This page is an English translation of the Korean original. If there is any discrepancy, the Korean version prevails.
| Item | Details |
|---|---|
| A. Classification codes | ICD-10: G12.8 / G12.9 | Special Copayment Reduction: V123 | Coverage period: 5 years |
| B. Synonyms | Other spinal muscular atrophies and related syndromes (G12.8), spinal muscular atrophy, unspecified (G12.9); non-5q SMA, non-SMN1 SMA, distal SMA, Finkel-type SMA, Scaffer-Volpe syndrome |
| C. Adult onset | Low to high likelihood of first onset in adulthood (heterogeneous group) — some types begin in the neonatal period, BICD2- and DYNC1H1-associated types in infancy, and the Finkel type (VAPB) in adulthood |
| 1. History | Non-SMN1 SMA was long clinically misclassified as SMN1 SMA; advances in NGS have spurred identification of causative genes |
| 2. Causes | Mutations in various genes such as BICD2, DYNC1H1, TRPV4, VAPB, SIGMAR1; some of unknown cause |
| 3. Genetics | Various inheritance patterns (autosomal dominant/recessive, X-linked); disease mechanisms differ by causative gene |
| 4. General epidemiology | Non-5q SMA is about 5% of all SMA; each subtype is very rare (less than 0.1 per 100,000) |
| 5. Pathophysiology | Gene-specific mechanisms such as impaired axonal transport (BICD2, DYNC1H1), endoplasmic reticulum stress (VAPB), and mechanosensory channels (TRPV4) |
| 6. Symptoms | Proximal or distal muscle weakness, respiratory impairment, joint contractures; phenotype varies by gene |
| 7. Signs | Flaccid paralysis, reduced deep tendon reflexes, muscle atrophy; scapular winging (scapula alata) is common with BICD2 |
| 8. Screening | Non-5q gene panel for SMA phenotypes negative for SMN1 deletion; reconfirmation of negative newborn SMA screening |
| 9. Diagnosis | NGS gene panel (including non-5q SMA), WES/WGS, EMG, muscle biopsy, muscle MRI |
| 10. Treatment | No established disease-modifying therapy; supportive care centered on rehabilitation, respiratory support, and assistive devices |
| 11. Prognosis | Varies greatly by gene; some have a mild course, some (infantile type) are severe; mostly progressive |
| 12. Prevention | Genetic counseling, prenatal diagnosis, and PGT-M after identification of the causative gene |
| 13. References | Peeters K et al. Brain 2014 and 9 other papers |
Spinal muscular atrophy (SMA) was long understood to consist mostly of 5q SMA caused by SMN1 gene abnormalities. However, some patients showed a lower motor neuron disease phenotype clinically similar to SMA despite having no SMN1 deletion, and they were grouped under the provisional classification "non-5q SMA" or "non-SMN1 SMA." Until the late 20th century, it was difficult to identify the causative gene in most of these patients, and they were classified as "other spinal muscular atrophy" or "unspecified spinal muscular atrophy." With advances in molecular genetics and next-generation sequencing (NGS) technology, since the 2000s various causative genes have been identified one after another, including BICD2 (Bicaudal D homolog 2), DYNC1H1 (dynein cytoplasmic 1 heavy chain 1), TRPV4 (transient receptor potential vanilloid 4), and VAPB (VAMP-associated protein B). Finkel SMA is an adult-onset autosomal dominant SMA caused by VAPB gene mutations, first described by Finkel in 1962. Scaffer-Volpe syndrome is a congenital proximal SMA caused by BICD2 mutations. Even now, the causative gene remains unidentified in about 55–70% of non-5q SMA patients, and discovery of new genes through whole-genome sequencing (WGS) is ongoing.
