Sage Journals: Discover world-class research

Abstract

The availability of disease modifying therapies for spinal muscular atrophy (SMA) has created an urgent need to identify clinically meaningful biomarkers. Biomarkers present a means to measure and evaluate neurological disease across time. Changes in biomarkers provide insight into disease progression and may reveal biologic, physiologic, or pharmacologic phenomena occurring prior to clinical detection. Efforts to identify biomarkers for SMA, a genetic motor neuron disease characterized by motor neuron degeneration and weakness, have culminated in a number of putative molecular and physiologic markers that evaluate biological media (eg, blood and cerebrospinal fluid [CSF]) or nervous system function. Such biomarkers include SMN2 copy number, SMN mRNA and protein levels, neurofilament proteins (NFs), plasma protein analytes, creatine kinase (CK) and creatinine (Crn), and various electrophysiology and imaging measures. SMN2 copy number inversely correlates with disease severity and is the best predictor of clinical outcome in untreated individuals. SMN mRNA and protein are commonly measured in the blood or CSF of patients receiving SMA therapies, particularly those aimed at increasing SMN protein expression, and provide insight into current disease state. NFs have proven to be robust prognostic, disease progression, and pharmacodynamic markers for SMA infants undergoing treatment, but less so for adolescents and adults. Select plasma proteins are altered in SMA individuals and may track response to therapy. CK and Crn from blood correlate with motor function and disease severity status and are useful for predicting which individuals will respond to therapy. Electrophysiology measures comprise the most reliable means for monitoring motor function throughout disease course and are sensitive enough to detect neuromuscular changes before overt clinical manifestation, making them robust predictive and pharmacodynamic biomarkers. Finally, magnetic resonance imaging and muscle ultrasonography are non-invasive techniques for studying muscle structure and physiology and are useful diagnostic tools, but cannot reliably track disease progression. Importantly, biomarkers can provide information about the underlying mechanisms of disease as well as reveal subclinical disease progression, allowing for more appropriate timing and dosing of therapy for individuals with SMA. Recent therapeutic advancements in SMA have shown promising results, though there is still a great need to identify and understand the impact of biomarkers in modulating disease onset and progression.

Keywords

Spinal muscular atrophy biomarker molecular biomarkers electrophysiology neurofilament proteins

Introduction

Spinal muscular atrophy (SMA) is a degenerative motor neuron disorder and a prominent genetic cause of infantile death, with an incidence of 1 in 11 000 live births.^1,2 SMA is caused by a homozygous deletion or mutation of the survival motor neuron 1 (SMN1) gene. SMN2, a nearly identical gene, is able to produce a small quantity of full-length SMN protein, enough to prevent lethality in humans born with SMA. At least 1 copy of SMN2 is required for survival in utero, with subsequent copies inversely correlating with disease severity.^3-7 Though SMN1 and SMN2 only differ by a C to T nucleotide substitution, this single base change causes skipping and exclusion of exon 7 during splicing, resulting in SMNΔ7 mRNA that produces a truncated, nonfunctional protein.^8-13 This protein is ultimately ubiquitinated for lysosomal degradation,^14-16 and thus is unable to compensate for the full-length SMN normally produced by SMN1. One notable exception to this general trend is the c.859G>C modifier that promotes exon 7 inclusion and allows for a moderately increased amount of full-length SMN mRNA in diseased individuals with 2 copies of SMN2, resulting in a mild type 3b phenotype.^13,17,18

Classified based on clinical features, there are 4 main subtypes of SMA, ranging from the earliest (ie, infantile), most severe onset (type 1) to childhood, non-ambulatory onset (type 2) to later childhood, ambulatory onset (type 3), to adult, mild onset (type 4). Sufficient SMN expression arising from increased SMN2 copy number can result in later-onset, mild forms of disease.

In the last decade, significant advancements have been made in SMA treatment, resulting in 3 FDA-approved therapies. Current approved therapies for SMA act by upregulating SMN protein in diseased individuals, either via increasing exon 7 inclusion in mature mRNA (Evrysdi [risdiplam]^19,20 and Spinraza [nusinersen]²¹) or AAV9-mediated gene replacement therapy (Zolgensma [onasemnogene abeparvovec]^22,23) with extraordinary success. However, considering the primary target of SMN-restoring therapies is the spinal cord, a feasible method for the longitudinal evaluation of SMN levels in surviving patient populations is still lacking. There remains an urgent need to develop biological markers (“biomarkers”) for improved detection and clinical management of the disease, as well as for understanding variability of therapeutic response between treated patients.

Biomarkers provide a powerful means of measuring and evaluating the spectrum of neurological disease, from diagnosis to death. These measures may include direct evaluations of biological media, such as blood or cerebrospinal fluid (CSF), as well as technologies measuring changes in the physiology or function of the nervous system. In this update, we review the current biomarkers available for detecting and monitoring biological and clinical features of SMA (Table 1). The ideal biomarker for SMA is one that is noninvasive, yet able to reflect systemic SMN levels, particularly those within protected nervous tissue. It must also be accurate and reliable. However, a single biomarker may be useful but not fulfill insights into all areas—disease progression, prognosis, and response to therapy—thus, a combination of biomarkers may provide greater insight and more accurate clinical assessment. Here, we refer to 5 classifications of biomarkers based on points of diagnosis and treatment: Diagnostic, Prognostic, Predictive, Disease Progression, and Pharmacodynamic. Markers of disease phenotype are considered “diagnostic” prior to diagnosis or “prognostic” prior to treatment initiation and afterwards are called “disease progression” markers. Likewise, markers of treatment response are considered “predictive” prior to treatment initiation and afterwards are “pharmacodynamic” markers.

Table 1.

Putative SMA biomarkers.

Putative biomarker	Type of biomarker	References
Molecular
Genetic testing	Diagnostic	Pearn,¹ Sugarman et al²
SMN2 copy number	Prognostic	Lefebvre et al,³ McAndrew et al,⁴ Monani et al,⁵ Feldkötter et al,⁶ Elsheikh et al,⁷ Schrank et al,²⁶ Hahnen et al,²⁷ Campbell et al,²⁸ Wirth et al,²⁹ Mailman et al,³⁰ Kolb et al,³¹ Kolb et al,³² Vezain et al,³³ Qu et al,³⁴ Stenson et al,³⁵ Wu et al,³⁶ Parsons et al³⁷
SMN mRNA and protein levels	Prognostic in some cases, pharmacodynamic for systemic therapies	Schrank et al,²⁶ Kolb et al,³¹ Kolb et al,³² Ramos et al,⁵⁷ Zaworski et al,⁵⁸ Burlet et al,⁵⁹ Giavazzi et al,⁶⁰ Poirier et al,⁶¹ Alves et al,⁶² Nash et al,⁶³ Bonati et al,⁷³ Arnold et al⁷⁹
Neurofilament proteins	Disease progression, prognostic, pharmacodynamic	Benatar et al,⁸⁶ Finkel et al,⁹² Darras et al,⁹³ De Vivo et al,⁹⁴ Tozawa et al,⁹⁵ Winter et al,⁹⁶ Spicer et al,⁹⁷ Wurster et al,⁹⁸ Wurster et al,⁹⁹ Totzeck et al,¹⁰⁰ Rich et al,¹⁰¹ Faravelli et al¹⁰²
Creatine kinase and creatinine	Predictive, pharmacodynamic, disease progression in infants with 3 copies of SMN2 (creatinine)	Rudnik-Schöneborn et al,¹⁰⁷ Alves et al,¹⁰⁹ Freigang et al,¹¹⁰ Deutsch et al¹¹¹
Physiologic
Compound muscle action potential (CMAP)	Disease progression, prognostic, predictive	Kolb et al,³¹ Kolb et al,³² Arnold et al,⁷⁹ Swoboda et al,¹¹⁴ David et al,¹¹⁸ Weng et al,¹¹⁹ Al-Zaidy et al,¹²² Kariyawasam et al,¹²³ Darras et al¹²⁴
Motor unit number estimation (MUNE)	Disease progression, prognostic, predictive	Swoboda et al,¹¹⁴ David et al,¹¹⁸ Al-Zaidy et al,¹²² Kariyawasam et al,¹²³ Darras et al¹²⁴
Motor unit number index (MUNIX) and motor unit size index (MUSIX)	Disease progression	Günther et al,¹³⁰ Querin et al,¹³¹ Nandedkar et al,¹³³ Sanchez and Rutkove,¹³⁶ Li et al¹⁴⁴
Electrical impedance myography (EIM)	Disease progression, predictive, pharmacodynamic	Kolb et al,³¹ Li et al,¹⁴⁴ Arnold et al,¹⁴⁵ Rutkove et al,¹⁴⁶ Rutkove et al,¹⁴⁷ Arnold et al¹⁴⁸
Magnetic resonance imaging (MRI)	Diagnostic, disease progression in some cases	Bonati et al,⁷³ Sproule et al,¹⁴⁹ Durmus et al¹⁵¹
Muscle ultrasonography	Diagnostic	Pillen et al,¹⁵² Wu et al,¹⁵³ Regensburger et al¹⁵⁵

Molecular Biomarkers

Survival motor neuron 2 (SMN2) copy number

SMN2 is an SMN1 paralog differing by 2 exonic base changes, which result in a significantly decreased ability to splice out intron 7 and thus to produce full-length, functional SMN mRNA, and protein from the SMN2 gene.^9,10,24,25 In humans, it is not uncommon for multiple copies of SMN2 to be present. The number of SMN2 copies, known as SMN2 copy number, is proportional to the amount of full-length SMN protein produced in humans and mouse models of SMA.^5,26 Further, an inverse relationship exists between SMN2 copy number and phenotypic severity, with a higher SMN2 copy number generally associated with a milder SMA phenotype.^4,27-30 In an infantile natural history study, SMA infants with fewer copies of SMN2 exhibited lower motor function scores (MFS) than those with greater copy numbers.^31,32 Reduced copy number also correlated with mortality and need for invasive ventilatory support. For copy number ⩾3, however, this relationship is less consistent. Due to its inverse correlation with disease severity, SMN2 copy number remains the best means for predicting future clinical outcome in an individual who has not received therapy and thus is a useful prognostic biomarker, particularly for individuals with fewer than 3 copies of SMN2. Additionally, SMN2 copy number is typically available in conjunction with genetic confirmation of disease. However, SMN2 copy number is fixed and will not change; therefore it has no value as a marker of disease progression or treatment response.

