Single Nucleotide SMN1 Variants in a Cohort of Individuals With Spinal Muscular Atrophy

Abstract

Background and Objectives

Spinal muscular atrophy 5q (SMA) is a motor neuron disorder caused by recessive pathogenic variants in the SMN1 gene, which encodes the survival motor neuron (SMN) protein. While the majority of patients with SMA exhibit homozygous deletions in SMN1, a minority (2%–5%) of patients with SMA harbor an SMN1 deletion plus a single nucleotide variant on the second allele, which can be identified through direct gene sequencing. The comprehensive characterization of patients with SMA is increasingly crucial considering emerging therapies and newborn screening initiatives.

Methods

Over the past 20 years, we confirmed a molecular diagnosis of SMA in 149 patients consisting of 138 postnatal and 11 prenatal cases, through a quantitative molecular approach (real-time PCR and/or multiplex ligation-dependent probe amplification) associated with direct sequencing.

Results

We identified homozygous SMN1 deletions in 142 probands (95%). The remaining 7 patients (5%) displayed heterozygous SMN1 deletion in compound with a different molecular defect. Notably, 1 patient presented with an intronic variant necessitating mRNA transcript analysis, a process that extended the time to diagnosis.

Discussion

The identification of small pathogenic variants in patients with SMA is of paramount importance for enhancing diagnosis and prognosis, deciphering response variations to existing treatments, and designing novel therapies tailored to address these genetic variants. We propose a paradigm shift from current guidelines, particularly for patients with a heterozygous SMN1 deletion and a clinically compatible SMA phenotype, especially when reduced SMN transcript levels are evident. In such cases, expedited therapy initiation, including reversible treatments like nusinersen or risdiplam, is recommended without waiting for the completion of the molecular testing, thus minimizing delays in crucial therapeutic interventions.

Introduction

Spinal muscular atrophy 5q (SMA) is a severe neuromuscular disorder characterized by progressive muscle wasting, weakness, and atrophy because of the deterioration of motor neurons in the anterior horn of the spinal cord. Its incidence ranges from 1 in 6,000 to 1 in 10,000 individuals.^1-3

Historically, and prior to disease-modifying treatment being available, SMA has been categorized into 4 types based on age at onset and the highest motor function achieved.⁴ This historical description by type of SMA is based on the age that symptoms began and the highest physical milestone achieved. Within each type, abilities can vary from person to person. SMA I is the most severe subtype, presenting within the first 6 months of life, associated with inability to achieve independent sitting posture and reduced survival. SMA II, or intermediate, displays an onset between 6 and 18 months and inability to achieve independent ambulation. SMA III typically becomes evident after 18 months of age, while SMA IV is the adult-onset form.^4,5 Within each type, a certain degree of clinical variability has been described.

The approval of disease-modifying therapies and the implementation of newborn screening (NBS) programs, which allow diagnosis and initiation of therapy before the onset of symptoms, have highlighted the limitations of this classification to describe the new generation of SMA infants and children treated early in life.⁶

SMA is genetically caused by biallelic defects in the survival motor neuron-1 (SMN1) gene located in the telomeric region of the long arm of chromosome 5 (chr5q13).⁷ A closely related paralogous gene, SMN2, situated in the centromeric region of 5q arm, produces a protein which can only partially compensate for the absence of SMN1.^7,8 In fact, the difference between SMN1 and SMN2, at the genomic level, consists of 5 nucleotide positions, 2 of which are located in exons 7 and 8 (Figure 1). SMN2 primarily generates transcripts lacking exon 7, leading to unstable SMN protein. In the absence of SMN1, the number of SMN2 copies inversely correlates with disease severity, making it a major genetic modifier of the phenotype.^9-11

Figure 1. Molecular Changes in SMN1 and SMN2 Sequences Described in the Text.

Schematic diagram of SMN1 (green) and SMN2 (red) genes at chromosomal region 5q13. Exon numbers are indicated. Nucleotide positions differentiating SMN1 and SMN2 are shown above the diagram. Single nucleotide variations in SMN1 and SMN2 found in our cohort of Patients with SMA are indicated below the diagram. SMA = spinal muscular atrophy 5q; SMN = survival motor neuron.

