Zolgensma is unquestionable.
AveXis voluntarily informed the regulator about the falsified data in Zolgensma’s NDA only because it obviously understood the situation. Firstly, if the regulator had uncovered the falsified data on its own, there would have been much more problems. Secondly, the fabricated data are not critical and do not affect the market status of the drug; Novartis stated that the data was used only during the initial testing of the drug and is not used during the current commercial release of Zolgensma.
By the way, after the FDA learned about the data manipulation, the AveXis production site was inspected which revealed certain irregularities but not significant.
Kaspar and Thomas Dee, founding CFO and CSO at AveXis, respectively, accused of tampering with Zolgensma data, were expelled in disgrace. In March 2021, the disgraced brothers were taking board roles at Endsulin, a small biotech searching for a better treatment for type 1 diabetes.
Zolgensma and Competitors
The appearance of the first gene therapy for spinal muscular atrophy will have a negative impact on Biogen’s revenue from the sale of Spinraza and the royalties that Ionis, the originator of this antisense oligonucleotide, receives. The current earnings here are decent: while nusinersen generated $884 million in 2017, it earned $1.72 billion in 2018, $2,10 billion 2019, and $2,05 billion in 2020.
However, Biogen investors should not be afraid yet, since Zolgensma is currently approved only for the treatment of the infant form of spinal muscular atrophy; Spinraza is suitable for the therapy of all patients including adults. Thus, according to Novartis estimates, there are approximately 400 new diagnoses of spinal muscular atrophy each year in the United States versus somewhere around 7,000 patients of all ages and types of the disease.
A major competitive blow came when the approval of Evrysdi (risdiplam), an alternative pre-mRNA splicing modifier for the SMA2 gene, brought to readiness by Roche and PTC Therapeutics, happened. First, Evrysdi is designed to treat all types of spinal muscular atrophy. Second, Evrysdi is superior to Spinraza and only slightly inferior to Zolgensma. Third, risdiplam is an oral low-molecular-weight drug made in the form of a solution, meaning it is not difficult to manufacture and easy to administer, as compared to Spinraza which is administered intrathecally requiring specially trained medical personnel. Fourth, Evrysdi is attractive in terms of price with a maximum cost of $340,000 per year.
EvaluatePharma predicts that demand for Zolgensma will cross the $2 billion mark by 2022 thereby catching up with interest in Spinraza. Sales of the latter will gradually decline giving way to Evrysdi. By 2026 Zolgensma and Evrysdi sales will almost equal each other.
In general, Novartis will have to try hard to at least recoup the $8.7 billion that was laid out for the AveXis acquisition. Zolgensma’s sales were $361 million in 2019 and $920 million in 2020.
Zolgensma: What’s Next
Novartis is trying to broaden the list of therapeutic indications for Zolgensma by adding pre-symptomatic patients as well as those diagnosed with milder forms of spinal muscular atrophy, types 1 and 3.
As of July 2018, spinal muscular atrophy has been added to the list of conditions recommended for screening for them in all newborns in the United States. In other words, Novartis can offer Zolgensma to every patient newly diagnosed. And this is very important because of the extremely narrow window of opportunity that provides a high chance of “resuscitating” a full set of motor neurons that quickly become dysfunctional. Their death means the point of no return.
When it comes to the treatment of spinal muscular atrophy in patients who have been living with it for quite a long time, it is still decisively unclear whether Zolgensma will help them. Perhaps the critical threshold of motor neuron loss has been passed and even gene therapy will not save them. Thus, there remains a need for basic research that will give rise to alternative treatment options for fully symptomatic patients with spinal muscular atrophy.
The ongoing SPR1NT (NCT03505099) phase 3 clinical trial (nonrandomized, open-label, multicenter, international) enrolled pre-symptomatic (no symptoms or signs of disease) patients under 6 weeks of age with a bi-allelic deletion of the SMN1 gene and two copies of the SMN2 gene (cohort 1) or three copies of the SMN2 gene (cohort 2) — those at risk for diagnosis of spinal muscular atrophy type 1 or 2, respectively. All participants received a single intravenous dose of Zolgensma.
The primary endpoint was the ability to sit independently for at least 30 seconds (cohort 1) or to stand without support for at least 3 seconds (cohort 2) fixed until 18 or 24 months after Zolgensma administration, respectively.
Secondary endpoints for cohort 1: event-free (mortality or need for permanent ventilation in the absence of acute reversible complications or perioperative changes) survival (by age 14 months) and ability to maintain body weight at or above the 3rd percentile (according to WHO norms), and being without non-oral/mechanical feeding support (at 18 months or younger).