Rare SMN2 variants have been reported that modify the SMA disease phenotype. The c.859G>C variant in exon 7, originally identified in 2 male patients with type 3 SMA, was found to promote exon 7 inclusion and reduce disease severity.³³ One of the 2 patients had only 2 copies of SMN2 (both variant), a rare finding in patients with type 3 SMA. The other patient had 3 copies; 1 variant and 2 wildtype. The c.863G>T variant in exon 7 was reported in 3 individuals (male, 2 copies, type 1; female, 3 copies, type 1b; female, 3 copies, type 2) and shown to disrupt inclusion of exon 7 during splicing, causing SMA.³⁴ Other small intragenic deletions and missense variants have been reported in individuals with SMA, with variable correlations between SMN2 copy number and clinical phenotype.^29,35-37 Consequently, SMN2 copy number is not a reliable prognostic biomarker on its own and can be improved when variants are sought. Additional genetic modifiers outside of the SMN2 locus (Plastin 3,^38-40 Coronin 1C,⁴¹ and Neurocalcin Delta⁴²) and splicing regulators (microRNAs,^43-45 long non-coding RNAs,^46,47 DNA methylation,^48-50 and other epigenetic modifications⁵¹) have been investigated in pre-clinical settings and have the potential to be valuable prognostic biomarkers in the future.

The pressing need to identify affected individuals in pre-symptomatic or early stages of SMA, coupled with the strong prognostic power of SMN2 copy number, has led to an expansion of newborn screening for SMA. Current approaches involve detection of SMN1 deletion or intragenic mutations and quantification of SMN1 and SMN2 copy number, through a variety of laboratory methods.^52,53 While specific newborn screening protocols vary by state and country, the vast majority of initial screening includes detection of SMN1 deletion for diagnostic purposes. In conjunction with or following initial screening, second-tier tests are performed to confirm positive cases, quantify copy numbers, and assess risk for infantile onset of the disease. Carrier screening is also commonly performed in prenatal settings. Identification of SMA individuals through newborn screening, particularly those with lower SMN2 copy numbers who show the greatest benefit from therapeutic intervention, allows for prompt diagnosis and treatment of affected individuals, improving clinical outcomes.

Survival motor neuron (SMN) mRNA and protein levels

SMN mRNA and protein levels reflect the extent of SMN2 gene expression, but also the resultant transcription and translation, respectively. Similar to a greater SMN2 copy number, increased mRNA and protein levels correspond to milder types of SMA.^5,26 Hence, SMA therapies have primarily been focused on the development of drugs that increase SMN protein expression, especially in nervous tissue. SMN is ubiquitously expressed and thus can be detected in all cell types,²⁴ though previous studies have mainly measured SMN levels in blood and, less frequently, CSF. SMN expression varies between tissue types due to developmental and cell-type specific regulation.^54-56 This, combined with the multitude of approaches for measuring SMN expression, has resulted in a wide range of reported expression levels and has made comparisons across studies particularly challenging.

An important study recently reported SMA levels in perinatal post-mortem tissues from SMA and control individuals.⁵⁷ SMN protein levels in the spinal cord were highest prenatally, then decreased 6.5-fold during the perinatal period (3 months before to 3 months after birth) and remained low through adolescence.⁵⁷ This is consistent with previously reported perinatal SMN declines in both mouse⁵⁸ and human.^59,60 SMN1 and SMN2 gene expression showed similar trends, with the decline in SMN2 slightly preceding that of SMN1, by approximately 0 to 3 months. Developmental expression patterns of both mRNA and protein were comparable across disease-relevant tissues, including spinal cord, cortex, iliopsoas muscle, and diaphragm muscle. Collectively, these studies suggest that there is a greater need for SMN neonatally during normal development.

The ease of access to blood has made it a popular biological fluid for SMN detection, particularly when monitoring disease progression and response to therapy. For example, the first-in-human study of risdiplam demonstrated treatment-induced increases in full-length SMN2 mRNA levels, decreases in SMN2Δ7 mRNA levels, and increases in SMN protein levels in whole blood.¹⁹ Risdiplam has recently been shown to increase SMN in CSF as well as peripheral blood.⁶¹ Another study showed whole blood SMN protein levels negatively correlated with severity of denervation assessed by muscle electrophysiology.⁶² Serum exosome-derived SMN protein levels were also reflective of disease state in mice and humans, with the quantity of SMN protein correlating with animal genotype and human disease subtype.⁶³ Still, there is conflicting data regarding the ability of peripheral blood to accurately reflect central SMN expression and therapeutic potential. SMN mRNA and protein levels increased following administration of valproic acid—a histone deacetylase inhibitor with therapeutic potential for SMA⁶⁴—in multiple human cell lines including fibroblasts, but did not affect blood SMN levels nor clinical course.^65-70 Endogenous SMN levels differed between tissue types, with levels in peripheral blood generally lower than those of motor neurons.⁷¹ Importantly, these studies showed that even with the intended pharmacodynamic effect of increased histone acetylation with valproic acid treatment, there was no significant change in SMN levels. Results such as these, where a drug has its intended molecular effect but no clinical effect, shape how negative results should be interpreted in a clinical trial. A direct test of how SMN levels respond when motor neuron levels are changed was performed in a postnatal porcine model of SMA induced by the intrathecal delivery of an shRNA to decrease the porcine Smn expression.⁷² While SMN mRNA levels were effectively reduced in motor neurons, as measured from laser captured microdissected motor neuron samples, there was no reduction in SMN mRNA level in blood.⁷² This demonstrated a key limitation of peripheral blood biomarkers: mRNA and protein levels measured from blood, while convenient to obtain, do not necessarily reflect biological mechanisms occurring at the motor neuron—highlighting the importance of increasing SMN expression in accessible, as well as relevant, cell and tissue types.

Results from the previously described natural history study of infantile SMA also demonstrated a reduction in SMN protein levels in SMA infants compared to age-matched healthy infants.³¹ There was no correlation between age and SMN levels in either cohort, though SMN protein levels varied widely. A later study confirmed reduced SMN levels in SMA patients, but levels remained stable despite rapidly changing motor function—increasing in healthy infants and decreasing in SMA infants—and thus no correlation with MFS was identified.³² Of note, SMA infants at enrollment were already exhibiting disease symptoms and had reduced motor function compared to controls. Another group also reported stability in SMN mRNA and protein levels over the course of a year, and levels did not correlate with either muscle function or structural muscle integrity as measured by functional clinical scales, such as Motor Function Measure (MFM), 6-Minute Walk Test (6MWT), and quantitative MRI (qMRI) of thigh muscles.⁷³ Further, in a cross-sectional study evaluating SMN gene and protein expression in whole blood, there was no difference in SMN expression levels nor any other changes in gene expression correlating with disease severity between SMA and healthy cohorts.⁷⁴ This suggests that the degree of SMN protein level decline seen in SMA patients is not sufficient to cause a comparable decline in gene expression or splicing in peripheral blood mononuclear cells.

Numerous groups have utilized pharmacologic agents to increase SMN levels,^75-78 2 of which ultimately achieved FDA approval (nusinersen and risdiplam).^19-21 An additional agent which has not received FDA approval, morpholino antisense oligonucleotide (ASO), was used in an SMNΔ7 mouse line to target ISS-N1, a negative splice regulator downstream of exon 7.⁷⁹ ASOs are able to block splice modulating gene sequences to allow for, in the case of SMN2, incorporation of exon 7 and restoration of SMN levels.^80-82 A single injection of ASO increased SMN levels in both SMA and heterozygote mice, though not to the levels of control animals. More remarkably, morpholino ASO increased the median survival of SMNΔ7 mice from 2 to over 14 weeks. However, ASO delivery and distribution was systemic rather than tissue-specific as intended, there was a high level of variability in SMN levels between tissue types, and there was an age-related decline in SMN expression in all tissues. This alteration of SMN levels in non-central nervous system (CNS) tissues raises the interesting question of how systemic restoration could affect the utility of peripheral biomarkers in monitoring disease progression or treatment response.

While SMN mRNA and protein may provide some insight into current disease state, they have not been shown to change over time, making them poor candidates for monitoring disease progression and response to therapy. Further, SMN measured in blood is not necessarily reflective of levels in motor neurons or other nervous tissue. Nevertheless, blood SMN levels may prove to be an accessible and useful tool in understanding the systemic response to CNS-targeted therapies aimed at increasing SMN levels.

Neurofilaments (NFs)

Neurofilament (NF) is a cytoskeletal protein that regulates axonal caliber and maintains the structural integrity of the axon. It is released from neurons following injury, and elevated NF levels can be detected in both blood and CSF.^83,84 With a half-life nearing 8 months, NF may offer insight into axonal changes occurring many weeks prior to measurement.⁸⁴ The majority of NFs are high molecular weight, phosphorylated proteins—known as phosphorylated neurofilament heavy chain (pNF-H)—which display enhanced resistance to protein degradation.^83,85 NF also exists in light (NF-L) and medium chain conformations. As markers of active axonal loss, NF proteins have previously been studied in the context of axonal injury, degeneration, and disease.⁸⁴ Disease-related elevations in both NF-L and pNF-H have been demonstrated in such disorders as amyotrophic lateral sclerosis (ALS),^86-88 Charcot-Marie-Tooth disease,⁸⁹ and Alzheimer’s disease.^90,91 Over the last few years, there has been a surge in studies investigating the utility of NF proteins as biomarkers for SMA, particularly in patients receiving nusinersen. Nusinersen, now FDA-approved for all patients with SMA, is an antisense oligonucleotide that modifies SMN2 splicing to promote the inclusion of exon 7 in the spliced mRNA transcript, thereby increasing the expression of full-length SMN protein. Due to ease of measurement in blood, NF proteins offer a feasible avenue for monitoring neurodegeneration.

Plasma and CSF NF levels have been shown to be elevated in SMA patients, predict disease severity prior to nusinersen treatment, and decline rapidly within 2 months of treatment onset. Using samples longitudinally collected from participants of the ENDEAR phase III clinical trial of nusinersen in infants with SMA type 1,⁹² baseline plasma pNF-H levels were higher in healthy infants (age <1 year) than in older healthy children (age 1-18 years), as well as in symptomatic SMA infants as compared to age-matched healthy infants.⁹³ This indicated early axonal degeneration consistent with the natural history of SMA. Higher baseline pNF-H correlated with earlier age of diagnosis and symptom onset and lower baseline motor function, and this inverse relationship persisted across age groups. Over time, pNF-H declined to baseline levels in both sham control and nusinersen-treated cohorts, though much more rapidly in the cohort of SMA patients receiving nusinersen therapy. pNF-H levels remained elevated at later timepoints in both cohorts of diseased individuals, regardless of treatment status. Thus, plasma pNF-H may inform and guide future clinical care from both prognostic and pharmacodynamic standpoints. In presymptomatic SMA infants treated with nusinersen as part of the single-arm phase II NURTURE trial, plasma pNF-H at baseline and day 64 correlated with future motor milestones and were the strongest predictors of future motor function achievement.⁹⁴ Two independent n-of-1 studies of non-ambulatory SMA type 1 infants with 2 SMN2 copies receiving nusinersen therapy further demonstrated neurofilament trends in CSF.^95,96 These studies recapitulate the rapid decline in NF levels observed following nusinersen administration, with additional evidence to correlate NF levels with clinical presentation and support the utility of NF proteins as prognostic and pharmacodynamic biomarkers for SMA.

