Muscular dystrophies (MD) and congenital myopathies (CM) consist of a wide variety of inherited muscle disorders that lead to significant disability in patients of different age groups (North et al., 2014; Mercuri et al., 2019). MD feature progressive loss of muscle strength along with abnormal pathological changes seen on muscle biopsy (Mercuri et al., 2019). They have traditionally been categorized based on age at onset, main clinical and biopsy characteristics, and results of immunostaining (Sewry et al., 2019). MD include conditions affecting proteins like dystrophin, sarcoglycans and dysferlin. Identifying the defective proteins can be challenging as multiple proteins may be affected. CM are a group of muscle diseases present from birth, usually causing hypotonia and weakness (North et al., 2014). They tend to be static or slowly progressive. CM have been classified according to the predominant morphological features seen on muscle biopsy (Witting et al., 2017).

MD are a heterogeneous group of inherited disorders characterized by progressive muscle weakness and degeneration. The most common types of MD include Duchenne muscular dystrophy (DMD), Becker muscular dystrophy (BMD), limb-girdle muscular dystrophies (LGMD), facioscapulohumeral muscular dystrophy (FSHD), and congenital muscular dystrophies (CMD) (Mercuri et al., 2019). DMD and BMD are caused by mutations in the dystrophin gene (DMD) and are characterized by progressive proximal muscle weakness, with DMD being more severe than BMD (Birnkrant et al., 2018). LGMD represents a group of disorders with predominant weakness in the shoulder and pelvic girdle muscles, caused by mutations in various genes encoding proteins involved in muscle function and structure (Straub et al., 2018). FSHD is an autosomal dominant disorder caused by contractions in the D4Z4 repeat region on chromosome 4q35, leading to characteristic facial and shoulder girdle muscle weakness (Tawil et al., 2015). CMDs are a group of early-onset muscular dystrophies with variable severity and genetic heterogeneity (Kang et al., 2015).

Thanks to significant advances in molecular genetics during the last 2 decades, molecular genetic testing has dramatically improved the diagnosis of neuromuscular disorders (North et al., 2014; Mercuri et al., 2019). Over 100 genes associated with neuromuscular disorders have been identified (Benarroch et al., 2020). The ability to correlate specific gene defects to clinical phenotypes has transformed our understanding of these conditions. Clinical manifestations of neuromuscular disorders in pediatric patients often lack specificity, presenting nonspecific symptoms, such as motor developmental delay, hypotonia, and weakness, which are inadequate for a definitive diagnosis. Many neuromuscular diseases have phenotypic heterogeneity, which makes diagnosis challenging. Next-generation sequencing (NGS) in the molecular diagnosis of neuromuscular diseases is becoming more common. The identification of causative genes may be achieved with enhanced efficiency by integrating the patient’s phenotype, pathological findings from muscle biopsy, and sequencing data (Böhm et al., 2013; Valencia et al., 2013). However, costs remain a barrier to implementing NGS technology in many settings.

Whole exome sequencing (WES) has emerged as a powerful diagnostic tool for patients with suspected genetic disorders during the last decade. WES focuses on gene protein-coding regions, allowing for comprehensive analysis of variants across many genes simultaneously. WES has been found to have a diagnostic yield of 25%–73% in pediatric patients with neuromuscular disorders (Richards et al., 2015). However, some variation is based on the patient cohort and the filtering strategies. WES enables genetic diagnosis even in patients with atypical symptoms or ultra-rare conditions.

The primary aim of this study was to investigate the diagnostic yield and clinical utility of WES in a cohort of Taiwanese pediatric patients with suspected CM or MD. We sought to determine the proportion of patients who received a genetic diagnosis through WES and to characterize the spectrum of genetic variants identified in this population. Furthermore, we aimed to explore the potential of WES to refine clinical diagnoses, guide management decisions, and inform genetic counseling for affected families.

This study adhered to the principles of the Helsinki Declaration. From January 2018 to August 2023, a descriptive, cross-sectional study was conducted on pediatric patients clinically diagnosed with MD and CM at MacKay Memorial Hospital in Taipei, Taiwan. MD and CM were clinically diagnosed by experienced pediatric neurologists based on medical histories, age of symptom onset, disease progression, physical examinations, and standard diagnostic tests. Patients with suggestive clinical phenotypes underwent single gene testing, including multiplex ligation-dependent probe amplification (MLPA) for spinal muscular atrophy (SMN1) and Duchenne muscular dystrophy (DMD). After preliminary studies, patients with no known molecular etiologies were recruited for WES analysis (Figure 1).

In our study, none of the 46 patients who underwent WES had a prior muscle biopsy. The decision to proceed directly to WES was made by the treating physicians based on a thorough clinical evaluation and the suspicion of a genetic etiology for the patients’ muscular disorders. This approach was aimed at streamlining the diagnostic process and minimizing invasive procedures for the patients.

The decision to segregate the MD and DMD groups was based on the distinct clinical and molecular characteristics of these conditions. DMD is a well-characterized X-linked disorder caused by mutations in the DMD gene, whereas the MD group encompasses a heterogeneous set of disorders with varying genetic etiologies and clinical presentations.

As for the MLPA-negative DMD group, these patients had initially undergone MLPA testing for the DMD gene, which did not identify any causative variants. However, given the strong clinical suspicion of DMD, they were included in our cohort to explore the potential of WES in detecting single nucleotide variants or small indels that may have been missed by MLPA.

