Spinal muscular atrophy treatments, newborn screening, and the creation of a neurogenetics urgency

Abstract

Spinal muscular atrophy (SMA) is a progressive neuromuscular disorder characterized by loss of motor neurons leading to muscle weakness and atrophy. The United States’ Food and Drug Administration’s (FDA) approval of nusinersen, onasemnogene abeparvovec, and risdiplam for SMA has challenged existing treatment paradigms with multiple treatment options, a new natural history of the disease, and an emerging understanding of the importance of early and pre-symptomatic treatment. The profound impact of early, pre-symptomatic treatment has led to the creation of a neurogenetics urgency for newly identified patients with SMA, a novel problem for neurologists more accustomed to a more methodical approach to diagnosis and care. Implementation of newborn screening programs has helped facilitate early diagnosis and treatment, but challenges remain in overcoming administrative and procedural hurdles that can lead to treatment delays. Herein I discuss two cases that highlight the importance of early treatment, as well as gaps in our understanding of the progression of SMA in pre-symptomatic infants.

INTRODUCTION

Spinal muscular atrophy (SMA) is a progressive disorder characterized by loss of motor neurons and subsequent muscle weakness and atrophy.¹ In addition to the physical disability associated with limb weakness, children with SMA develop weakness of the diaphragm and accessory muscles of respiration leading to respiratory insufficiency. Progressive bulbar weakness leads to dysphagia and a risk of aspiration. Clinical subtypes of SMA are defined by maximal motor function, but the emergence of treatments in the last few years is challenging these clinical distinctions. Traditionally, infants with type 1 SMA have onset in the first weeks or months of life and never achieve independent sitting. SMA type 2 has onset in the first year, and affected individuals attain independent sitting, but do not walk independently. Children with type 3 SMA achieve independent ambulation, but have variable onset and progression. Some lose the ability to ambulate in early childhood, and others maintaining the ability to ambulate into adulthood. Regardless of the type of SMA, children undergo progressive loss of motor neurons. This loss is early and catastrophic in infants with type 1 SMA, slower in children with type 2 SMA, and more variable in children with type 3 SMA.²

SMA is caused by the absence of the survival motor neuron (SMN) protein encoded by the SMN1 gene on chromosome 5q.³ In 95% of cases, SMA is caused by a homozygous deletion of exon 7 of the SMN1 gene. With a population frequency of this mutation of 1/50 individuals, the resulting prevalence of SMA is about 1/10,000.⁴ The remaining 5% of cases are compound heterozygous for the common exon 7 deletion and an SMN1 point mutation.⁵ Type 1 SMA is the most common form of SMA and accounts for almost half of the cases.⁵ In most cases, infants with type 1 SMA do not survive beyond their second birthday, making SMA one of the most common genetic causes of infant mortality.⁶

The severity of SMA is determined in part by a pseudogene, SMN2, immediately distal of SMN1 on chromosome 5q.¹ SMN2 is a nearly identical copy of SMN1 except for a C to T transition in exon 7 that does not change the amino acid sequence, but disrupts a splicing enhancer.⁵ The resulting transcript excludes exon 7 and results in a non-functional SMN protein that is degraded. Despite the mutation on SMN2, a small amount normal transcript is produced and results in a normal SMN protein.⁷ The severity of SMA is largely controlled by SMN2 copy number with type 1 patients with SMA most often having 2 copies of SMN2, type 2 patients with 3 copies, and type 3 patients with 4 or more copies.¹