Though NF proteins appear to be consistently informative biomarkers for infants with SMA, they have been less successful as markers for individuals with chronic forms of SMA. CSF NF protein levels are not elevated in later-onset adolescent and adult SMA patients, and this has been further demonstrated in mice.⁹⁷ In a study of 25 adolescent and adult patients with SMA type 2 (n = 9) or 3 (n = 16), no significant difference was identified between median CSF pNF-H or NF-L levels between SMA patients and controls, either at baseline or after 4 injections of nusinersen.⁹⁸ The same group then conducted a follow-up study to investigate whether this finding was because NF levels in CSF are less elevated than in blood. Again looking at SMA type 2 (n = 20) and 3 (n = 26) patients receiving nusinersen therapy, they compared NF-L levels in CSF and serum and found no correlation.⁹⁹ Additionally, after 14 months of treatment, serum NF-L levels remained nearly constant from baseline levels, demonstrating no change with treatment. Interestingly, higher serum NF-L levels were associated with worse motor performance, though changes in motor function did not correlate with changes in serum NF-L. A third study monitored NF levels in adults with SMA type 3, specifically at the time they were receiving loading doses of nusinersen.¹⁰⁰ Serum and CSF were collected from 11 adults prior to loading with nusinersen and analyzed for several markers of axonal injury, including pNF-H, tau protein, S100B protein, and neuron-specific enolase. SMN2 copy number ranged from 3 to 7 copies, and 9 of the 11 patients had retained the ability to walk at the start of the study. Three nusinersen loading doses were administered at 14-day intervals and then a fourth dose 30 days later. The study found that none of the markers were significantly elevated in CSF and there was no correlation with clinical presentation. In both healthy and SMA adults, an age-related increase in baseline CSF NF levels was observed, though neither plasma nor CSF NF levels changed in SMA adults in response to nusinersen treatment and levels were comparable to healthy controls.¹⁰¹ A recent publication also demonstrated that serum pNF-H levels in SMA type 2 and 3 patients are slightly reduced compared to healthy controls, which may reflect CNS motor pool exhaustion resulting from chronic disease.⁹⁷ Taken together, these studies suggest that NF proteins may not be informative biomarkers for monitoring disease progression or treatment response in adolescent or adult types of SMA, at any point throughout the disease course.

Due to the slower disease progression of SMA types 2 and 3 compared to type 1, in addition to the greater variability in clinical course—especially in type 3—subtle declines in motor function may elicit correspondingly subtle elevations in markers of axonal degeneration. Furthermore, due to the long disease duration prior to receiving treatment, considering that nusinersen did not gain FDA approval until 2016, preexisting loss of motor neurons in the adolescent and adult SMA patients studied as part of the studies outlined above may obscure the true utility of NF proteins as prognostic or pharmacodynamic biomarkers. For example, 1 study assessed changes in pNF-H and NF-L levels in the CSF of 12 adults with SMA type 3—who were on average 6 years younger and had disease durations nearly 10 years shorter than those patients examined in the above adult SMA studies—before and after nusinersen therapy and found significant declines in pNF-H and NF-L after just 6 months of nusinersen therapy.¹⁰² CSF samples were collected immediately prior to the first administration and again 6 months later on the last administration. Except for 1 patient with only 2 copies, all patients had either 3 or 4 copies of SMN2. Further, 9 patients were independently ambulant at initiation of the study. Baseline CSF levels of pNF-H and NF-L were comparable to controls and significantly below those of SMA type 1 patients previously reported. After 6 months of treatment, both pNF-H and NF-L levels significantly decreased while motor function, as assessed by the Hammersmith Functional Motor Scale Expanded (HFMSE) and 6MWT, only slightly improved. However, no correlation existed between NF levels and motor function scores. An important consideration is that many older SMA patients experience significant complications such as scoliosis and joint contractures, in addition to potential floor and ceiling effects, that limit the accuracy of motor function tests. Nonetheless, NF proteins may indicate early biochemical effects of nusinersen treatment on axonal degeneration that precede changes in motor performance. As patients begin to be treated at much earlier timepoints in their disease course given the current availability of approved therapies, there remains promise that NF proteins may provide insight into disease progression and treatment response for SMA patients of all types.

Spinal muscular atrophy multi-analyte panel (SMA-MAP) protein analytes

In 2012, an unbiased effort to identify novel candidate biomarkers through proteomic, metabolomic, and transcriptomic approaches resulted in the identification of protein analytes that were associated with motor function in SMA. The Biomarkers for SMA (BforSMA) study was a cross-sectional-omics study that evaluated blood and urine protein analytes in children (age 2-12 years) with genetically confirmed SMA and age-matched healthy controls.⁷⁴ Analytes were assessed for association to the Modified Hammersmith Functional Motor Scale (MHFMS), an instrument for measurement of disease severity.¹⁰³ A resulting 200 candidate biomarkers were found to correlate with MHFMS scores, and the most significant markers across all outcome measures were plasma protein analytes. When analyzed in conjunction with MHMFS scores, 3 analytes (dipeptidyl peptidase-IV, osteopontin, tetranectin) showed increased plasma levels and positive correlations with MHMFS, while 2 (fetuin-A, vitronectin) showed decreased plasma levels and negative correlations with MHMFS.^74,104 These parallel alterations suggest that changes in SMA phenotype may not be directly attributable to the disease itself or to changing SMN levels, but instead represent compensatory molecular changes that may affect other pathways indirectly influencing the disease course. Twenty-seven of the BforSMA protein analytes were ultimately included in a commercial multi-analyte panel (SMA-MAP) based on correlations with motor function and/or non-motor outcome measures, such as pulmonary function and quality of life.¹⁰⁴ These analytes are listed in Table 2.

Table 2.

SMA-MAP protein analytes.

Putative biomarker	Key function(s)	Change in SMA infants*	Change in SMA mice (response to therapy?)
Putative biomarker	Key function(s)	Kolb (2016)	Arnold (2016)	Strathmann (2018)
Apolipoprotein B	Metabolic processes, cholesterol homeostasis	No	—	—
AXL receptor tyrosine kinase	Cellular functions (growth, migration, aggregation, anti-inflammation)	Decreased	No (No)	—
C-reactive protein	Complement activation, acute phase response	No	—	—
Cadherin-13	Signal transduction, regulation of axonal growth	Decreased	—	—
Cartilage oligomeric matrix protein	Extracellular matrix organization, cartilage development, apoptosis	Decreased	Decreased (No)	Decreased (No)
Cathepsin D	Proteolysis	No	—	—
Chitinase-3-like protein 1	Inflammation, tissue remodeling	Increased	No (No)	—
Complement component C1q receptor	Cell-cell adhesion, phagocytosis	Decreased	—	—
Complement factor H-related protein 1	Complement activation	No	—	—
Dipeptidyl peptidase IV	Glucose and insulin metabolism, immune regulation	Decreased	Increased (Yes)	Decreased (No)
Endoglin	Vasculogenesis, cellular functions (adhesion, migration)	Decreased	—	—
Fetuin-A	Vascular calcification (inhibitor), anti-inflammation	No	Decreased (Yes)	No (No)
Fibulin-1C	Extracellular matrix organization, platelet adhesion	No	—	—
Human epidermal growth factor receptor 2	Cell proliferation	Decreased	—	—
Insulin-like growth factor binding protein 6	Signal transduction, metabolic processes	Decreased	—	—
Leptin	Energy homeostasis, endocrine functions	No	—	—
Monocyte chemotactic protein 1	Inflammatory response, monocyte and macrophage migration	No	—	—
Myoglobin	Oxygen transport	No	—	—
Osteopontin	Biomineral tissue development, inflammatory response	Decreased	Increased (Yes)	Decreased (No)
Peptidase D	Proteolysis, amino acid metabolism	Decreased	—	—
Placenta growth factor	Angiogenesis, cell proliferation	No	—	—
Serum amyloid P-component	Complement activation, chaperone-mediated protein complex assembly	No	—	—
Tenascin-X	Extracellular matrix maturation, collagen metabolism	Decreased	—	—
Tetranectin	Bone mineralization, ossification	Decreased	Increased (Yes)	No (No)
Thrombospondin-4	Cell-cell interactions, signal transduction, behavioral response to pain	Decreased	—	—

Excluding SMN2 > 2.

A subset of the protein analytes were studied in an SMNΔ7 mouse model, where a few were shown to be responsive to both SMA phenotype and postnatal SMN restoration. Five analytes (dipeptidyl peptidase-IV, fetuin-A, osteopontin, tetranectin, vitronectin) were significantly altered in SMA mice compared to wildtype controls, and levels normalized following morpholino ASO treatment and subsequent SMN restoration.⁷⁹ Four of these 5 analytes—excluding fetuin-A—also correlated with SMN levels in brain, spinal cord, liver, and quadriceps muscle. Further, vitronectin correlated with compound muscle action potential (CMAP) amplitude, a well-established measure of motor function. Thus, there is a pre-clinical suggestion that these analytes may serve as biomarkers of treatment response. At P90, well after ASO efficacy has diminished,⁸⁰ tetranectin and vitronectin still showed a significant change in plasma level directionally consistent with earlier time points and with prior studies mentioned here. Thus, these 2 protein analytes may be suitable candidates for pharmacodynamic biomarkers, monitoring long-term response to therapy and assessing both need for and timing of retreatment.

The NeuroNEXT SMA infant biomarker study identified a handful of protein analytes that were altered in the serum of SMA infants (age <6 months)—a much younger population than in the original BforSMA study—compared to healthy infants of the same age.³¹ In SMA infants, 5 analytes showed reduced levels (cadherin-13, cartilage oligomeric matrix protein, insulin-like growth factor binding protein 6, peptidase D, tetranectin) while 2 showed increased levels (myoglobin, chitinase-3-like protein 1). These analytes distinguished between healthy and SMA infants at both baseline and later time points. Six analytes showed a negative correlation with age in both cohorts (AXL receptor tyrosine kinase, cartilage oligomeric matrix protein, complement component C1q receptor, fibulin-1C, tenascin-X, thrombospondin-4). Serum concentrations of many analytes were reduced with age in both healthy and SMA infants, suggesting a general trend of declining analyte concentration with increasing age independent of disease status, with 2 notable exceptions being apolipoprotein B and serum amyloid P-component in SMA infants. While no general conclusions can be drawn about the SMA-MAP protein analytes as a whole, the ability of specific analytes to reliably distinguish between SMA and control cohorts supports their utility as potential prognostic markers in infantile SMA.