Other and unspecified spinal muscular atrophy (non-5q SMA) is a heterogeneous group caused by mutations in various genes other than SMN1 gene abnormalities. The main identified causative genes and associated syndromes are as follows. BICD2 (Bicaudal D homolog 2) mutations cause SMALED2 (spinal muscular atrophy with lower extremity dominance type 2), characterized by lower-limb-predominant proximal muscle weakness and scapular winging. DYNC1H1 mutations cause SMALED1, which may be accompanied by learning disability. TRPV4 (TRP vanilloid 4 channel) mutations show diverse phenotypes, including scapuloperoneal SMA and congenital arthrogryposis. VAPB mutations cause a spectrum including adult-onset Finkel-type SMA and ALS8. SIGMAR1 (sigma-1 receptor) mutations are associated with juvenile-onset distal SMA. SETX (Senataxin), GARS (glycyl-tRNA synthetase), and others can also cause SMA-like phenotypes. Fazio-Londe progressive bulbar palsy has been shown to be caused by SLC52A2 (riboflavin transporter 2) or SLC52A3 mutations. Some non-5q SMA occurs sporadically (de novo or unidentified genes).
Non-5q SMA is not defined by a single inheritance pattern; it is a heterogeneous group that includes autosomal dominant (AD), autosomal recessive (AR), and X-linked (XL) inheritance depending on the causative gene. BICD2 mutations are autosomal dominant, and de novo mutations are also common. DYNC1H1 mutations are also autosomal dominant. TRPV4 mutations are autosomal dominant and characterized by variable penetrance and phenotypic diversity. VAPB mutations are autosomal dominant, with wide variation in age of onset within families. SIGMAR1 mutations are autosomal recessive. Fazio-Londe bulbar palsy due to riboflavin transporter deficiency is autosomal recessive. When NGS gene panels are applied to non-5q SMA patients, the diagnostic yield is about 30–45%, consistent with findings that the three genes BICD2, DYNC1H1, and TRPV4 account for about 2/3 of confirmed cases. WES (whole-exome sequencing) or WGS (whole-genome sequencing) can additionally search for novel causative gene variants not found on existing panels. Variant pathogenicity classification (ACMG criteria) and bioinformatic analysis are important in the diagnostic workflow.
Non-5q SMA is a very rare group of diseases accounting for less than about 5% of all SMA. Individual gene-associated subtypes are ultra-rare diseases with only tens to hundreds of families reported worldwide. The prevalence of all non-5q SMA is difficult to estimate precisely but is estimated at 0.05–0.2 per 100,000. Sex bias varies by subtype, and the BICD2-, DYNC1H1-, and TRPV4-associated types occur in both males and females. Age of onset is highly variable, ranging by subtype from the neonatal period (type with congenital arthrogryposis), infancy, and early childhood (BICD2, DYNC1H1) to adolescence (SIGMAR1) and adulthood (VAPB, some TRPV4). Epidemiological data on non-5q SMA in Asian countries including Korea are extremely limited, and accumulation of data from rare disease registry programs is needed. The rate of gene identification varies by country depending on medical infrastructure and access to genetic testing. The possibility of non-5q SMA should always be considered in infants with an SMA phenotype whose SMN1 newborn screening is negative. When there is a family history, early identification of the causative gene is important for preventive genetic counseling within the family.
The pathophysiology of non-5q SMA leads to selective death or dysfunction of lower motor neurons through different molecular mechanisms depending on the causative gene. BICD2 and DYNC1H1 are both components of, or related adaptor proteins for, the dynein-dynactin complex responsible for intracellular retrograde axonal transport. Mutations in these genes disrupt the transport of mitochondria, vesicles, and neurotrophic factor receptors between the cell body and the growth cone, leaving motor neurons in a state of energy deficiency and deficient trophic factor signaling. Learning disability accompanies DYNC1H1 mutations because dynein is essential for cortical neuronal migration. TRPV4 is a nonselective cation channel that senses mechanical stimuli, temperature, and osmotic changes. Mutant TRPV4 is constitutively activated (gain-of-function), causing excessive calcium ion influx, which leads to excitotoxic death of motor neurons. VAPB (VAMP-associated protein B) regulates lipid transport, the UPR (endoplasmic reticulum stress response), and calcium homeostasis at endoplasmic reticulum–mitochondria contact sites (MAM, mitochondria-associated membrane). Mutant VAPB forms aggregates and interferes with normal VAPB function (dominant-negative mechanism), causing endoplasmic reticulum stress and mitochondrial dysfunction.