Secondary endpoint for cohort 2: ability to walk independently (to take at least 5 steps independently displaying coordination and balance) [by age 24 months].
Novartis relies on the specific pathogenesis of spinal muscular atrophy. SMN protein deficiency, as the root cause of the disease, is not only reflected by a decrease in motor function but also leads to the selective death of motor neurons. And once they have gone, no drug will bring them back to life. In other words, therapy should be initiated at a very early age. Additional evidence for the validity of this approach comes from the results collected by Spinraza in the treatment of pre-symptomatic patients.
However, if one tries to administer Zolgensma too early, there is a definite possibility of encountering a biological barrier in the form of maternal IgG antibody against AAV9 transmitted to the child through the placenta. Since AAV9 is a wild non-synthetic virus, most adults, including the mothers of those children with spinal muscular atrophy, already have immunity against it in the form of appropriate antibodies that inactivate Zolgensma. Fortunately, maternal anti-AAV9 antibodies are only temporarily present in the fetus and are characterized by an approximately three-week half-life. Still, in about 5–10% of infants under 6 months of age, the antibody titer against AAV9 exceeds the threshold level of 1:50 thus making treatment, including early treatment with Zolgensma, impossible.
Based on data at the end of September 2018, Zolgensma infusion was given to 7 patients aged 8–37 days (median 12 days, mean 21 days) and with a baseline CHOP INTEND score of the mean of 41.7.
The increase in CHOP INTEND score was the mean of 6.8 (n=4), 11.0 (n=3), 18.0 (n=3), and 22.5 (n=2) points after 14 days, one, two, and three months, respectively.
At the beginning of March 2019, with the median duration of follow-up of 5.4 months and the median age of subjects of 6.1 months, all patients (n=18/18) remained alive and in an event-free status.
Among patients with spinal muscular atrophy type 1, the median CHOP INTEND score increase was 8.9 points after one month, and the median Bayley-III (The Bayley Scales of Infant and Toddler Development, third edition) score increase for assessment of gross motor skills was 8.4 points after two months.
All patients showed CHOP INTEND gain or retention of 50 points, with four having a score of 60 and three reaching a maximum of 64 on this scale.
Patients demonstrated age-appropriate motor skills, including two who could sit independently for at least 30 seconds and one who could stand with support for at least 2 seconds.
As of the end of May 2019, 10 patients with two copies of the SMN2 gene, 12 with three, and one with four had received gene therapy Zolgensma. All remained alive and did not require permanent ventilation.
After six months of follow-up of the cohorts with two and three copies of SMN2, all patients had normal swallowing functions and oral food intake.
In the cohort with two copies of the SMN2 gene, administration of Zolgensma resulted in a CHOP INTEND score of 50 or higher, with seven patients scoring at least 60, including five participants who recorded a maximum score of 64. At the end result, 60% of patients (n=6/10) gained the ability to sit independently for at least 30 seconds, at the mean age of 7.6 months, and 30% of patients (n=3/10) learned to stand with support, at the mean age of 10.1 months.
At the end of December 2019, 14 and 15 pre-symptomatic patients with two and three copies of the SMN2 gene, respectively, had received Zolgensma. After the mean of 10.5 months (5.1–18.0) and 8.7 months (2.0–13.9) after a single dose of gene therapy, they had the mean age of 11.2 months (6.0–18.6) and 9.7 months (3.3–15.1). Another individual had four copies of the SMN2 gene but his data were not included in the treatment efficacy analysis.
At the indicated date, all patients remained alive and did not require permanent ventilation. All participants did not require additional nutritional support. Nearly all (97%, n=28/29) had body weight within the normal range.
Half of the patients (50%, n=7/14) with two copies of the SMN2 gene and all subjects (100%, n=15/15) with three copies of the SMN2 gene showed the development of gross motor function at a level similar to same-age peers without spinal muscular atrophy.
All patients (100%, n=14/14) with two copies of SMN2 gene increased their CHOP INTEND score to 50 or more, while 86% (n=12/14) showed a score of 60 or more.
Motor gains in patients with two copies of the SMN2 gene:
- 57% (n=8/14) were able to sit independently for at least 30 seconds, and 88% (n=7/8) of them achieved this in an age-appropriate time frame
- 21% (n=3/14) were able to stand independently for at least 10 seconds, and two of them have achieved this at an age-appropriate time frame
- 29% (n=4/14) were able to walk independently, and three of them did so at an age-appropriate time frame.
Motor gains in patients with three copies of the SMN2 gene:
- 27% (n=4/15) were able to stand independently for at least 3 seconds, and all were able to do so at age-appropriate time frame
- 13% (n=2/15) started walking independently
- 60% (n=9/15) were able to stand with assistance.