Creatine kinase (CK) and creatinine (Crn)

The creatine kinase (CK) system is essential for maintenance of energy homeostasis and skeletal muscle function.^105,106 CK is an enzyme that catalyzes the reversible transfer of phosphate to creatine, generating phosphocreatine, which acts as a rapidly mobilizable energy reserve primarily in skeletal muscle. Elevated CK activity has been observed in SMA, especially in patients with milder, chronic forms of the disease, likely due to greater preserved muscle mass or degree of activity.¹⁰⁷ Creatinine (Crn) is a metabolic waste product of the CK system and a marker of muscle mass, previously shown to correlate with disease severity in other denervating motor neuron diseases such as spinal and bulbar muscular atrophy (SBMA).¹⁰⁸

Serum Crn inversely correlates with disease severity in children and adolescents, when corrected for age and lean mass. In a study of 238 patients with SMA, Crn was highest in type 3 patients and lowest in type 1 patients, and levels decreased with age regardless of clinical subtype.¹⁰⁹ This trend persisted when patients were grouped based on SMN2 copy number, motor function (categorized as walkers, sitters, or nonsitters based on HFMS performance), and severity of denervation measured through maximum CMAP amplitude and MUNE. Pre-symptomatic infants (n = 22) were further assessed for changes in serum Crn throughout the first year of life. Both CMAP and Crn declined over a period of approximately 3 months, with Crn changes slightly preceding CMAP changes in the sub-population of infants with 3 SMN2 copies. At this early time period, when compensatory reinnervation may be aiding in stabilization of CMAP, serum Crn may present a more sensitive marker of early denervation and disease progression in this specific population.

In adults, serum CK and Crn correlate with motor function and disease severity status, and baseline levels can distinguish nusinersen treatment responders from non-responders. In a study of treatment-naïve adults with SMA types 2 (n = 70) and 3 (n = 136), CK was elevated in SMA type 3 and ambulatory patients, as well as in patients with SMN2 copy number ⩾4.¹¹⁰ In patients treated with nusinersen (n = 68-85), improvements in motor function were observed over a period of 18 months, in conjunction with a decline in CK and a slight increase in Crn. Baseline measures of CK and Crn were much higher in treatment responders (⩾1 increase on HFMSE score) compared to non-responders (⩾1 decrease on HFMSE score). CK declined more significantly in responders, while Crn increased more notably in non-responders. An additional study found that urine Crn can distinguish healthy (n = 3) from pre-treated SMA (n = 24) cohorts, but that levels are not significantly altered in response to nusinersen therapy.¹¹¹ Taken together, these studies suggest that serum CK and serum Crn may be useful predictive and pharmacodynamic biomarkers for SMA.

Electrophysiology and Imaging Biomarkers

Compound muscle action potential (CMAP) and motor unit number estimation (MUNE)

Electrophysiology has long been used as a method for the quantification of neuromuscular function in the setting of motor neuron disorders.^112-114 It allows for in vivo assessment of motor unit connectivity and can provide insight into disease progression and the relationship between muscle connectivity and function.^115-117 CMAP represents the total electrical output of the motor units innervating a single muscle or group of muscles following supramaximal nerve stimulation. The output of a single motor unit, the single motor unit potential (SMUP), is instead the average of 10 incremental, submaximal nerve stimulations. Motor unit number estimate (MUNE), a quantification of the number of motor units innervating a muscle, can then be easily calculated as ratio of the output of all motor units to that of a single motor unit, equivalent to CMAP divided by SMUP.

Electrophysiology assessments of neuromuscular function during the early postnatal period demonstrated that both CMAP and MUNE were reduced in SMA mice by postnatal week 2, and that these responses could be corrected by day 30 via treatment with morpholino ASO that restored SMN levels.¹¹⁸ Electrophysiology measurements were performed on the right hindlimb, specifically the triceps surae muscle, of SMNΔ7 mice under anesthesia. Consistent with these findings, a follow-up study verified that CMAP was significantly reduced at day 12 in SMA mice.⁷⁹

In humans, CMAP and MUNE are also decreased in SMA and correlate with SMN2 copy number, SMA subtype, and patient age. A study of SMA type 1, 2, and 3 patients (age 1 month-50 years) showed that lower copy number correlates with lower CMAP values and worse functional outcomes.¹¹⁴ This finding was replicated in an additional study of 12 SMA type 1 infants, identified through NBS, where CMAP was significantly lower in infants with 2 copies of SMN2 than 3 copies.¹¹⁹ Further, CMAP declines in an age-dependent manner. In infants, CMAP positively correlates with motor function, as assessed by the Test of Infant Motor Performance Screening Items (TIMPSI)¹⁸ and The Children’s Hospital of Philadelphia Infant Test of Neuromuscular Disorders (CHOP-INTEND),^120,121 in SMA infants but not in healthy infants.^31,32 However, this correlation was not replicated when analyzing only the subgroup of SMA infants with 2 copies of SMN2, which typically corresponds to SMA type 1. CMAP was later assessed in 12 SMA type 1 infants receiving AAV9-mediated gene replacement therapy with onasemnogene abeparvovec and was shown to improve with treatment after just a few months.¹²² Treated infants were compared to untreated SMA type 1 infants from the infantile SMA natural history study. Both cohorts of infants had similar baseline CMAP values. SMA infants receiving onasemnogene abeparvovec treatment demonstrated increases in CMAP peak area and peak amplitude values, alongside improved motor function and motor milestone achievement. In treated SMA infants, CMAP values showed the largest increases from baseline (6 months) to 12 months of age—increasing from 1.1 to 2.8 mV—and continued to increase or stabilized through 24 months. The motor function decline seen in untreated SMA type 1 infants was halted, and even reversed to an extent, though not to normal, healthy infant capacity. In a small study of infants (n = 12), changes in CMAP of the ulnar nerve were monitored prior to initiation of nusinersen therapy and periodically throughout treatment.¹¹⁹ Interestingly, a pre-treatment CMAP amplitude of 2 mV appeared to delineate infants who achieved normal motor milestones (>2 mV) from those who still showed functional improvements, yet at a slower rate (<2 mV). Those that had more rapid increases in CMAP post-treatment tended to have better functional outcomes. Thus, CMAP may be a sensitive method for detection of disease onset and transition from presymptomatic to symptomatic, and may help inform optimal timing for treatment.

In a prospective study of 20 symptomatic SMA children (ages 4 months-16 years) undergoing nusinersen treatment, individuals with shorter disease durations showed greater increases in CMAP and MUNE, and these measures correlated with functional changes assessed by CHOP-INTEND and HFMSE.¹²³ In children with later-onset SMA (ages 2-15 years), CMAP and MUNE fluctuated slightly but were generally stable over 3 years of nusinersen treatment.¹²⁴ SMA type 2 children (n = 11, average age 15.4 months at diagnosis) showed progressive improvements in motor function and the majority demonstrated clinically meaningful improvements by the end of the study, while SMA type 3 children (n = 17, average age 43.6 months at diagnosis) showed only modest, and few clinically meaningful, functional improvements. Additionally, MUNE progressively decreased in type 3 children, despite therapeutic intervention, indicating that age and severity of denervation at the time of treatment initiation may significantly influence treatment response. These studies suggest that electrophysiological measures may be markers of treatment response, predicting which individuals will be functionally stable and which will achieve significant gains in motor function following treatment, though limitations regarding disease duration must be considered.

As techniques and interrater reliability continue to improve, the studies outlined above instill confidence in electrophysiology techniques such as CMAP and MUNE as reliable markers of motor unit loss accompanying disease progression in SMA.

Motor unit number index (MUNIX) and motor unit size index (MUSIX)

MUNIX is a non-invasive electrophysiology technique that utilizes surface electromyography to estimate the number of functional motor units within a muscle. Requiring minimal nerve stimulation compared to other current MUNE methods, MUNIX is able to quantify and monitor motor unit loss longitudinally with appropriate reliability and validity.^125-128 MUSIX provides information regarding the size of individual motor units and is computed as CMAP amplitude divided by MUNIX for a given muscle.¹²⁹ MUNIX and MUSIX have recently been used as outcome measures for patients with SMA to monitor functional changes in hand muscles.

MUNIX of the hand muscles strongly correlated with disease score independent of stage and could provide insight into overall disability and disease progression in adults with SMA types 3 and 4.¹³⁰ MUNIX correlated with both muscle strength and disability in adult SMA patients and was able to detect motor neuron loss with greater sensitivity than CMAP.¹³¹ Only severely affected muscles showed reduced CMAP, whereas MUNIX was reduced in all hand muscles compared to controls independent of symptom severity, indicating its utility in detecting motor neuron loss at varying disease stages. Further, MUSIX was increased in patients with SMA, which was attributed to active re-innervation and collateral axonal sprouting similar to what has been observed in ALS.^132-135 Interestingly, motor unit loss in hand muscles was not evenly distributed, resulting in a specific hand muscle wasting pattern in SMA patients (“reverse split hand,” preserved thenar index)—highlighting a pattern of motor unit loss distinct from that of ALS (“split hand,” thenar weakness with hypothenar sparing). A preserved thenar index was able to discriminate SMA from ALS and controls with high sensitivity and specificity. MUNIX was able to be recorded even at very advanced stages of the disease and thus may present a new method by which to monitor disease progression in SMA patients of any age and throughout the course of disease.

Electrical impedance myography (EIM)

Electrical impedance myography (EIM) is a bioimpedance-based non-invasive technique that allows for voltage measurement of a localized muscle region following the application of a high-frequency electrical current.¹³⁶ This metric has been used in the context of motor neuron disease to reflect changes in the overall composition and health of muscles. EIM has been used in neurogenic disorders with marked muscle atrophy such as ALS due to easy assessment of multiple limbs and reliable tracking of disease progression in both mice and humans.^137,138 EIM also correlates with electrophysiologic measures including CMAP and MUNE in mouse models of ALS.^139,140 Further, EIM has widely been applied in myogenic disorders with relatively limited muscle atrophy, in particular in Duchenne muscular dystrophy (DMD). EIM discriminates between DMD and healthy male children, associates with disease progression, and is sensitive to treatment.^141-143 EIM is especially useful in pediatric settings given its ease of application and simplicity compared to standard electrophysiological measures.