The clinical symptoms of non-5q SMA vary greatly by causative gene and subtype. Common to all, muscle weakness and atrophy due to lower motor neuron involvement are central, and the absence of upper motor neuron signs is characteristic. In BICD2-associated SMALED2, lower-limb-predominant proximal muscle weakness appears from birth or infancy, accompanied by scapular winging (scapula alata), hip dislocation, and foot deformities. Some are mild and maintain walking ability, whereas severe cases require respiratory support. In DYNC1H1-associated SMALED1, mild to moderate proximal muscle weakness appears together with congenital hip dislocation and foot deformities, and some have cognitive impairment or cortical dysplasia. TRPV4-associated scapuloperoneal SMA is characterized by atrophy and weakness of the shoulder and calf regions. Finkel-type SMA (VAPB) begins in adulthood with proximal muscle weakness and progresses very slowly. In Fazio-Londe progressive bulbar palsy, bulbar symptoms (dysarthria, dysphagia, facial paralysis) and respiratory muscle weakness are mainly prominent. Arthrogryposis may accompany severe congenital subtypes.
Physical examination findings in non-5q SMA commonly show flaccid paralysis, reduced or absent deep tendon reflexes, and muscle atrophy reflecting lower motor neuron lesions. Upper motor neuron signs (spasticity, positive Babinski, hyperreflexia) are absent, an important point of distinction from ALS and cervical myelopathy. In the BICD2-associated type, scapular winging (scapula alata; the scapula protrudes on the back when the arms are extended forward) is a characteristic sign reflecting weakness of the serratus anterior. A point distinguishing BICD2 from SMN1 SMA is that deep tendon reflexes are sometimes preserved or even brisk. In the DYNC1H1-associated type, proximal lower-limb atrophy, weakness of ankle plantar flexion, and in some cases foot deformities (equinus, clubfoot) are observed. In the severe neonatal-onset type with congenital arthrogryposis, fixed flexion or extension deformities of limb joints are evident from birth. In the TRPV4-associated scapuloperoneal type, atrophy of the deltoid, biceps, and gastrocnemius is prominent. Overall, sensory function is normal in non-5q SMA because anterior horn cells (motor neurons) are selectively involved. Mild sensory neuropathy may accompany some subtypes (DYNC1H1).
No specific newborn screening for non-5q SMA is currently established. There are two main screening scenarios. First, in infants who are negative on newborn SMA screening (SMN1 exon 7 deletion test) but show a clinically SMA-like phenotype (hypotonia, muscle weakness, flaccid paralysis, fasciculations), non-5q SMA should be strongly suspected and additional gene panel testing performed. Second, when there is a family history of non-5q SMA, targeted genetic testing for the familial variant is offered. In the non-5q SMA diagnostic workflow, a rapid diagnostic pathway is more important than screening; to this end, a protocol is recommended in which, for a suspected SMA phenotype, SMN1 deletion is checked and, if negative, a broad neuromuscular gene panel including a non-5q SMA panel (BICD2, DYNC1H1, TRPV4, VAPB, etc.) is performed immediately. The longer the diagnosis is delayed, the more likely inappropriate treatment (e.g., nusinersen administration) may be attempted, so accurate genetic confirmation is essential for treatment decisions. Electromyography and nerve conduction studies serve as selective supplementary tools to confirm lower motor neuron lesions and guide the diagnostic direction.