About 97% of patients with SMA exhibit homozygous deletions or spontaneous SMN1-to-SMN2 conversion at quantitative analysis (SMN1-null genotype), whereas the remaining cases are compound heterozygotes, with an SMN1 deletion on one allele and a different molecular defect on the other one.¹ Several SMN1 pathogenic variants (classified as pathogenic or likely pathogenic based on American College of Medical Genetics and Genomics classification) have been documented so far including deletions, duplications, insertions, splice site variants, and nucleotide substitutions introducing aminoacidic changes or premature termination codons.¹² The ClinVar database (last access December 2024), public archive of interpretations of clinically relevant variants, reports 60 pathogenic defects, 28 likely pathogenic defects, and 50 variants of uncertain significance (VOUS) in the SMN1 gene. The c.815A>G, p.(Tyr272Cys) missense variant stands out as the most common occurring in 20% of cases^13,14 and is mainly associated with the most severe form of SMA (SMA I).¹⁵

Achieving a genetic diagnosis in SMA is of paramount importance not only for genetic counseling but also to guarantee access to the newly available disease-modifying therapies. Despite the international approval of 3 therapies for SMA (nusinersen, onasemnogen abeparvovec, and risdiplam), access to these treatments remains challenging for patients with unidentified molecular defects, including single nucleotide variants.^16-18

Advancements in gene sequencing technology have made genetic diagnosis a reliable supplementary method for SMA, especially in cases where a second pathogenic variant is elusive. Quantifying SMN transcript levels can be valuable in predicting the disease severity and can aid in genetic diagnosis and carrier identification. Combining traditional diagnostic (quantitative) methods, such as multiplex ligation-dependent probe amplification (MLPA) and fluorescent or real-time quantitative PCR protocols, with classical or next-generation sequencing methods, is expected to significantly enhance diagnostic accuracy.

Here, we describe the clinical and molecular features of 7 patients with SMA presenting heterozygous SMN1 deletion plus a single nucleotide variant and 5 patients with SMA with an SMN1-null genotype in which direct sequencing allowed the detection of SMN2 variants acting as genetic modifier.

Methods

Patient Enrollment and Sample Collection

This study was carried out at Istituto di Ricovero e Cura a Carattere Scientifico (IRCCS) Fondazione Ca' Granda Maggiore Hospital and Neuromuscular Omnicenter (NEMO) Clinical Center of Milan. We retrospectively analyzed patients found positive at SMA testing in a period from January 2003 to December 2023. Molecular testing was performed on blood-extracted DNA in 138 cases (postnatal testing) and in 11 chorionic villus samples (prenatal testing).

This study was approved by the Institutional Review Board of IRCCS Fondazione Ca' Granda Maggiore Hospital (0017300 Rebineu).

Legal guardians of these cases or patients themselves provided their voluntary consent for genetic testing and signed the informed consent forms. The entire experimental process strictly adhered to the ethical standards established by both the institutional and national research committees.

Quantitative PCR

We assessed the copy number of SMN1 and SMN2 via quantitative PCR (qPCR) experiments, using specific primers (targeting exon 7) that can differentiate between telomeric and centromeric genes.¹⁹ Using the 7500 Applied Biosystems International (ABI) PCR system, we conducted relative quantification experiments with at least 2 genomic reference genes including CFTR, BCKDHA, and RNASEP. We used 25 ng of genomic DNA as template, with each determination carried out in triplicate.

MLPA Analysis

MLPA analysis, evaluating exons 7 and 8 of SMN1 and SMN2 genes, was used as a confirmation method in the patients described in the text. We used the synthetic oligonucleotide amplification with ligation-specific analysis (SALSA) MLPA Kit P060 (MRC-Holland, Amsterdam, Netherlands) according to manufacturer instructions. The resulting MLPA products were analyzed using an ABI PRISM 3130 genetic analyzer (Applied Biosystems International Inc., CA) and processed with Gene Mapper version 3.5 software (Thermo Fisher Scientific Inc., Waltham). To determine the relative peak height for each sample, a comparison was made with 4 normal control samples using Coffalyser version 9 software (Coffalyser MLPA, Amsterdam, Netherlands).