Specific EIM metrics have been shown to distinguish between SMA and healthy cohorts in both pre-clinical and clinical settings. EIM 50 kHz phase was reduced in Smn1^c/c mice, a mild animal model of SMA, but not in wildtype mice.¹⁴⁴ EIM correlated with motor neuron counts in the spinal cord as well as SMN levels in mice at 1 year, in both SMA and control animals. Therefore, EIM may be a sensitive measure of muscle status in mild forms of SMA. In the more severe SMNΔ7 mouse model, EIM detected significant differences between SMA controls and SMA mice treated with ASOs to increase SMN protein for late (symptomatic) but not early (pre-symptomatic) treatment groups.¹⁴⁵ EIM measures also correlated with CMAP and MUNE in the late treatment group, reflecting a reduction in the number of motor units and a delay in the normalization of muscle potential underlying symptomatic ASO treatment. The ability of EIM to discriminate treatment groups, and thus the effect of ASO interventional timing on resulting motor neuron function, makes EIM a promising biomarker of treatment stratification and effect for SMA patients undergoing treatment.

Healthy children demonstrate increased EIM measures, an indication of muscle growth and development, compared to children with SMA type 2 and type 3 who display relatively stable measures over time.¹⁴⁶ However, SMA infants, the youngest cohort to be studied at the time of publication, have also shown increasing EIM measures over time.³¹ Of the EIM outcome measures, only high-frequency reactance slope was able to discriminate between SMA infants and controls, contrary to previous studies in older children.^146,147 This may have been due to differing impedance spectral characteristics between infants and older children that were not clarified until after this study was conducted.¹⁴⁸ In general, EIM positively correlated with age for both cohorts. There were no correlations between EIM and motor function in the SMA cohort, regardless of copy number. More studies in humans are necessary to determine the utility of EIM as an SMA biomarker, though its clinical feasibility and historical accuracy in tracking disease progression in other motor neuron diseases imbue optimism.

Overall, electrophysiologic measures have proven to be reliable markers of SMA disease progression, especially when multiple measures are used in conjunction and when evaluations assess multiple regions of motor neuron output. Techniques are becoming increasingly precise at detecting neuromuscular changes even before symptoms arise. The ability to quantify these changes at such early time points will guide clinical therapies and allow for pharmacologic interventions earlier in the disease course.

Magnetic resonance imaging and muscle ultrasonography

Quantitative magnetic resonance imaging (qMRI) has been identified in previous studies as a non-invasive technique yielding more sensitive and less biased measures of muscle physiology than conventional functional scales in the context of neuromuscular disease.^149,150 A year-long observational study of 18 ambulatory patients with SMA type 3 (mean SMN2 copy number of 3.8) concluded that qMRI, while reliable, was not predictive of SMA disease progression.⁷³ SMA patients showed a reduction in thigh muscle mass and density, as well as an increase in the fat to muscle ratio, when compared to healthy patients. There was no difference in biomarkers between SMA and healthy patients, and qMRI measures did not correlate with molecular markers measured in blood (SMN1 and SMN2 copy numbers, SMN1 and SMN2 expression, SMND7 expression, and SMN protein). While there was no observed change in motor function for either cohort, qMRI measures negatively correlated with MFS at all time points, suggesting that functional changes as measured by MFS may be directly reflective of underlying intramuscular structural changes assessed through non-invasive imaging methods. An additional study of 25 ambulatory SMA type 3b patients demonstrated selective involvement of proximal upper (triceps brachii) and lower (gluteus maximus) extremity muscles on MRI, a pattern that appears specific for SMA.¹⁵¹ Longer disease durations correlated with worse MRI scores, evaluated using a qualitative rating scale, in proximal muscles.

Muscle ultrasonography is a non-invasive imaging technique that can be easily performed at the patient bedside and used to quickly survey multiple muscles. It has been applied as a diagnostic tool in various neuromuscular disorders including Duchenne muscular dystrophy, ALS, polyneuropathy, and specific myopathies.¹⁵² In SMA, quantitative ultrasonography of the biceps brachii, wrist extensors, quadriceps, and tibialis anterior muscles was able to discriminate between healthy (n = 21) and SMA (n = 25) individuals, and more specifically between SMA type 2 and type 3 subtypes, based on muscle echogenicity.¹⁵³ The ratio of muscle luminosity to subcutaneous fat luminosity was highest in type 2 SMA, intermediate in type 3 SMA, and lowest in healthy individuals. However, SMA patients demonstrated increased skin-subcutaneous fat thickness compared to healthy patients and fat thickness varied with disease severity, which presented a potential confounding factor in the study. Further limitations to this technique include considerable variation in ultrasonic properties between machines, additional confounding factors such as age and sex of individuals, and inter- and intra-rater reliability.¹⁵⁴ A small feasibility study of 3 SMA individuals (1 of each subtype) and 3 age-matched controls reported decreased median nerve fascicle number and density in the SMA type 1 patient on ultra-high-frequency (UHF) ultrasound imaging.¹⁵⁵ No differences were observed between SMA type 2, SMA type 3, and controls, indicating that UHF ultrasonography may be most applicable in severe cases of SMA where changes in individual fascicle number and density may be most prominent. Given the extremely small sample size, however, future studies are warranted before any definitive conclusions can be drawn.

Overall, imaging measures have been shown to reliably differentiate between SMA and healthy patients but are ineffective at monitoring disease progression over time. Thus, MRI and ultrasonography currently lack utility as independent measures of SMA phenotype and disease course, but show promise as complementary measures to conventional functional scales with the ability to elucidate anatomo-functional relationships and contribute to future knowledge of clinical endpoints.

Conclusion

Efforts to identify biomarkers for detecting and monitoring SMA disease course have been thorough, though more work remains to be done. Genetic confirmation of SMA and SMN2 copy number remain the most accurate measures of detection, with copy number also being the strongest prognostic biomarker in untreated SMA individuals. SMN mRNA and protein levels provide insight into current disease state, but do not appear to change over time and are inconsistent between peripheral (eg, blood) and central (eg, CSF) biological samples. However, SMN levels in peripheral blood are useful to monitor the systemic response to CNS-targeting therapies that increase SMN. NF proteins have recently emerged as leading markers of SMA prognosis, disease progression, and treatment response, especially in SMA infants undergoing treatment. NF levels rapidly decline following treatment and correlate with clinical presentation in this population. Conversely, NFs are not informative biomarkers in adolescent or adult SMA populations. Certain plasma protein analytes can distinguish between SMA and control cohorts across time, are altered in SMA mice and infants, and respond to therapy in SMA mice, thus rendering them possible pharmacodynamic biomarkers. In pre-symptomatic infants with 3 copies of SMN2, serum Crn may be a sensitive marker of early denervation and disease progression, though its utility is restricted outside of this specific SMA subpopulation. In adults, serm CK and Crn reliably predict which individuals will respond to nusinersen treatment and thus may be meaningful predictive and/or pharmacodynamic biomarkers. Electrophysiology techniques such as CMAP, MUNE, and EIM reliably track motor function throughout disease course in both healthy and SMA individuals of all ages and have strong predictive ability regarding treatment effect. Electrophysiology measures may also be able to detect neuromuscular changes before overt clinical symptoms arise, allowing for earlier therapeutic intervention and guidance of clinical therapies. Imaging modalities can discriminate between SMA and healthy individuals and may reflect anatomo-functional changes, but are limited in widespread application due to current lack of equipment and rater reliability.

No single biomarker may necessarily be sufficient to monitor disease progression and treatment efficacy, but there is great potential in a combination of robust biomarkers that together allow for more accurate clinical assessment. Additional progress toward discovering and characterizing SMA biomarkers will be made as technology, natural history data, and an understanding of processes occurring at the motor neuron improve. SMA patients will continue to be treated and monitored in clinical settings. There is still much work to be done to understand variability in therapeutic efficacy and ultimately optimize treatment response. This will require a collaborative effort between clinicians and laboratory scientists to study humans as well as animals and cell models in parallel.

Footnotes

Funding:

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests:

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: SJK has received compensation for consulting from Biogen, Roche, and Novartis.

Author Contributions

MGP performed the literature search, designed the tables, and wrote the manuscript under the supervision of SJK. All authors contributed to manuscript revision, read, and approved the submitted version.

ORCID iDs

Megan G Pino

Kelly A Rich

Stephen J Kolb

References

Pearn

. The gene frequency of acute Werdnig-Hoffmann disease (SMA type 1). A total population survey in North-East England. J Med Genet. 1973;10: 260-265.

Sugarman

Nagan

Zhu

, et al. Pan-ethnic carrier screening and prenatal diagnosis for spinal muscular atrophy: clinical laboratory analysis of >72,400 specimens. Eur J Hum Genet. 2012;20: 27-32.

Lefebvre

Burlet

Liu

, et al. Correlation between severity and SMN protein level in spinal muscular atrophy. Nat Genet. 1997;16: 265-269.

McAndrew

Parsons

Simard

, et al. Identification of proximal spinal muscular atrophy carriers and patients by analysis of SMNT and SMNC gene copy number. Am J Hum Genet. 1997;60: 1411-1422.

Monani

Sendtner

Coovert

, et al. The human centromeric survival motor neuron gene (SMN2) rescues embryonic lethality in Smn(-/-) mice and results in a mouse with spinal muscular atrophy. Hum Mol Genet. 2000;9: 333-339.

Feldkötter

Schwarzer

Wirth

Wienker

Wirth

. Quantitative analyses of SMN1 and SMN2 based on real-time lightCycler PCR: fast and highly reliable carrier testing and prediction of severity of spinal muscular atrophy. Am J Hum Genet. 2002;70: 358-368.

Elsheikh

Prior

Zhang

, et al. An analysis of disease severity based on SMN2 copy number in adults with spinal muscular atrophy. Muscle Nerve. 2009;40: 652-656.

Gennarelli

Lucarelli

Capon

, et al. Survival motor neuron gene transcript analysis in muscles from spinal muscular atrophy patients. Biochem Biophys Res Commun. 1995;213: 342-348.

Lorson

Hahnen

Androphy

Wirth

. A single nucleotide in the SMN gene regulates splicing and is responsible for spinal muscular atrophy. Proc Natl Acad Sci USA. 1999;96: 6307-6311.

10.

Monani

Lorson

Parsons

, et al. A single nucleotide difference that alters splicing patterns distinguishes the SMA gene SMN1 from the copy gene SMN2. Hum Mol Genet. 1999;8: 1177-1183.

11.