Diagnosis of non-5q SMA is made by comprehensive genetic testing after SMN1 negativity is confirmed. In step 1, homozygous SMN1 exon 7 deletion is excluded by MLPA or qPCR. In step 2, a neuromuscular disease–associated multigene panel (including 200–400 genes) is performed, including non-5q SMA–associated genes (BICD2, DYNC1H1, TRPV4, VAPB, SETX, SIGMAR1, SLC52A2/3, etc.). If the panel is not diagnostic, testing is expanded to whole-exome sequencing (WES) or whole-genome sequencing (WGS). On electrophysiological testing, electromyography shows widespread denervation findings, and nerve conduction studies mostly show normal sensory nerve conduction and reduced motor CMAPs. Mild abnormalities of sensory nerve conduction may be observed in some subtypes. Muscle MRI shows patterns of selective involvement of specific muscle groups, helping classify phenotypes by subtype. Muscle biopsy shows denervation atrophy (grouped atrophy) and reinnervation findings but is no longer essential. Serum CK is mildly elevated or normal. Fundus examination (retinal abnormalities possible in some congenital subtypes) and brain MRI (to identify cortical dysplasia in the DYNC1H1-associated type) are also performed as needed.
There is currently no approved disease-modifying therapy for non-5q SMA. Treatments approved for SMN1 SMA, such as nusinersen, risdiplam, and onasemnogene, are not effective in non-5q SMA, so accurate genetic diagnosis is essential to prevent attempts at unnecessary treatment. In Fazio-Londe bulbar palsy (SLC52A2/3 mutations), high-dose riboflavin (vitamin B2) supplementation has shown marked clinical improvement in some patients, an important therapeutic exception; in this specific subtype, an early trial of riboflavin treatment is strongly recommended. In TRPV4 mutation–associated SMA, TRPV4 antagonists are being studied in animal models. Overall supportive care forms the foundation of treatment, including maintenance of muscle strength and preservation of function through physical and occupational therapy, assistive devices (lower-limb orthoses, spinal orthoses, walkers, wheelchairs), scoliosis correction surgery, noninvasive ventilation (when respiratory muscle strength declines), management of dysphagia (speech therapy, dietary modification, gastrostomy), and support for school and social integration. Collaboration of a multidisciplinary team (neurology, rehabilitation medicine, orthopedics, pulmonology, nutrition) is important. In research, development of gene-specific targeted therapies is actively underway.
The prognosis of non-5q SMA varies greatly by causative gene and clinical subtype. BICD2-associated SMALED2 has a broad phenotypic spectrum: in mild forms, walking ability is maintained throughout life and life expectancy is close to normal, whereas in severe congenital forms respiratory support is needed from the neonatal period and mortality is high. DYNC1H1-associated SMALED1 mostly shows a mild to moderate progressive course, and walking ability is often maintained. When cognitive abnormalities are present, education and social functioning are additionally affected. VAPB-associated Finkel-type SMA begins in adulthood and progresses very slowly; life expectancy is close to normal, but functional decline accumulates over decades. Fazio-Londe bulbar palsy (SLC52A2/3) had a very poor prognosis before treatment, but can improve markedly with riboflavin treatment, so early genetic diagnosis determines the prognosis. The type with congenital arthrogryposis is severe, and early supportive care is important for improving prognosis. Overall, in non-5q SMA, accurate genetic diagnosis is the starting point for predicting prognosis and planning treatment.
Prevention of non-5q SMA is possible through genetic counseling and reproductive medicine approaches when the causative gene has been identified. The recurrence risk is calculated according to the causative gene and inheritance pattern, and a prevention strategy suited to that risk is selected. In autosomal dominant types (BICD2, DYNC1H1, TRPV4, VAPB), children of the proband have a 50% risk. Preimplantation genetic testing (PGT-M), performed together with in vitro fertilization, can prevent the birth of an affected child by transferring embryos that do not carry the variant. Prenatal genetic diagnosis (amniocentesis or chorionic villus sampling) can determine whether the fetus carries the variant during pregnancy. In autosomal recessive types (SIGMAR1, SLC52A2/3), when both parents are carriers, the risk of an affected child is 25%. In cases due to de novo mutations (common in BICD2, DYNC1H1), even if no abnormality is found on parental genetic testing and the recurrence risk in subsequent pregnancies is theoretically low, the possibility of recurrence due to gonadal mosaicism cannot be completely excluded. In terms of secondary prevention, starting riboflavin treatment immediately upon diagnosis of Fazio-Londe bulbar palsy is decisive for preserving function.