DNA Sequencing

For patients presenting homozygous SMN1 deletions, sequence analyses of SMN2 intron 6 and exon 7 were performed, aiming to detect potential modifier variants. For patients presenting a single heterozygous SMN1 deletion, we proceeded with Sanger sequencing. During the primary sequencing phase, both the SMN1 and SMN2 genes were targeted using unspecific primers. When rare variants were pinpointed, nested PCR sequencing of an SMN1-specific long-range PCR amplicon was used.²⁰ Sanger sequencing was performed using the BigDye Terminator v3.1 Cycle Sequencing Kit from Thermo Fisher.

Databases

SMN1: OMIM 600354; NM_000344.3. SMN2: OMIM 601627; NM_017411.3.

Data Availability

Deidentified data are available on reasonable request.

Results

Clinical and Molecular Characterization of Patients

SMA was molecularly confirmed in 149 patients including 138 postnatal cases and 11 prenatal samples. Based on the initial quantitative analysis, a total of 142 probands exhibiting homozygous deletion of SMN1 exon 7 were classified as patients with SMA.

In 7 patients, a single SMN1 copy was detected by the quantitative approach. In these cases, direct sequencing allowed the detection of single nucleotide pathogenic or likely pathogenic variants (Table 1). Specifically, a null allele was observed in 2 SMA I cases: c.469C>T, p.(Gln157*) (Patient 4) and c.511G>T, p.(Glu171*) (Patient 2, eFigure 1). With the limited support of 2 SMN2 copies, these 2 changes likely contribute to the severe clinical phenotype observed. The splice site change c.888+1G> C was detected in a single SMA case (Patient 1). This variant, previously found to hamper exon 7 inclusion,⁸ and the presence of a single copy of SMN2 explains the dramatic (congenital) clinical presentation of Patient 1.

Table 1.

Clinical and Molecular Features of Patients With SMA Described in the Text

Patients with heterozygous SMN1 deletion paired with a single nucleotide variant
Patient #	Age at onset	Age at last clinical update	Sign(s) at onset	Clinical diagnosis	Therapy (age at first administration)	Effect of therapy on phenotype	SMN1 exon 7a copy number	SMN1 exon 8a copy number	SMN2 exon 7a copy number	SMN2 exon 8a copy number	Single nucleotide variant	ACMG	References for variant	References for the patient (ID in the text)
P1	Birth	4 mo (death)	Hypotonia, lack of spontaneous limb movements, areflexia	SMA I	NA	NA	1	1	1	1	c.888 + 1G>C, r.spl (SMN1)	Pathogenic	8	8
P2	<6 mo	10–15 y	Hypotonia, generalized muscle weakness	SMA I	Nusinersen (6 mo)	Clinical stability	1	1	2	2	c.511G>T p. (E171a) (SMN1)	Likely pathogenic	Novel	This study
P3	Birth	NA	Hypotonia, generalized muscle weakness, severe respiratory impairment	SMA I	NA	NA	1	1	2	2	c.815A>G, p. (Y272C) (SMN1)	Pathogenic	13,29,44-46	29 (patient 3)
P4	Birth	NA	Hypotonia, generalized muscle weakness, severe respiratory impairment	SMA I	NA	NA	1	1	2	2	c.469C>T, p. (Q157a) (SMN1)	Pathogenic	13,44,45	29 (patient 4)
P5	1–2 y	40–45 y	Inability to walk (sitter)	SMA II	NA	NA	1	1	2	2	c.815A>G p. (Y272C) (SMN1)	Pathogenic	13,29,44-46	This study
P6	1–2 y	15–20 y (alive)	Inability to sit (nonsitter)	SMA II	NA	NA	1	1	2	2	c.815A>G p. (Y272C) (SMN1)	Pathogenic	13,29,44-46	This study
P7	14 y	34 y (alive)	Walker, but loss of ability to walk	SMA III	Nusinersen (28 y)	Clinical stability	1	1	1	1	c.389A>G, p. (Y130C) (SMN1)	Likely pathogenic	29,30	29 (patient 23)

Abbreviations: ACMG = American College of Medical Genetics and Genomics classification; MLPA = multiplex ligation-dependent probe amplification; NA = not available; NIV = noninvasive ventilation; SMA = spinal muscular atrophy 5q.

^aAssayed by MLPA analysis targeting SMN1 and SMN2 exons 7 and 8.