Cartegni

Krainer

. Disruption of an SF2/ASF-dependent exonic splicing enhancer in SMN2 causes spinal muscular atrophy in the absence of SMN1. Nat Genet. 2002;30: 377-384.

12.

Kashima

Manley

. A negative element in SMN2 exon 7 inhibits splicing in spinal muscular atrophy. Nat Genet. 2003;34: 460-463.

13.

Prior

Krainer

Hua

, et al. A positive modifier of spinal muscular atrophy in the SMN2 gene. Am J Hum Genet. 2009;85: 408-413.

14.

Lorson

Strasswimmer

Yao

, et al. SMN oligomerization defect correlates with spinal muscular atrophy severity. Nat Genet. 1998;19: 63-66.

15.

Lorson

Androphy

. An exonic enhancer is required for inclusion of an essential exon in the SMA-determining gene SMN. Hum Mol Genet. 2000;9: 259-265.

16.

Burnett

Muñoz

Tandon

Kwon

Sumner

Fischbeck

. Regulation of SMN protein stability. Mol Cell Biol. 2009;29: 1107-1115.

17.

Bernal

Alias

Barcelo

, et al. The c.859G>C variant in the SMN2 gene is associated with types II and III SMA and originates from a common ancestor. J Med Genet. 2010;47: 640-642.

18.

Krosschell

Maczulski

Scott

, et al.; Project Cure SMA Investigators Network. Reliability and validity of the TIMPSI for infants with spinal muscular atrophy type I. Pediatr Phys Ther. 2013;25: 140-148. discussion 149.

19.

Ratni

Ebeling

Baird

, et al. Discovery of Risdiplam, a selective survival of motor neuron-2 (SMN2) gene splicing modifier for the treatment of Spinal muscular Atrophy (SMA). J Med Chem. 2018;61: 6501-6517.

20.

Sturm

Günther

Jaber

, et al. A phase 1 healthy male volunteer single escalating dose study of the pharmacokinetics and pharmacodynamics of risdiplam (RG7916, RO7034067), a SMN2 splicing modifier. Br J Clin Pharmacol. 2019;85: 181-193.

21.

Wurster

Ludolph

. Nusinersen for spinal muscular atrophy. Ther Adv Neurol Disord. 2018;11: 1756285618754459.

22.

Lee

Collins

Lewis

, et al. Combination therapy with nusinersen and AVXS-101 in SMA type 1. Neurology. 2019;93: 640-641.

23.

Dabbous

Maru

Jansen

, et al. Survival, motor function, and motor milestones: comparison of AVXS-101 relative to nusinersen for the treatment of infants with spinal muscular atrophy type 1. Adv Ther. 2019;36: 1164-1176.

24.

Lefebvre

Bürglen

Reboullet

, et al. Identification and characterization of a spinal muscular atrophy-determining gene. Cell. 1995;80: 155-165.

25.

Bürglen

Lefebvre

Clermont

, et al. Structure and organization of the human survival motor neurone (SMN) gene. Genomics. 1996;32: 479-482.

26.

Schrank

Götz

Gunnersen

, et al. Inactivation of the survival motor neuron gene, a candidate gene for human spinal muscular atrophy, leads to massive cell death in early mouse embryos. Proc Natl Acad Sci USA. 1997;94: 9920-9925.

27.

Hahnen

Schönling

Rudnik-Schöneborn

Zerres

Wirth

. Hybrid survival motor neuron genes in patients with autosomal recessive spinal muscular atrophy: new insights into molecular mechanisms responsible for the disease. Am J Hum Genet. 1996;59: 1057-1065.

28.

Campbell

Potter

Ignatius

Dubowitz

Davies

. Genomic variation and gene conversion in spinal muscular atrophy: implications for disease process and clinical phenotype. Am J Hum Genet. 1997;61: 40-50.

29.

Wirth

Herz

Wetter

, et al. Quantitative analysis of survival motor neuron copies: identification of subtle SMN1 mutations in patients with spinal muscular atrophy, genotype-phenotype correlation, and implications for genetic counseling. Am J Hum Genet. 1999;64: 1340-1356.

30.

Mailman

Heinz

Papp

, et al. Molecular analysis of spinal muscular atrophy and modification of the phenotype by SMN2. Genet Med. 2002;4: 20-26.

31.

Kolb

Coffey

Yankey

, et al.; NeuroNEXT Clinical Trial Network and on behalf of the NN101 SMA Biomarker Investigators. Baseline results of the NeuroNEXT spinal muscular atrophy infant biomarker study. Ann Clin Transl Neurol. 2016;3: 132-145.

32.

Kolb

Coffey

Yankey

, et al.; the NeuroNEXT Clinical Trial Network on behalf of the NN101 SMA Biomarker Investigators. Natural history of infantile-onset spinal muscular atrophy. Ann Neurol. 2017;82: 883-891.

33.

Vezain

Saugier-Veber

Goina

, et al. A rare SMN2 variant in a previously unrecognized composite splicing regulatory element induces exon 7 inclusion and reduces the clinical severity of spinal muscular atrophy. Hum Mutat. 2010;31: E1110-E1125.

34.

Bai

Cao

, et al. A rare variant (c.863G>T) in exon 7 of SMN1 disrupts mRNA splicing and is responsible for spinal muscular atrophy. Eur J Hum Genet. 2016;24: 864-870.

35.

Stenson

Mort

Ball

Shaw

Phillips

Cooper

. The human gene mutation database: building a comprehensive mutation repository for clinical and molecular genetics, diagnostic testing and personalized genomic medicine. Hum Genet. 2014;133: 1-9.

36.

Y-L

Cheng

N-Y

, et al.835-5T>G variant in SMN1 gene causes transcript exclusion of exon 7 and spinal muscular atrophy. J Mol Neurosci. 2018;65: 196-202.

37.

Parsons

McAndrew

Iannaccone

Mendell

Burghes

Prior

. Intragenic telSMN mutations: frequency, distribution, evidence of a founder effect, and modification of the spinal muscular atrophy phenotype by cenSMN copy number. Am J Hum Genet. 1998;63: 1712-1723.

38.

Stratigopoulos

Lanzano

Deng

, et al. Association of plastin 3 expression with disease severity in spinal muscular atrophy only in postpubertal females. Arch Neurol. 2010;67: 1252-1256.

39.

Yanyan

Yujin

Jinli

Yuwei

Hong

Fang

. Correlation of PLS3 expression with disease severity in children with spinal muscular atrophy. J Hum Genet. 2014;59: 24-27.

40.

McGovern

Massoni-Laporte

Wang

, et al. Plastin 3 expression does not modify spinal muscular atrophy severity in the ∆7 SMA mouse. PLoS One. 2015;10: e0132364.

41.

Hosseinibarkooie

Peters

Torres-Benito

, et al. The power of human protective modifiers: PLS3 and CORO1C unravel impaired endocytosis in spinal muscular atrophy and rescue SMA phenotype. Am J Hum Genet. 2016;99: 647-665.

42.

Riessland

Kaczmarek

Schneider

, et al. Neurocalcin delta suppression protects against spinal muscular atrophy in humans and across species by restoring impaired endocytosis. Am J Hum Genet. 2017;100: 297-315.

43.

Catapano

Zaharieva

Scoto

, et al. Altered levels of MicroRNA-9, -206, and -132 in spinal muscular atrophy and their response to antisense Oligonucleotide therapy. Mol Ther Nucleic Acids. 2016;5: e331.

44.

Chen

T-H

Chen

J-A

. Multifaceted roles of microRNAs: from motor neuron generation in embryos to degeneration in spinal muscular atrophy. eLife. 2019;8: e50848. doi:10.7554/elife.50848

45.

Chen

T-H

. Circulating microRNAs as potential biomarkers and therapeutic targets in spinal muscular atrophy. Ther Adv Neurol Disord. 2020;13: 1756286420979954.

46.

Fenoglio

Ridolfi

Galimberti

Scarpini

. An emerging role for long non-coding RNA dysregulation in neurological disorders. Int J Mol Sci. 2013;14: 20427-20442.

47.

d’Ydewalle

Ramos

Pyles

, et al. The antisense transcript SMN-AS1 regulates SMN expression and Is a novel therapeutic target for spinal muscular atrophy. Neuron. 2017;93: 66-79.

48.

Hauke

Riessland

Lunke

, et al. Survival motor neuron gene 2 silencing by DNA methylation correlates with spinal muscular atrophy disease severity and can be bypassed by histone deacetylase inhibition. Hum Mol Genet. 2009;18: 304-317.

49.

Cao

, et al. Association between SMN2 methylation and disease severity in Chinese children with spinal muscular atrophy. J Zhejiang Univ Sci B. 2016;17: 76-82.

50.

Maretina

Egorova

Baranov

Kiselev

. DYNC1H1 gene methylation correlates with severity of spinal muscular atrophy. Ann Hum Genet. 2019;83: 73-81.

51.

Zheleznyakova

Voisin

Kiselev

, et al. Genome-wide analysis shows association of epigenetic changes in regulators of Rab and Rho GTPases with spinal muscular atrophy severity. Eur J Hum Genet. 2013;21: 988-993.

52.

Phan

Taylor

Hannon

Howell

. Newborn screening for spinal muscular atrophy: anticipating an imminent need. Semin Perinatol. 2015;39: 217-229.

53.

Glascock

Sampson

Haidet-Phillips

, et al. Treatment algorithm for infants diagnosed with spinal muscular atrophy through Newborn screening. J Neuromuscul Dis. 2018;5: 145-158.

54.

Bella

Cisterni

Salaün

Pettmann

. Survival motor neuron (SMN) protein in rat is expressed as different molecular forms and is developmentally regulated. Eur J Neurosci. 1998;10: 2913-2923.

55.

Battaglia

Princivalle

Forti

Lizier

Zeviani

. Expression of the SMN gene, the spinal muscular atrophy determining gene, in the mammalian central nervous system. Hum Mol Genet. 1997;6: 1961-1971.

56.

Groen

EJN

Perenthaler

Courtney

, et al. Temporal and tissue-specific variability of SMN protein levels in mouse models of spinal muscular atrophy. Hum Mol Genet. 2018;27: 2851-2862.

57.

Ramos

d’Ydewalle

Gabbeta

, et al. Age-dependent SMN expression in disease-relevant tissue and implications for SMA treatment. J Clin Investig. 2019;129: 4817-4831.

58.

Zaworski

von Herrmann

Taylor

, et al. SMN protein Can Be reliably measured in whole blood with an Electrochemiluminescence (ECL) immunoassay: implications for clinical trials. PLoS One. 2016;11: e0150640.

59.