APPROVED TREAMENTS FOR SMA

The discovery of the genes causing SMA, and the role of SMN2 in severity, lead investigators to study targets that increase normal transcription from the SMN2 gene. Nusinersen, an antisense oligonucleotide SMN2 modulator, was approved by The United States’ Food and Drug Administration (FDA) in December 2016 after randomized blinded clinical trials showed efficacy in both early onset SMA (ENDEAR: A Study to Assess the Efficacy and Safety of Nusinersen (ISIS 396443) in Infants With Spinal Muscular Atrophy), and later onset SMA (CHERISH: A Study to Assess the Efficacy and Safety of Nusinersen (ISIS 396443) in Participants With Later-onset Spinal Muscular Atrophy).^{8, 9} Nusinersen is approved for all patients with SMA regardless of age and is delivered intrathecally with 4 loading doses in 2 months and maintenance doses every 4 months. The approval of nusinersen marked a significant breakthrough. It offered patients with SMA a disease modifying therapy for the first time, and challenged providers to adapt standards of care to a new natural history of the condition.^{10, 11} While initial studies in nusinersen focused on type 1 and relatively young type 2 patients with SMA, evidence of benefit is emerging across all ages and classes of patients with SMA, including older type 2 and 3 patients and those how have already reached milestones of disease progression such as ventilator dependence.^12-21

Soon after the approval of nusinersen came another shift in treatment paradigms with FDA’s approval of onasemnogene abeparvovec in May 2019 for patients with SMA under 2 years of age. Onasemnogene abeparvovec is traditional gene transfer therapy designed to deliver an SMN1 construct to motor neurons using an adeno-associated virus (AAV9) vector.²² Treatment with onasemnogene abeparvovec involves a single IV dose of 1.1 × 10¹⁴ vector genomes per kilogram delivered over an hour-long infusion.²³ FDA’s approval was based largely on a single center, open label dose finding trial in 15 infants with type 1 SMA (2 copies of SMN2).²² None of the treated infants reached permanent ventilation status during the study, and the overall cohort had longer survival and better attainment of motor milestones as compared to historical, similarly affected cohorts, especially in the 12 patients in the higher dose cohort. In a 2-year follow-up study involving the 12 patients in the higher dose cohort, patients required lower nutrition and pulmonary support, had improved motor function, and had lower hospitalization rates than historical controls.²⁴ Delivery of this traditional gene therapy is complex and requires expertise to address serious potential complications such as liver inflammation.²³

Risdiplam, an oral SMN2 modulator, was approved by the FDA for patients with SMA over 2 months of age in August 2020. In an open label, phase 2-3 study, 21 infants with type 1 SMA were treated with risdiplam for 12 months and showed increased expression of the SMN protein.²⁵ Additional data regarding efficacy of risdiplam are not yet published. Data from two clinical trials included on the FDA label suggest significant benefits in treated patients with improvement in motor function and survival without permanent ventilation in infants with type 1 SMA and improvement in motor function in patients with later onset (type 2 and type 3) SMA (NCT02913482 and NCT02908685).²⁶ The approval of risdiplam provides an additional option for treatment that avoids the complexity of frequent lumbar punctures and addresses the possibility of treatment of the extra-CNS manifestations of SMA.²⁷

NEBWORN SCREENING, NEW CARE MODELS, AND THE CREATION OF A NEUROGENTEICS URGENCY

The rapid approval of multiple treatment options for SMA over a short period of time has excited the community and challenged providers to implement new standard of care models. Prior to the approvals of nusinersen, onasemnogene abeparvovec, and risdiplam for SMA, treatment paradigms had evolved from a primarily palliative or reactive approach, to a more proactive care model with attention to a multitude of needs including respiratory, nutrition, and orthopedic. Consensus statements on standard of care, initially published in 2007 and revised in 2018, reflect these changes. But those documents do not address care standards in light of newly approved treatments.^{10, 11, 28} Comprehensive reviews detailing these rapid changes in the care for patients with SMA have recently been published.^29-31

One of the most significant developments in the new models of care for SMA is the implementation of newborn screening. While there had been longstanding interest in the community of patients and families of patient with SMA for adding SMA to newborn screening programs, the condition was not added to the recommended uniform screening panel (RUSP) of the United States after a preliminary review by the Secretary's Advisory Committee on Heritable Disorders and Genetic Diseases in Newborns and Children (ACHDNC) in 2008. The decision to defer a formal evidence review for SMA was based largely on the absence of a clearly beneficial treatment.³² The effort was re-initiated after the approval of nusinersen and growing evidence for better outcomes in patients with earlier treatment.⁹ After a formal re-review, the ADHCNS recommended inclusion of SMA on the RUSP in 2018, and several American states began efforts to implement screening. As of May 2021, 36 American states have implemented newborn screening programs for SMA, covering 74% of infants born in the United States.³³