At the early June 2020, follow-up of patients from two cohorts (n=14 and n=15) with an mean age of 15.6 months (8.8-18.8) and 15.2 months (3.3-21.1) continued. Following a mean of 14.1 months (8.0-18.4) and 13.1 months (2.0-19.9), respectively, after a single injection of Zolgensma, all patients remained alive and did not require respiratory or nutritional support, and their clinical parameters improved.
All patients (100%, n=14/14 and n=15/15) demonstrated a sustained gain in the mean score on the Bailey-III scale for assessment of fine and large motor skills.
Moreover, 64% (n=9/14) of subjects with two copies of the SMN2 gene and 100% (n=15/15) of those with three copies of the SMN2 gene showed improvements in large motor skills similar to those seen in children of the same age without spinal muscular atrophy.
Motor gains in patients with two copies of the SMN2 gene:
- 79% (n=11/14) were able to sit independently for at least 30 seconds, with 91% (n=10/11) of them achieving this within their age-appropriate time frame.
- 29% (n=4/14) were able to stand independently for at least 10 seconds, with 50% (n=2/4) of them achieving this skill in an age-appropriate time frame.
- 36% (n=5/14) learned to walk independently, and 80% (n=4/5) of them did so within an age-appropriate time frame.
- All (100%, n=14/14) achieved or maintained a CHOP INTEND score of 50 points or more, while 86% (n=12/14) had a score of 60 or more.
Motor gains of patients with three copies of the SMN2 gene:
- 53% (n=8/15) gained the ability to stand independently for at least 3 seconds, and all (100%, n=8/8) of them developed this ability within their age-appropriate time frame.
- 40% (n=6/15) began to walk independently, and all (100%, n=6/6) of them achieved this motor skill within an age-appropriate time frame.
- 80% (n=12/15) were able to stand with outside support, and 83% (n=10/12) of them did so within their age-appropriate time frame.
The STRONG (NCT03381729) phase 1/2 clinical trial (nonrandomized, open-label, multicenter), investigating the efficacy and safety of intrathecal administration of Zolgensma in spinal muscular atrophy type 2, is ongoing.
Novartis had hoped that the data collected in STRONG would be sufficient to submit an application to expand the patient population suitable for treatment with Zolgensma. However, in late September 2020, the FDA demanded a confirmatory clinical trial. As a result, the regulatory verdict will be delayed, it is believed, until as late as 2023.
- Intravenous administration of Zolgensma provided for spinal muscular atrophy type 1, as the most severe form of this neuromuscular disease, involves systemic delivery of gene therapy to motor neurons in the brain and spinal cord and other locations. Proper expression of SMN protein throughout the body, including peripheral tissues, is critical in infancy.
- Intrathecal administration of Zolgensma means injecting the drug into the spinal canal or subarachnoid space so that it reaches the cerebrospinal fluid. And it is reasonable for spinal muscular atrophy type 2 or 3 which is usually encountered in older children, adolescents, and adults. Intrathecal administration provides a more specific targeting of the drug when the main target is motor neurons throughout the spinal cord (including its cervical region) as well as in the brainstem and motor cortex. This is of particular interest for patients because these areas control the muscles involved in voluntary movements, breathing, and swallowing. In addition, it becomes possible to administer the drug in a reduced dose lowering the risks of an immune response to the viral vector in the peripheral tissues.
In October 2019, the FDA placed a partial hold on the clinical program testing intrathecal administration of Zolgensma. The regulator raised concerns about the results of preclinical animal studies: intrathecal administration of Zolgensma led to inflammation of the dorsal root ganglia of the spinal cord (due to infiltration of mononuclear cells) and was sometimes accompanied by degeneration or loss of neuronal cells. In response, Novartis provided safety data for Zolgensma showing no clinical cases of sensory neuropathy in 335 patients who received onasemnogen abeparvovec intravenously or intrathecally before the end of 2019. As of mid-March 2021, the STRONG clinical trial remained on pause.
Participants in the STRONG clinical trial, who were 6 months to 5 years old at the start of treatment, had to be able to sit for at least 10 seconds without support, but not to stand or walk. Among the mandatory inclusion criteria for the study:
- three copies of the SMN2 gene
- homozygous absence of exon 7 of the SMN1 gene (found in 95% of patients diagnosed with spinal muscular atrophy: corresponding aberrant mRNA transcripts, lacking exon 7, encode a truncated and unstable SMNΔ7 protein)
- absence of mutation c.859G>C in exon 7 of the SMN2 gene (its presence is indicative of a mild disease phenotype)
- manifestation of clinical signs and symptoms of spinal muscular atrophy before 12 months of age.