The missense variant c.389A>G, p.(Tyr130Cys) was found in a patient with SMA III (Patient 7), while the c.815A>G, p.(Tyr272Cys) substitution occurred in 1 SMA I (Patient 3) and 2 SMA II probands (Patients 5 and 6).

In addition, direct sequencing of SMN2 exon 7 in all SMN1-null patients detected the presence of the modifier variants c.859G>C, p.(Gly287Arg) in 4 patients (homozygous in Patients 8 and 9 and heterozygous in Patients 10 and 11) and c.835-44A>G in Patient 12.^21,22

Clinical and molecular features of these patients are reported in Table 2. Specifically, age at onset was earlier for the heterozygous Patient 10 (within the first years of her 2nd decade of life) compared with those of the homozygous siblings Patients 8 and 9 (at the end of their 3rd decade of life). Nowadays, all these 3 adult patients cannot walk without aid or cannot perform sit-to-stand and stand-to-sit transitions independently. Patient 8 also requires nocturnal noninvasive ventilation. Nusinersen treatment is ongoing with effect of overall clinical stability.

Table 2.

Clinical and Molecular Features of SMN1-Null Patients Presenting a SMN2 Modifier Variant

SMN1-null patients with SMN2 modifier variants
Patient #	Age at onset	Age at last clinical update	Sign(s) at onset	Clinical diagnosis	Therapy (age at first administration)	Effect of therapy on phenotype	SMN1 exon 7a copy number	SMN1 exon 8a copy number	SMN2 exon 7a copy number	SMN2 exon 8a copy number	Modifier variant	ACMG	References for variant	References for the patient (ID in the text)
P8	20–30 y	45–50 y (alive)	Walker, but loss of the ability to walk	SMA III	Nusinersen (30 y)	Clinical stability overall. No NIV. Manual wheelchair	0	0	2	2	c.859G>C, p. (G287R) homozygous (SMN2)	VUS	21	This study
P9, brother of P8	20–30 y	50–55 y (alive)	Walker, but loss of the ability to walk	SMA III	Nusinersen (33 y)	Clinical stability overall. Nocturnal NIV. Manual wheelchair	0	0	2	2	c.859G>C, p. (G287R) homozygous (SMN2)	VUS	21	This study
P10	10–15 y	40–45 y (alive)	Walker, but loss of the ability to walk	SMA III	Nusinersen (42 y)	Clinical stability overall. No NIV. Manual wheelchair	0	0	2	2	c.859G>C, p. (G287R) heterozygous (SMN2)	VUS	21	This study
P11	1–2 y	40–45 y (alive)	Inability to achieve independent standing and walking (sitter)	SMA II	Nusinersen (38 y)	Clinical stability overall. Nocturnal NIV. Manual wheelchair	0	0	2	2	c.859G>C, p. (G287R) heterozygous (SMN2)	VUS	21	This study
P12	1–2 y	15–20 y (alive)	Walker, but emergence of difficulties in walking and in postural transitions	SMA III	Nusinersen (13 y)	Improved ambulation. No NIV. Manual wheelchair only for long distances	0	1	3	2	c.835-44A>G heterozygous (SMN hybrid gene)	Likely benign	21	This study

Abbreviations: ACMG = American College of Medical Genetics and Genomics classification; MLPA = multiplex ligation-dependent probe amplification; NA = not available; NIV = noninvasive ventilation; SMA = spinal muscular atrophy 5q; SMN = survival motor neuron.

^aAssayed by MLPA analysis targeting SMN1 and SMN2 exons 7 and 8.

Results by MLPA analysis evaluating exons 7 and exons 8 from SMN1 to SMN2 genes in Patient 12 are compatible with the presence of an hybrid SMN gene (SMN1 presenting SMN2 exon 7), the only case in our cohort.