Data continue to emerge suggesting marked benefits of early, pre-symptomatic treatment as compared to symptomatic treatment. In the open-label NURTURE trial (A Study of Multiple Doses of Nusinersen (ISIS 396443) Delivered to Infants With Genetically Diagnosed and Presymptomatic Spinal Muscular Atrophy), 15 subjects with 2 copies of SMN2 and 10 subjects with 3 copies of SMN2 were treated pre-symptomatically with nusinersen.³⁴ All 25 subjects achieved the ability to sit without support. By comparison, only 6/73 symptomatic infants with type 1 SMA treated with nusinersen on the ENDEAR study (symptomatic type 1 SMA treated with nusinersen) achieved this milestone.⁹ Furthermore, 22/25 subjects achieved walking independently including 12 of the 15 subjects with 2 copies of SMN2. Similar studies in pre-symptomatic SMA using onasemnogene abeparvovec (clinicaltrials.gov: NCT03505099) and risdiplam (clinicaltrials.gov: NCT03779334) are ongoing. Recent data from nascent newborn screening programs for SMA in Australia and Germany also demonstrated benefits in motor development from early diagnosis and treatment.^{35, 36}

With rapid loss of motor neurons in the first weeks of life and significant benefits of early treatment, there is increasing pressure to treat patients as early as possible. To facilitate decision making for providers managing newborn screening for SMA, a treatment algorithm was developed in 2018 based on expert consensus.³⁷ At that time, nusinersen was the only approved treatment and the proposed algorithm was based on the complexity of a treatment regimen with ongoing lumbar punctures for a lifetime. The recommendation was for immediate treatment for newborns with 2 or 3 copies of SMN2, although a specific timeline was not proposed. There was disagreement about immediate treatment vs. delayed treatment with screening for symptoms for newborns with 4 or more copies of SMN2. Ultimately, consensus was achieved for a recommendation of delayed treatment with careful screening for presentation of symptoms for patients with more than 4 copies of SMN2, but consensus was not reached for those with exactly 4 copies of SMN2. Screening recommendations included neurophysiologic studies (electromyography and nerve conductions), formal motor function assessments such as the Children’s Hospital of Philadelphia Infant Test of Neuromuscular Disorders (CHOP-INTEND),³⁸ and careful neurologic examination every 3-6 months until age two and every 6-12 months thereafter. With the availability of emerging clinical data and real-world experience, the working group re-convened in 2019. After approval of onasemnogene abeparvovec, the group’s updated recommendation is to initiate immediate treatment for newborns with 4 copies of SMN2.³⁹

Implementation of newborn screening programs have helped to facilitate early diagnosis and treatment, but many challenges remain. Recently, the ADHCNS released a report on the status of newborn screening for SMA, highlighting a number of challenges with implementation in the three years after the addition of SMA to the RUSP.⁴⁰ Two issues standout: 1- obtaining clinical care for diagnostic confirmation and guidance on time-sensitive treatment decisions can be challenging outside major centers without experience in caring for patients with SMA; and 2-insurance authorization for treatment can be lengthy and as such delay the treatment. Navigating these challenges can have lasting consequences, even when the delay is “only” by a few days.

DOES 1 WEEK MATTERS?

Two patients recently treated at our center highlight the urgency to treat promptly. With the start of newborn screening for SMA in the state of Utah in 2018, our center developed goals for treatment that include: a- neuromuscular subspecialty consultation within 1-2 days following notification of a positive case from the Department of Health: and b- initiation of treatment by 4 weeks of life.