Burlet

Huber

Bertrandy

, et al. The distribution of SMN protein complex in human fetal tissues and its alteration in spinal muscular atrophy. Hum Mol Genet. 1998;7: 1927-1933.

60.

Giavazzi

Setola

Simonati

Battaglia

. Neuronal-specific roles of the survival motor neuron protein: evidence from survival motor neuron expression patterns in the developing human central nervous system. J Neuropathol Exp Neurol. 2006;65: 267-277.

61.

Poirier

Weetall

Heinig

, et al. Risdiplam distributes and increases SMN protein in both the central nervous system and peripheral organs. Pharmacol Res Perspect. 2018;6: e00447.

62.

Alves

CRR

Zhang

Johnstone

, et al. Whole blood survival motor neuron protein levels correlate with severity of denervation in spinal muscular atrophy. Muscle Nerve. 2020;62: 351-357.

63.

Nash

McFall

Perozzo

, et al. Survival motor neuron protein is released from cells in exosomes: a potential biomarker for spinal muscular atrophy. Sci Rep. 2017;7: 13859.

64.

Kernochan

Russo

Woodling

, et al. The role of histone acetylation in SMN gene expression. Hum Mol Genet. 2005;14: 1171-1182.

65.

Mohseni

Zabidi-Hussin

Sasongko

. Histone deacetylase inhibitors as potential treatment for spinal muscular atrophy. Genet Mol Biol. 2013;36: 299-307.

66.

Swoboda

Scott

Reyna

, et al. Phase II open label study of valproic acid in spinal muscular atrophy. PLoS One. 2009;4: e5268.

67.

Brichta

Hofmann

Hahnen

, et al. Valproic acid increases the SMN2 protein level: a well-known drug as a potential therapy for spinal muscular atrophy. Hum Mol Genet. 2003;12: 2481-2489.

68.

Tsai

L-K

Tsai

M-S

Ting

. Multiple therapeutic effects of valproic acid in spinal muscular atrophy model mice. J Mol Med. 2008;86: 1243-1254.

69.

Piepers

Cobben

J-M

Sodaar

, et al. Quantification of SMN protein in leucocytes from spinal muscular atrophy patients: effects of treatment with valproic acid. J Neurol Neurosurg Psychiatry. 2011;82: 850-852.

70.

Harahap

ISK

Saito

San

, et al. Valproic acid increases SMN2 expression and modulates SF2/ASF and hnRNPA1 expression in SMA fibroblast cell lines. Brain Dev. 2012;34: 213-222.

71.

Renusch

Harshman

, et al. Spinal muscular atrophy biomarker measurements from blood samples in a clinical trial of Valproic acid in ambulatory adults. J Neuromuscul Dis. 2015;2: 119-130.

72.

Iyer

Wang

Renusch

, et al. SMN blood levels in a porcine model of spinal muscular atrophy. J Neuromuscul Dis. 2017;4: 59-66.

73.

Bonati

Holiga

Hellbach

, et al. Longitudinal characterization of biomarkers for spinal muscular atrophy. Ann Clin Transl Neurol. 2017;4: 292-304.

74.

Finkel

Crawford

Swoboda

, et al.; on behalf of the Pilot Study of Biomarkers for Spinal Muscular Atrophy (BforSMA) Trial Group. Candidate proteins, metabolites and transcripts in the Biomarkers for Spinal Muscular Atrophy (BforSMA) clinical study. PLoS One. 2012;7: e35462.

75.

Andreassi

Angelozzi

Tiziano

, et al. Phenylbutyrate increases SMN expression in vitro: relevance for treatment of spinal muscular atrophy. Eur J Hum Genet. 2004;12: 59-65.

76.

Mercuri

Bertini

Messina

, et al. Randomized, double-blind, placebo-controlled trial of phenylbutyrate in spinal muscular atrophy. Neurology. 2007;68: 51-55.

77.

Mattis

Rai

Wang

Chang

C-WT

Coady

Lorson

. Novel aminoglycosides increase SMN levels in spinal muscular atrophy fibroblasts. Hum Genet. 2006;120: 589-601.

78.

Liang

W-C

Yuo

C-Y

Chang

J-G

, et al. The effect of hydroxyurea in spinal muscular atrophy cells and patients. J Neurol Sci. 2008;268: 87-94.

79.

Arnold

Duque

Iyer

, et al. Normalization of patient-identified plasma biomarkers in SMNΔ7 mice following postnatal SMN restoration. PLoS One. 2016;11: e0167077.

80.

Porensky

Mitrpant

McGovern

, et al. A single administration of morpholino antisense oligomer rescues spinal muscular atrophy in mouse. Hum Mol Genet. 2012;21: 1625-1638.

81.

Passini

Roskelley

, et al. CNS-targeted gene therapy improves survival and motor function in a mouse model of spinal muscular atrophy. J Clin Investig. 2010;120: 1253-1264.

82.

Passini

Richards

, et al. Antisense oligonucleotides delivered to the mouse CNS ameliorate symptoms of severe spinal muscular atrophy. Sci Transl Med. 2011;3: 72ra18.

83.

Petzold

. Neurofilament phosphoforms: surrogate markers for axonal injury, degeneration and loss. J Neurol Sci. 2005;233: 183-198.

84.

Yuan

Rao

Veeranna, Nixon

R. A

. Neurofilaments and neurofilament proteins in health and disease. Cold Spring Harb Perspect Biol. 2017;9: a018309.

85.

Goldstein

Sternberger

. Phosphorylation protects neurofilaments against proteolysis. J Neuroimmunol. 1987;14: 149-160.

86.

Benatar

Wuu

Andersen

Lombardi

Malaspina

. Neurofilament light: a candidate biomarker of presymptomatic amyotrophic lateral sclerosis and phenoconversion. Ann Neurol. 2018;84: 130-139.

87.

Benatar

Wuu

Lombardi

, et al. Neurofilaments in pre-symptomatic ALS and the impact of genotype. Amyotroph Lateral Scler Frontotemporal Degener. 2019;20: 538-548.

88.

Weydt

Oeckl

Huss

, et al. Neurofilament levels as biomarkers in asymptomatic and symptomatic familial amyotrophic lateral sclerosis. Ann Neurol. 2016;79: 152-158.

89.

Sandelius

Zetterberg

Blennow

, et al. Plasma neurofilament light chain concentration in the inherited peripheral neuropathies. Neurology. 2018;90: e518-e524.

90.

Mattsson

Cullen

Andreasson

Zetterberg

Blennow

. Association between longitudinal plasma neurofilament light and neurodegeneration in patients with Alzheimer Disease. JAMA Neurol. 2019;76: 791-799.

91.

Benatar

Zhang

Wang

, et al.; CReATe Consortium. Validation of serum neurofilaments as prognostic and potential pharmacodynamic biomarkers for ALS. Neurology. 2020;95: e59-e69.

92.

Finkel

Mercuri

Darras

, et al.; ENDEAR Study Group. Nusinersen versus sham control in infantile-onset spinal muscular atrophy. New Engl J Med. 2017;377: 1723-1732.

93.

Darras

Crawford

Finkel

, et al. Neurofilament as a potential biomarker for spinal muscular atrophy. Ann Clin Transl Neurol. 2019;6: 932-944.

94.

De Vivo

Bertini

Swoboda

, et al.; NURTURE Study Group. Nusinersen initiated in infants during the presymptomatic stage of spinal muscular atrophy: interim efficacy and safety results from the phase 2 NURTURE study. Neuromuscul Disord. 2019;29: 842-856.

95.

Tozawa

Kasai

Tatebe

, et al. Intrathecal nusinersen treatment after ventriculo-peritoneal shunt placement: a case report focusing on the neurofilament light chain in cerebrospinal fluid. Brain Dev. 2020;42: 311-314.

96.

Winter

Guenther

Ludolph

Hermann

Otto

Wurster

. Neurofilaments and tau in CSF in an infant with SMA type 1 treated with nusinersen. J Neurol Neurosurg Psychiatry. 2019;90: 1068-1069.

97.

Spicer

Catapano

, et al. The altered expression of neurofilament in mouse models and patients with spinal muscular atrophy. Ann Clin Transl Neurol. 2021;8: 866-876.

98.

Wurster

Winter

Wollinsky

, et al. Intrathecal administration of nusinersen in adolescent and adult SMA type 2 and 3 patients. J Neurol. 2019;266: 183-194.

99.

Wurster

Steinacker

Günther

, et al.; the MND-Net. Neurofilament light chain in serum of adolescent and adult SMA patients under treatment with nusinersen. J Neurol. 2020;267: 36-44.

100.

Totzeck

Stolte

Kizina

, et al. Neurofilament heavy chain and tau protein are not elevated in cerebrospinal fluid of adult patients with spinal muscular atrophy during loading with Nusinersen. Int J Mol Sci. 2019;20: 5397.

101.

Rich

Fox

Yalvac

, et al. Neurofilament levels in CSF and serum in an adult SMA cohort treated with Nusinersen. Personal Communication. 2021.

102.

Faravelli

Meneri

Saccomanno

, et al. Nusinersen treatment and cerebrospinal fluid neurofilaments: an explorative study on spinal muscular atrophy type 3 patients. J Cell Mol Med. 2020;24: 3034-3039.

103.

Krosschell

Maczulski

Crawford

Scott

Swoboda

. A modified Hammersmith functional motor scale for use in multi-center research on spinal muscular atrophy. Neuromuscul Disord. 2006;16: 417-426.

104.

Kobayashi

Shi

Stephen

, et al.; the Biomarkers for Spinal Muscular Atrophy Study Group. SMA-MAP: a plasma protein panel for spinal muscular atrophy. PLoS One. 2013;8: e60113.

105.

Saks

Kongas

Vendelin

Kay

. Role of the creatine/phosphocreatine system in the regulation of mitochondrial respiration. Acta Physiol Scand. 2000;168: 635-641.

106.

Wallimann

Tokarska-Schlattner

Schlattner

. The creatine kinase system and pleiotropic effects of creatine. Amino Acids. 2011;40: 1271-1296.

107.

Rudnik-Schöneborn

Lützenrath

Borkowska

Karwanska

Hausmanowa-Petrusewicz

Zerres

. Analysis of creatine kinase activity in 504 patients with proximal spinal muscular atrophy types I–III from the point of view of progression and severity. Eur Neurol. 1998;39: 154-162.

108.

Lombardi

Querin

Ziff

, et al. Muscle and not neuronal biomarkers correlate with severity in spinal and bulbar muscular atrophy. Neurology. 2019;92: e1205-e1211.

109.

Alves

CRR

Zhang

Johnstone

, et al. Serum creatinine is a biomarker of progressive denervation in spinal muscular atrophy. Neurology. 2020;94: e921-e931.

