Adrenal& Bone Complications in Paediatric Patients Living With Spinal Muscular Atrophy (SMA)

Spinal muscular atrophy (SMA) is an autosomal recessive neuromuscular disorder characterised by the dysfunction and degeneration of specialised nerve cells, the motor neurons1 This leads to progressive muscle wasting and weakness affecting mobility, respiration and bulbar function. The worldwide incidence of SMA is approximately 1 in 10,000 live births 2,3. The disease is caused by homozygous deletion and/or mutations of the survival motor neuron 1 (SMN1) gene on chromosome 5, encoding SMN protein. SMN protein is vital for the survival and health of motor neurons.

In humans, SMN protein is also encoded by a paralogous gene, SMN2, also located on chromosome 5 in varying copy numbers. Due to a single nucleotide difference from SMN1 (a change of cytosine to thymidine in exon 7), the majority of SMN protein encoded is truncated and unstable. Approximately 10% of SMN protein produced from SMN2 is functional and can compensate for SMN1 mutations.3 Disease severity of SMA is therefore largely dependent on the number of SMN2 copies, with greater copy numbers associated with milder forms of the disease. SMA has been classified into five types (0,1,2,3,4) based on age of onset and best motor function. Type 0 is a very rare severe type of SMA with symptoms beginning in-utero. Type 1, a severe and the most common form (50-60% of cases), with onset before 6 months, is associated with one to two copies of SMN2. Untreated type 1 SMA usually leads to death before the age of 2. In contrast, type 4, the mildest form with adult onset, is associated with more than four copies of SMN2.

There are now three commercially available drugs shown to improve survival and mobility of patients with SMA1.

They work either by increasing functional SMN protein production via targeting SMN2 activity (risdiplam and nusinersen) or by introducing a fully functional copy of the SMN1 gene (onasemnogene abeparvovec). The drugs can be administered locally, to target key functional areas (intrathecal nusinersen), or systemically (orally administered risdiplam and intravenously administered onasemnogene abeparvovec).

Since introduction of commercially available drugs for spinal muscular atrophy (SMA) in the UK in 2018, mostpatients with SMA, particularly the paediatric population (under 16 years old) have tried at least one of the three disease modifying drugs that increases SMN protein production. These drugs evidently improve survival and ventilatory outcomes along with changes to other domains of function as well (e.g., mobility). Patients are now surviving longer than in the natural history of the condition. Cell culture systems suggest SMN protein interacting with osteoclast stimulating factor (OSF) leading to signaling cascade inducing stimulators for osteoclast (cells that degrade bone) formation5. Smn-/- SMN2 mouse models of SMA had osteoporotic bone phenotype on microcomputed tomography analysis. Results also show decreased osteoblast differentiation markers, osteocalcin, osteopontin and osterix mRNA expression with increased rate of osteoclast formation and bone resorption capacity, all these indicating SMA mice have higher bone turnover with increased bone resorption and osteoclast formation. Immobility or reduced mobility and the subsequent loss of muscle loading of bone in SMA can lead to secondary osteoporosis with low bone mass and increased fracture risk. There have been case reports of congenital fractures in SMA7,8, but also in older patients9. Furthermore, when bone mineral density (BMD) is compared between different paediatric neuromuscular disorders including Duchenne muscular dystrophy, children with SMA had the lowest BMD, although use of steroids or vertebral fractures were not controlled in the Duchenne muscular dystrophy group10. BMD also differed significantly between walkers and non-walkers (Z-score of -1.07 vs -2.70, P\< 0.05) in the SMA cohort with no difference in the mean age of the walkers and non-walkers10.

The largest natural history cohort studies to date, prior to use of SMA specific drugs, of children with SMA and bone health was performed in 2017, and included 85 children with type 1 to 3 SMA (24 SMA type 1, 44 SMA type 2, and 17 SMA type 3) but excluded children less than 12 months old and those on drugs or medical conditions known to affect bone metabolism e.g. corticosteroids11. The study concluded that SMA children had high prevalence of low BMD and fractures at a young age, with SMA type 1 having the lowest areal BMD at all skeletal sites. Low BMD was common at the lateral distal femur where fractures often occur in these patients, but only 12.9% met the criteria of osteoporosis in children (The International Society for Clinical Densitometry, ISCD, 2013 peadiatric criteria).

There has only been one small study (N=32)12 published which included patients on novel drugs for SMA looking at bone health in children with SMA. This study only had BMD data on two type 1 SMA patients, the most common and severe type of SMA. The study did not have any patients who received onasemnogene abeparvovec treatment.

Low BMD in the paediatric SMA population has been reported variably between 16.7 to 93.3% over the last twenty years8 and there is no agreement on factors that are associated with low BMD. Femur fractures seem the predominant fracture type, with the distal femur being involved most commonly11,12. There is only one study specifically on intravenous bisphosphonate use in children with SMA for osteoporosis and low BMD and included 8 children (6 type 1 and 2 type 2 SMA) who were all on nusinersen13. The study showed that intravenous bisphosphonates appeared safe and effective in reducing fractures in children with SMA.

A retrospective analysis of 28 children with SMA (16 type 1, 7 type 2, 4 type 3 and 1 pre-symptomatic) receiving neuromuscular care in Sheffield Children's NHS Foundation Trust (SCFT), all of whom have now had SMA drug treatment with either one or more of SMA drugs, nusinersen, risdiplam or onasemnogene abeparvovec, showed that 13/28 (46%) had osteopenia reported radiologically, some as early as 1 year of age. These are seen in various bones on x-ray e.g. ribs, pelvis, femur, vertebrae and feet and can also be detected in some on dual energy Xray absorptiometry (dual-energy X-ray absorptiometr (DXA) scan) with or without vertebral fracture assessment (VFA). 1 in 7 (4 children) had a fracture and the majority (4 of 6) were in the femur, some with very trivial impact. One child had two fractures on the same femur at different timepoints and a vertebral fracture. Two children had bisphosphonate treatment for osteoporosis, vertebral fractures were present in both.

There are currently no national or international guidelines, nor consensus, on the management of secondary osteoporosis or fragility fractures in patients living with SMA14, particularly after introduction of SMA modifying drugs. There are no licensed treatments for the management of osteoporosis in children. The mainstay of treatment is with intravenous bisphosphonates and its duration is important to explore. The need for immunosuppression with prednisolone post-omnasemnogene aberparvovec infusion may have secondary effects to reduce bone integrity. There is no evidence to date on the effect of a course of prednisolone on bone health in infants and very young children with SMA.

Investigators propose reviewing data available on the paediatric SMA REACH UK database (currently about 600 active patients) on bone health (DXA scan results, vitamin D levels and supplements, fractures, endocrinologist input, bisphosphonate use) and associated parameters that may contribute to bone health (SMA type, SMN2 copy numbers, ambulation, motor scores, use of SMA drugs, and use of other drugs including steroids contributing to bone health, growth parameters). Investigators would then obtain further information on patients with bone health events or risk factors for suboptimal bone health (fractures, DXA scan, bisphosphonate treatment and long-term steroid treatment) from the local neuromuscular centres that are not recorded on the paediatric SMA REACH UK database. Understanding bone health status in this retrospective observational dataset will allow us to consider important factors in bone health in SMA patients and design prospective studies to improve bone health outcomes and inform standards of care in this field.

The primary hypothesis to be tested is that patients with spinal muscular atrophy (SMA) have low bone density and increased fracture risk. The secondary hypothesis is that treating SMA with commercially available drugs that increase SMN protein production improves bone density and reduces fracture risks in patients with SMA.