Patient 1 was identified by the newborn screening at 5 days of life and was seen by our team the following day. The patient was part of an out-of-state adoption (born in Utah) to adoptive parents living in a different state where newborn screening is not available. Confirmatory testing was available within 1 week showing absent SMN1 and 2 copies of SMN2. The child’s physical exam was unremarkable. Anti-AAV9 antibodies were absent, and all screening labs were normal. The parents elected to pursue treatment with onasemnogene abeparvovec. CHOP-INTEND score at day of life 7 was 50 and at day of life 23 (1 day prior to infusion) was 52 (scores range from 0-64 with higher scores indicating better motor function). A provider was identified within driving distance of the family’s home and plans were made to transfer care, but the treatment would be delayed by 3-4 weeks. With some complexity in arranging transfer of care, and frustration at the potential delays, the questions was posed, “Does 1-week matters?” The family opted to return to our center and the child was treated 1 week later with onasemnogene abeparvovec at 24 days of life. The post gene-transfer course was unremarkable, and at 1 year of age, the child is meeting all developmental milestones on time, including walking independently at 11 months of age.

Patient 2 was also identified by the newborn screening at 5 days of life and seen by us the following day. Confirmatory testing showed absent SMN1 and 2 copies of SMN2. In this case, however, elevated anti-AAV9 antibodies (presumably maternal) were identified. Other tests were unremarkable, and the physical and neurologic examinations were normal. In this case, we requested prior authorization from the patient’s insurance for treatment with nusinersen and then conversion to treatment with onasemnogene abeparvovec after the anti-AAV antibodies were cleared by the body. However, the insurance policy explicitly denied any treatment with onasemnogene abeparvovec in a patient previously treated with nusinersen. After discussion with the family and some days lost in discussions with the insurance provider, we proceeded with a plan for treatment with nusinersen. While the clinical and neurologic examinations were normal at the last clinic visit on day of life 12, the patient had become notably hypotonic 1 week later with typical diaphragmatic breathing pattern seen in patients with type 1 SMA patients. The CHOP-INTEND score had dropped from 36 at day of life 12, to 31 on day of life 19. Nusinersen was successfully initiated on day of life 20, but the patient suffered an aspiration event soon after and was transferred to the pediatric intensive care unit where he was intubated for most of the next 3 weeks. Ultimately, he recovered and completed four doses of nusinersen and transitioned to onasemnogene abeparvovec at 4 months of age. At one year of age he continues to make developmental progress, but has marked weakness, persistent dysphagia with aspiration, and prominent gross motor delays.

As these two patients illustrate, there are a number of gaps in our understanding of early and, “pre-symptomatic” SMA, in part because it is not entirely clear that there is a true pre-symptomatic state. We lack effective motor function assessments and biomarkers that could differentiate early progressors from later progressors in a timeframe where even a few days can make a difference. We are intrigued by the 16-point difference in baseline CHOP-INTEND score between these two patients (50 vs. 36), and the change in score from baseline to the day prior to treatment (+2 vs. −5). More detailed studies of motor function, electrophysiology and biomarkers may help to identify early progressing patients in a shorter timeframe.

CONCLUSION

The treatment and care for patients with SMA has changed dramatically over the last five years. This change was born by with three FDA approved treatments, implementation of newborn screening programs in many American states, and a rapidly changing natural history of the disease. Providers have been challenged to implement these changes in a complex medical environment where rapid change can conflict with established care models and a slow-moving infrastructure. Neuromuscular providers have adapted the moniker “time is brain” - coined in 1993 to promote early treatment of stroke - to “time is muscle” highlight the urgency to treat quickly.⁴¹ The recommendation to treat patients immediately after they have been identified by the newborn screening as having 2, 3, or 4 copies of SMN2 is complicated by institutional, administrative, and insurance related barriers. But we are hopeful that the moniker, “time is muscle,” will facilitate a better understanding of the urgency to treat patients with SMA. Based on our experience and the patients highlighted here, we have revised our expectations. Patients with SMA identified through the newborn screening program should have treatment available within 14 days of life.