Discussion

SMA is a neuromuscular disorder characterized by the loss of motor neurons, often leading to significant morbidity and mortality. 95%–98% of patients diagnosed with SMA5q exhibit a homozygous deletion of the SMN1 gene. The remaining 2%–5% cases display alternative SMN1 defects.²³ Early diagnosis is crucial for timely intervention and improved clinical outcomes. As presented in this study, over the past 20 years, our team has achieved a molecular diagnosis of SMA in 149 cases. From this cohort, 142 probands (95%) demonstrated homozygous SMN1 deletions, while 7 patients (5%) exhibited a heterozygous SMN1 deletion paired with a single nucleotide substitution. The phenotypic classification of these 7 patients is as follows: SMA type I (n = 4%, 57.1%), SMA type II (n = 2%, 28.6%), and SMA type III (n = 1%, 14.3%). In the patient with SMA III (P7), we found the missense change c.389A>G, p.(Tyr130Cys) while the c.815A>G, p.(Tyr272Cys) substitution occurred in 3 probands (1 SMA I, 2 SMA II; P3, P5-P6).

The advent of molecular genetic testing allowing the independent quantitative analysis of SMN1 and SMN2 has revolutionized the diagnostic approach to SMA. In addition, sequencing becomes paramount for patients showcasing a heterozygous deletion in the SMN1 gene to identify another concomitant defect.²⁴

The timing and accuracy of SMA diagnosis, particularly in terms of genotype elucidation, is of utmost significance, given the availability of new therapeutic strategies for SMA. This urgency is underscored by data suggesting that older patients exhibit heightened resistance to treatments compared with their younger counterparts and that therapeutic efficacy is markedly reduced when not initiated at the proper time because of delayed diagnosis or misdiagnosis.^18,25 The best clinical responses have been shown to occur when treatment is initiated presymptomatically.^26,27 This is potentially attributed to irreversible tissue deterioration observed in aged and already symptomatic individuals.

NBS for SMA has been shown to be successful in allowing infants to be treated before symptoms onset and has resulted in improved clinical outcomes.^12,28 However, a relevant number of SMA diagnosis can escape NBS screening because it does not allow the reporting of heterozygous SMN1 deletion carriers nor the identification of SMN1 and SMN2 single nucleotide variants.^12,28 A two-pronged molecular diagnostic approach, with initial MLPA or qPCR followed by meticulous gene sequencing and transcript analysis, emerges as fundamental for diagnostic purposes and is also mandatory to refine prognosis and eventually comprehend variations in treatment responses.

The c.389A>G, p.(Tyr130Cys) variant, located in exon 3, partially impairs SMN protein oligomerization, and it has been previously associated with milder SMA III phenotypes irrespective of SMN2 copies.^29,30 This molecular defect was exclusively found in a single patient with SMA III. By contrast, the c.815A>G, p.(Tyr272Cys) substitution in exon 6 severely impairs SMN protein oligomerization and function. This substitution partially destroys the so-called YG box domain (residues 254–280), which contains the (YxxG)3 motif.³¹ The YG box domain is essential for oligomerization of SMN protein. The c.815A>G is more frequently found in the most severe SMA I cases,^14,32 even though it can also be found in patients with SMA II.¹⁵

In our cohort, this variant was found in 1 patient with SMA I and in 2 patients with SMA II (Table 1). The rate of patients harboring the p.(Tyr272Cys) change aligns literature data which identifies the p.(Tyr272Cys) variant in 20% of patients with SMA I presenting a small variant.^14,33

Except for p.(Glu171*), all the other SMN1 small variants detected in our cohort have been previously reported and functionally validated in other patients with SMA, facilitating their clinical interpretation. The identification of novel SMN1 defects in suspected SMA patients with heterozygous SMN1 deletion poses the problem of the assessment of their pathogenic burden. The first step is segregation testing in available family members and the accurate attribution of the variant to SMN1 (and not SMN2) genomic location. Nonsense variants are usually classified as likely pathogenic or pathogenic and their predicted effects on transcript stability could be assessed by using qRT-PCR analysis on blood-derived RNA or in patient's cells before and after cycloheximide treatment, a reagent used to inhibit nonsense-mediated mRNA decay of transcripts harboring premature termination codons.³⁴ The pathogenicity of novel splice sites changes can be assessed by performing qualitative and quantitative analysis of blood-derived RNA and/or by applying in silico predictors. Finally, the interpretation of novel variants leading to aminoacidic substitutions represents a challenge because different complementary methods are required to ascertain their pathogenic behavior such as evaluation of steady state levels and assembly of mutant SMN protein. These investigations might require the establishment of ad hoc cellular models or the in vitro expression of mutant SMN constructs.