The patients were allocated into two groups according to the age of treatment start: from 6 months to 2 years (group 1) and from 2 years inclusively to 5 years (group 2). They received three different doses of Zolgensma: dose A (6.0×1013 vg/kg), dose B (1.2×1014 vg/kg), and dose C (2.4×1014 vg/kg).
Treatment efficacy measures were to be taken after 12 months of Zolgensma administration. The primary endpoint for the first group was the ability to stand independently for at least 3 seconds on the Bayley-III scale, and for the second group, change in score on the Hammersmith Functional Motor Scale Expanded (HFMSE). The secondary endpoint for both groups was the ability to walk at least 5 steps independently with proper coordination and balance on the Bayley-III.
- For reference, if spinal muscular atrophy type 2 is left untreated, HFMSE inevitably worsens over time, patients can never walk independently, and they are often confined to a wheelchair; over 30% of patients die by the age of 25.
At the beginning of March 2019, all patients (n=30) with the mean follow-up of 6.5 months remained alive. 63% of participants (n=19/30) showed an increase in their HFMSE score of the mean of 4.2 points. Two were able to stand independently, one was able to walk with support, and another was able to walk independently.
At the end of May 2019, when one of the above doses of Zolgensma was administered to 31 patients, the results are as follows.
Among the 16 patients who received dose A or dose B, 18 motor gains were recorded, including the fact that two people gained the ability to stand independently and one of them was able to walk independently.
At follow-up for the mean of 9.3 months, an increase in HFMSE score of the mean of 5.9 points was demonstrated, and this improvement nearly doubled the threshold of clinical significance set by an increase in HFMSE score of at least 3.0 points.
Half of the patients (n=6/12) showed an HFMSE increase of at least 3.0 points, attested within the first month after the Zolgensma infusion.
Among the 12 subjects who received dose B, there were four motor gains, including one participant’s ability to walk with assistance.
The results of gene therapy Zolgensma collected at the beginning of December 2019, covering data from 32 patients, were as follows (information on participants who received dose B [n=25] is given).
Group 1 (n=13)
One patient reached the primary and secondary efficacy endpoints.
83% of patients (n=5/6) of those who reached an age that was acceptable for the HFMSE assessment showed an increase of 3.0 points recorded at any time point after 24 months of age, and this was consistent with motor gains in group 2 which encompassed older patients.
46% of patients (n=6/13) demonstrated 18 motor gains including one patient who was able to stand and walk independently.
Group 2 (n=12)
Achieved the primary efficacy endpoint of treatment relative to the natural history of disease left untreated (p<0.0021).
The patients showed clinically significant improvement in motor function as reflected by a 6.0 point increase in HFMSE scale, 12 months after Zolgensma administration.
At any one follow-up period, nearly all subjects (92%, n=11/12) demonstrated a clinically meaningful HFMSE score increase of at least 3.0 points confirming a sustained response to treatment and a clear improvement in disease course over those left untreated (p<0.0001).
Three of those receiving dose B showed four motor gains including the ability to walk with support in one subject.
The ability to walk at least 5 steps on their own has not yet been observed in any of the participants.
Once the STRONG clinical trial has accumulated the necessary data, the REACH phase 1 clinical trial will be initiated, which will enroll patients aged 6 months to 18 years with spinal muscular atrophy type 1, 2, or 3 who do not meet the criteria for participation in all other Zolgensma trials. These are those patients who have antibody titers against AAV9 above 1:50.
Intrathecal administration of Zolgensma will be preceded by a plasmapheresis procedure which will eliminate these antibodies. In any case, trials in primates have confirmed the validity of plasmapheresis in the task of getting rid of the anti-AV9 antibodies that are harmful in this case.
Does Zolgensma Cure Spinal Muscular Atrophy Once and for All?
Despite the impressive successes of Zolgensma, the question remains as to the duration of the healing period of spinal muscular atrophy. The conducted clinical trials are still short enough to be able to unequivocally declare a complete cure that will last throughout the patient’s whole life. It is possible that SMN expression will decrease over time, with no possibility of repeating the gene therapy procedure because of the high probability of antibodies against the capsid proteins of the adenoviral vector AAV9. Furthermore, the spinal muscular atrophy phenotype may scale up as the child grows older affecting other organs and tissues and it is currently unknown whether gene therapy has a positive effect on other cell types.
Zolgensma (onasemnogene abeparvovec). Prescribing information. US. [PDF]
Zolgensma (onasemnogene abeparvovec). Prescribing information. Europe. [PDF]
Zolgensma (onasemnogene abeparvovec). European public assessment report (EPAR). [PDF]