110.

Freigang

Wurster

Hagenacker

, et al. Serum creatine kinase and creatinine in adult spinal muscular atrophy under nusinersen treatment. Ann Clin Transl Neurol. 2021;8: 1049.

111.

Deutsch

Osredkar

Plavec

Stres

. Spinal muscular atrophy after nusinersen therapy: improved physiology in pediatric patients with No significant change in urine, serum, and liquor 1H-NMR metabolomes in comparison to an age-matched, healthy cohort. Metabolites. 2021;11: 206.

112.

Buchthal

Olsen

. Electromyography and muscle biopsy in infantile spinal muscular atrophy. Brain. 1970;93: 15-30.

113.

Hausmanowa-Petrusewicz

Karwańska

. Electromyographic findings in different forms of infantile and juvenile proximal spinal muscular atrophy. Muscle Nerve. 1986;9: 37-46.

114.

Swoboda

Prior

Scott

, et al. Natural history of denervation in SMA: relation to age,SMN2 copy number, and function. Ann Neurol. 2005;57: 704-712.

115.

Lewelt

Krosschell

Scott

, et al. Compound muscle action potential and motor function in children with spinal muscular atrophy. Muscle Nerve. 2010;42: 703-708.

116.

Kaufmann

McDermott

Darras

, et al.; On behalf of the Muscle Study Group (MSG), The Pediatric Neuromuscular Clinical Research Network for Spinal Muscular Atrophy (PNCR). Prospective cohort study of spinal muscular atrophy types 2 and 3. Neurology. 2012;79: 1889-1897.

117.

Finkel

. Electrophysiological and motor function scale association in a pre-symptomatic infant with spinal muscular atrophy type I. Neuromuscul Disord. 2013;23: 112-115.

118.

David Arnold

Porensky

McGovern

, et al. Electrophysiological biomarkers in spinal muscular atrophy: proof of concept. Ann Clin Transl Neurol. 2014;1: 34-44.

119.

Weng

W-C

Hsu

Y-K

Chang

F-M

, et al. CMAP changes upon symptom onset and during treatment in spinal muscular atrophy patients: lessons learned from newborn screening. Genet Med. 2021;23: 415-420.

120.

Glanzman

Mazzone

Main

, et al. The Children’s Hospital of Philadelphia Infant Test of Neuromuscular Disorders (CHOP INTEND): test development and reliability. Neuromuscul Disord. 2010;20: 155-161.

121.

Krosschell

Bosch

Nelson

, et al.; the NeuroNEXT Clinical Trial Network and on behalf of the NN101 SMA Biomarker Investigators. Motor function test reliability during the NeuroNEXT spinal muscular atrophy infant biomarker study. J Neuromuscul Dis. 2018;5: 509-521.

122.

Al-Zaidy

Kolb

Lowes

, et al. AVXS-101 (Onasemnogene Abeparvovec) for SMA1: comparative study with a prospective natural history cohort. J Neuromuscul Dis. 2019;6: 307-317.

123.

Kariyawasam

D’Silva

Howells

, et al. Motor unit changes in children with symptomatic spinal muscular atrophy treated with nusinersen. J Neurol Neurosurg Psychiatry. 2021;92: 78–85.

124.

Darras

Chiriboga

Iannaccone

, et al.; ISIS-396443-CS2/ISIS-396443-CS12 Study Groups. Nusinersen in later-onset spinal muscular atrophy: long-term results from the phase 1/2 studies. Neurology. 2019;92: e2492-e2506.

125.

Neuwirth

Nandedkar

Stålberg

Weber

. Motor unit number index (MUNIX): a novel neurophysiological technique to follow disease progression in amyotrophic lateral sclerosis. Muscle Nerve. 2010;42: 379-384.

126.

Neuwirth

Nandedkar

Stålberg

, et al. Motor unit number index (MUNIX): a novel neurophysiological marker for neuromuscular disorders; test–retest reliability in healthy volunteers. Clin Neurophysiol. 2011;122: 1867-1872.

127.

Swash

. MUNIX in the clinic in ALS: MUNE comes of age. Clin Neurophysiol. 2017;128: 482-483.

128.

Neuwirth

Braun

Claeys

, et al. Implementing motor unit number index (MUNIX) in a large clinical trial: real world experience from 27 centres. Clin Neurophysiol. 2018;129: 1756-1762.

129.

Fatehi

Delmont

Grapperon

A-M

, et al. Motor unit number index (MUNIX) in patients with anti-MAG neuropathy. Clin Neurophysiol. 2017;128: 1264-1269.

130.

Günther

Neuwirth

Koch

, et al. Motor Unit Number Index (MUNIX) of hand muscles is a disease biomarker for adult spinal muscular atrophy. Clin Neurophysiol. 2019;130: 315-319.

131.

Querin

Lenglet

Debs

, et al. The motor unit number index (MUNIX) profile of patients with adult spinal muscular atrophy. Clin Neurophysiol. 2018;129: 2333-2340.

132.

Ahn

S-W

Kim

S-H

Kim

J-E

, et al. Reproducibility of the motor unit number index (MUNIX) in normal controls and amyotrophic lateral sclerosis patients. Muscle Nerve. 2010;42: 808-813.

133.

Nandedkar

Barkhaus

Stålberg

. Motor unit number index (MUNIX): principle, method, and findings in healthy subjects and in patients with motor neuron disease. Muscle Nerve. 2010;42: 798-807.

134.

Neuwirth

Barkhaus

Burkhardt

, et al. Tracking motor neuron loss in a set of six muscles in amyotrophic lateral sclerosis using the motor unit number index (MUNIX): a 15-month longitudinal multicentre trial. J Neurol Neurosurg Psychiatry. 2015;86: 1172-1179.

135.

Neuwirth

Barkhaus

Burkhardt

, et al. Motor unit number index (MUNIX) detects motor neuron loss in pre-symptomatic muscles in amyotrophic lateral sclerosis. Clin Neurophysiol. 2017;128: 495-500.

136.

Sanchez

Rutkove

. Electrical impedance myography and its applications in neuromuscular disorders. Neurother. 2017;14: 107-118.

137.

Rutkove

Zhang

Schoenfeld

, et al. Electrical impedance myography to assess outcome in amyotrophic lateral sclerosis clinical trials. Clin Neurophysiol. 2007;118: 2413-2418.

138.

Rutkove

Caress

Cartwright

, et al. Electrical impedance myography as a biomarker to assess ALS progression. Amyotroph Lateral Scler. 2012;13: 439-445.

139.

Wang

Spieker

Rutkove

. Electrical impedance myography for monitoring motor neuron loss in the SOD1 G93A amyotrophic lateral sclerosis rat. Clin Neurophysiol. 2011;122: 2505-2511.

140.

Sung

Rutkove

. Electrophysiologic biomarkers for assessing disease progression and the effect of riluzole in SOD1 G93A ALS mice. PLoS One. 2013;8: e65976.

141.

Rutkove

Darras

. Electrical impedance myography for the assessment of children with muscular dystrophy: a preliminary study. J Phys Conf Ser. 2013;434: 012069.

142.

Zaidman

Wang

Connolly

, et al.; the DART-EIM Clinical Evaluators Consortium. Electrical impedance myography in duchenne muscular dystrophy and healthy controls: a multicenter study of reliability and validity. Muscle Nerve. 2015;52: 592-597.

143.

Sanchez

Yim

Pacheck

Widrick

Rutkove

. Evaluation of electrical impedance as a biomarker of myostatin inhibition in wild type and muscular dystrophy mice. PLoS One. 2015;10: e0140521.

144.

Geisbush

Arnold

Rosen

Zaworski

Rutkove

. A comparison of three electrophysiological methods for the assessment of disease status in a mild spinal muscular atrophy mouse model. PLoS One. 2014;9: e111428.

145.

Arnold

McGovern

Sanchez

, et al. The neuromuscular impact of symptomatic SMN restoration in a mouse model of spinal muscular atrophy. Neurobiol Dis. 2016;87: 116-123.

146.

Rutkove

Gregas

Darras

. Electrical impedance myography in spinal muscular atrophy: a longitudinal study. Muscle Nerve. 2012;45: 642-647.

147.

Rutkove

Shefner

Gregas

, et al. Characterizing spinal muscular atrophy with electrical impedance myography. Muscle Nerve. 2010;42: 915-921.

148.

Arnold

Taylor

Nagy

Sanchez

Rutkove

. Electrical impedance myography detects age-related muscle change in mice. PLoS One. 2017;12: e0185614.

149.

Sproule

Punyanitya

Shen

, et al. Muscle volume estimation by magnetic resonance imaging in spinal muscular atrophy. J Child Neurol. 2011;26: 309-317.

150.

Morrow

Sinclair

CDJ

Fischmann

, et al. MRI biomarker assessment of neuromuscular disease progression: a prospective observational cohort study. Lancet Neurol. 2016;15: 65-77.

151.

Durmus

Yilmaz

Gulsen-Parman

, et al. Muscle magnetic resonance imaging in spinal muscular atrophy type 3: selective and progressive involvement. Muscle Nerve. 2017;55: 651-656.

152.

Pillen

Arts

IMP

Zwarts

. Muscle ultrasound in neuromuscular disorders. Muscle Nerve. 2008;37: 679-693.

153.

Darras

Rutkove

. Assessing spinal muscular atrophy with quantitative ultrasound. Neurology. 2010;75: 526-531.

154.

Pillen

van Alfen

Sorenson

, et al. Assessing spinal muscular atrophy with quantitative ultrasound. Neurology. 2011;76: 933. author reply 933-934.

155.

Regensburger

Wagner

Hanslik

, et al. Ultra–high-frequency ultrasound in patients with spinal muscular atrophy: a retrospective feasibility study. Muscle Nerve. 2020;61: E18-E21.

Update on Biomarkers in Spinal Muscular Atrophy

Abstract

Keywords

Introduction

Molecular Biomarkers

Survival motor neuron 2 (SMN2) copy number

Survival motor neuron (SMN) mRNA and protein levels

Neurofilaments (NFs)

Spinal muscular atrophy multi-analyte panel (SMA-MAP) protein analytes

Creatine kinase (CK) and creatinine (Crn)

Electrophysiology and Imaging Biomarkers

Compound muscle action potential (CMAP) and motor unit number estimation (MUNE)

Motor unit number index (MUNIX) and motor unit size index (MUSIX)

Electrical impedance myography (EIM)

Magnetic resonance imaging and muscle ultrasonography

Conclusion

Footnotes

Funding:

Declaration of Conflicting Interests:

Author Contributions

ORCID iDs

References