Currently, the copy number of SMN2 is the key factor in therapeutic decisions for patients with SMA.²³ SMN2 copy number can also predict clinical severity, decreasing it in a dose-dependent manner,¹¹ even if this correlation is sometimes not absolute.^18,35 Apart from SMN2 copy number, 2 main variants in the SMN2 gene sequence have been reported to influence clinical presentation, acting as positive modifiers in the SMA phenotype: c.859G>C, p.(Gly287Arg) in exon 7 and c.835-44A>G in intron 6.³⁵ The presence of 1 of these 2 variants in SMN2 results in a clinical presentation with later onset and milder course.²¹

In our cohort, the SMN2 c.859G>C variant in exon 7 was found in 4 patients of which 2 related (P8-P11). We also detected the c.835-44A>G variant in SMN1-null Patient 12.

The c.859G>C change is known to create a new exonic splicing enhancer element, thereby increasing SMN2 exon 7 inclusion (from 40% to 50%–70% in vitro) and subsequently the amount of full-length transcripts originating from SMN2 alleles.^11,36,37 Also, the patient's phenotype appears to be milder in cases with a greater number of SMN2 copies harboring the c.859G>C variant, confirming an additive effect.³⁷ In patients with 0 SMN1 and 2 SMN2 copies, in fact, those who are heterozygous for the c.859G>C variant in SMN2 typically exhibit a SMA II phenotype, as in the case of our Patient 11. Conversely, homozygosity for this variant generally results in an SMA III phenotype.³⁶ In our cohort, the presence of the c.859G>C variant in homozygosity appears to be linked to a milder phenotype compared with those patients with the same variant in heterozygosity. These findings support an additive effect of the c.859G>C variant on top of the SMN2 copy number.^11,36,37

The SMN2 c.835-44A>G variant has been recognized as an additional positive disease modifier.²² In patients with 0 SMN1 copies and 3 SMN2 copies, the c.835-44A>G variant was demonstrated to improve the inclusion of exon 7 in the SMN2 gene. Detailed studies revealed that the c.835-44A>G change reduces the binding affinity of the RNA binding protein HuR to the −44 region. HuR normally acts as a splicing repressor, so its decreased binding because of the c.835-44A>G transition results in a moderate increase in exon 7 inclusion. This effect is beneficial because it leads to higher levels of functional SMN protein, which is crucial for mitigating the severity of the disease phenotype. In addition, results by MLPA analysis evaluating exons 7 and exons 8 from SMN1 to SMN2 genes in Patient 12 are compatible with the presence of an hybrid SMN gene (SMN1 presenting SMN2 exon 7), the only case in our cohort.

Hybrid gene formation is regarded as a fusion event between the flanking region of SMN2 exon 7 and the flanking region of SMN1 exon 8.^38,39 The presence of hybrid genes has been recently associated with milder SMA phenotypes respect to those expected based on SMN2 copy number.³⁸ However, some exception exists, and MLPA analysis seems to detect only a limited number of potential hybrid structures.³⁹ Patient 12 displays an SMA III phenotype, with a slow disease progression because the age of first symptoms onset, at 15 months, when mild walking difficulties appeared. Nusinersen therapy, started at age 13 years, is still ongoing with a moderate clinical improvement, especially on ambulation. Nowadays, independent walking and standing are possible, and the patient uses a manual wheelchair only for long distances, suggesting a positive outcome of nusinersen on his peculiar genotype.

Patients with 3 copies of the SMN2 gene are known to have a more variable clinical presentation compared with patients with 1, 2, 4, or more copies, with phenotypes spanning from SMA type I to SMA IIIc.⁴⁰ As already stated, this variability cannot be attributed solely to SMN2 copy number.

Our study highlights the importance of rapidly identifying rare sequence variations within both the SMN1 and SMN2 genes. SMN1 sequencing is crucial to confirm SMA diagnosis in cases with heterozygous deletion. In addition, detecting SMN2 copy number provides prognostic value, while identifying SMN2 modifier variants can help explain the variability in phenotype severity and treatment response. As stressed before, the modifier SMN2 gene deserves in-depth analysis via sequencing to enable more accurate genotype-phenotype correlations and to support SMA management.⁴¹ In agreement with 2018 care recommendations,⁴² we also suggest performing sequence analysis of (at least) SMN2 exon 7 for the identification of genetic modifier in patients presenting SMN1-null genotypes.

Overall, these aspects might contribute to a significant diagnostic delay in patients with SMA not presenting the classical SMN1 null genotype. This aspect is highlighted in Patient 1, diagnosed at 1 month of age in 2015 (before the advent of nusinersen). The availability of blood-derived RNA from the proband and his parents and the application of multiple molecular assays were fundamental to prove the severe detrimental effect of the c.888+1G>C variant which contributed to a fatal outcome 3 months after the diagnosis. The time required for these investigations resulted in a diagnostic delay that must be minimized in the present era of molecular therapies for patients with SMA.

We favor the collection of blood-derived RNA and patient's primary cells (fibroblasts or lymphocytes) for transcript and protein analysis to speed up the identification of the second molecular defect in suspected SMA patients with single SMN1 deletions. In principle, this approach could be also extended to suspected SMA patients with 2 or more intact (not deleted) SMN1 copies to check for the presence of biallelic small variants. It is noteworthy that this circumstance, once considered extremely rare and restricted to conceptions in consanguineous pedigrees, has been recently observed to be more frequent than expected, as evidenced by novel diagnosis because of innovative bioinformatic pipelines of next-generation sequencing diagnostic scans.⁴³

Finally, we envision a reorientation from prevailing guidelines, especially in scenarios involving patients with a heterozygous SMN1 deletion manifesting a clinical phenotype consistent with SMA. This is accentuated when there is a conspicuous reduction in SMN transcript levels. For these patients, we advocate for a swift commencement of therapy, encompassing reversible treatments such as nusinersen or risdiplam. This approach negates the necessity of waiting for dual pathogenic variant identification, thus minimizing delays in vital therapeutic interventions.

Glossary

MLPA

multiplex ligation-dependent probe amplification

NBS

newborn screening

qPCR

quantitative PCR

SMA

spinal muscular atrophy 5q

SMN

survival motor neuron

Author Contributions

M. Rimoldi: drafting/revision of the manuscript for content, including medical writing for content; analysis or interpretation of data. F. Magri: major role in the acquisition of data. M. Meneri: major role in the acquisition of data. D. Gagliardi: major role in the acquisition of data. V. Ada SANSONE: major role in the acquisition of data. E. Albamonte: major role in the acquisition of data L Ottoboni: major role in the acquisition of data. G.P. Comi: drafting/revision of the manuscript for content, including medical writing for content; analysis or interpretation of data. E. Mercuri: drafting/revision of the manuscript for content, including medical writing for content. F.D. Tiziano: drafting/revision of the manuscript for content, including medical writing for content. D. Ronchi: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; study concept or design; analysis or interpretation of data. S. Corti: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; study concept or design; analysis or interpretation of data.

Study Funding

This study was (partially) funded by Italian Ministry of Health–Current research IRCCS Ca' Granda Ospedale Maggiore Policlinico (PI: Stefania Corti) and Italian Ministry of Health– Ricerca Finalizzata RF-2019-12370334 to GPC and EM.The PNC “Hub Life Science-Diagnostica Avanzata (HLS-DA), PNC-E3-2022-23683266- CUP: C43C22001630001” is funded by the Italian Minister of Health. The support of Italian Ministry of Education and Research “Dipartimenti di Eccellenza Program 2023–2027”–Dept of Pathophysiology and Transplantation, University of Milan to L. Ottoboni, G.P. Comi, D. Ronchi and S. Corti is gratefully acknowledged.

Disclosure

M. Rimoldi, F. Magri, M. Meneri, and D. Gagliardi reports no disclosures relevant to the manuscript; V. Sansone served on the scientific advisory board for Novartis, Roche, Biogen and Sarepta; E. Albamonte, and L. Ottoboni reports no disclosures relevant to the manuscript; G.P. Comi served on the scientific advisory board for Novartis, Roche, Biogen and Sarepta; E.M. Mercury, and F.D. Tiziano served on the scientific advisory board for Novartis, Roche, Biogen and Sarepta; D.Ronchi reports no disclosures relevant to the manuscript; S. Corti served on the scientific advisory board for Novartis, Roche, Biogen and Sarepta. Go to Neurology.org/NG for full disclosures.