Regarding the MD group, these patients did not receive MLPA, single-gene sequencing, or gene panel analysis prior to WES. They were included in our cohort based on the clinical diagnosis of muscular dystrophies but without a confirmed molecular etiology. The decision to proceed directly to WES for these patients was made to streamline the diagnostic process and avoid sequential testing, which can be time-consuming and costly.

Genomic DNA was extracted from peripheral blood samples using the QIAamp DNA Blood Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. Exome capture was performed using the SureSelect Human All Exon V5 Kit (Agilent Technologies, Santa Clara, CA, United States), which targets approximately 50 Mb of the human exonic regions. Sequencing libraries were prepared using the SureSelect XT Library Prep Kit (Agilent Technologies) and sequenced on the Illumina HiSeq 4000 platform (Illumina, San Diego, CA, United States) with 150-bp paired-end reads.

Variants were filtered based on the following criteria: (a) located in exons or flanking introns, (b) resulting in amino acid changes, (c) minor allele frequency less than 1% in 1000 Genomes database, (d) allele frequency less than 0.1% in Genome Aggregation Database (gnomAD), (e) missense variants predicted to be deleterious by SIFT and PolyPhen algorithms, and (f) variants associated with neuromuscular phenotypes.

Gene variants linked to neuromuscular diseases were first analyzed, followed by classification of potential variants by clinical geneticists and neurologists based on American College of Medical Genetics and Genomics guidelines (Straub et al., 2018). Variants were considered “novel” if previously unreported in PubMed database.

Cases were categorized as “solved” if pathogenic or likely pathogenic variants were identified in phenotype-associated genes with matching inheritance patterns. Cases with only one variant associated with recessive disease were designated as “partially solved”.

In a cohort of 22 patients with CM, WES successfully provided a genetic diagnosis in 15 patients (68.2%). Among these 15 solved cases, six patients (40.0%) had autosomal recessive variants, another six patients (40.0%) had de novo variants, and the remaining three patients (20.0%) had X-linked variants. In total, 15 distinct pathogenic variants were identified, each located in a different gene as follows: ZSWIM6, OCRL, ATRX, COL12A1, SYNE1, NIPBL, VARS, CNTNAP1, TUBB4A, OFD1, COL6A1, ATN1, NEB, AK9, and CHRNA1 (n = 1 variant per gene). This study demonstrates the use of WES in diagnosing genetically heterogeneous conditions such as CM by allowing the discovery of causal variants across many genes (Table 2).

WES enabled genetic diagnosis in 12 (57.1%) of 21 patients with MD. Among the 12 solved cases, autosomal recessive inheritance was identified in seven patients (58.3%), de novo variants in four patients (33.3%), and X-linked inheritance in one patient (8.4%). A total of 12 pathogenic variants were detected, each in a different gene as follows: ZSWIM6, SLC37A4, PLEC, GJB2, SH3TC2, ARIDG, ACSL4, MCM3AP, FUK, STXBP1, BPTF, and PRS6KA3 (n = 1 for each) (Table 3).

Whole exome sequencing revealed three pathogenic variants in the DMD gene in the MLPA-negative DMD group, confirming the diagnosis of DMD. These variants included two missense mutations and one nonsense mutation (Table 4).

In this study, WES revealed causative variants in 65.2% (30/46) of our cohort. The diagnostic yield was 100% (3/3) in the MLPA-negative DMD group, 57.1% (12/21) in the MD group, and 68.2% (15/22) in the CM group (Figure 2). The most commonly mutated gene was ZSWIM6, implicated in neurodevelopmental disorder with movement abnormalities, abnormal gait, and autistic features. Other recurrently affected genes included DMD in the MD group and SYNE1, ATRX, and COL6A1 in the CM group. The key clinical and molecular findings of the genetically diagnosed patients are summarized in Table 2, Table 3 and Table 4.

In our cohort, we identified several de novo mutations in patients whose genetic findings revealed a diagnosis different from the initial clinical suspicion of a muscular disorder. Specifically, in the congenital myopathies group, 6 patients (27.3%) harbored de novo variants in the following genes: ZSWIM6, OCRL, ATRX, SYNE1, NIPBL, and TUBB4A. In the muscular dystrophies group, 4 patients (19.0%) had de novo variants in the genes ARIDG, MCM3AP, STXBP1, and BPTF.

The modes of inheritance observed for the identified mutations included autosomal dominant, autosomal recessive, and X-linked inheritance patterns. We found variants in 9 genes associated with autosomal dominant conditions (COL12A1, NIPBL, TUBB4A, SYNE1, ARIDG, BPTF, STXBP1, ZSWIM6, GJB2), 13 genes linked to autosomal recessive disorders (VARS, CNTNAP1, COL6A1, NEB, AK9, CHRNA1, ATN1, SLC37A4, PLEC, SH3TC2, MCM3AP, FUK, COL12A1), and 6 genes related to X-linked diseases (OCRL, ATRX, OFD1, ACSL4, PRS6KA3, DMD).

Out of the 30 patients with a genetic diagnosis, 10 patients (33.3%) had findings consistent with their initial clinical diagnosis of a neuromuscular disorder. However, in 20 patients (66.6%), the genetic testing revealed a different diagnosis, such as neurodevelopmental disorders (e.g., ZSWIM6-related disorder), deafness (GJB2-related), or Lowe syndrome (OCRL-related). These findings underscore the importance of genetic testing in refining clinical diagnoses and highlight the potential for phenotypic overlap between neuromuscular disorders and other neurological conditions.

To assess the potential functional impact of the identified variants, we performed a conservation analysis using the University of California Santa Cruz (UCSC) Genome Browser and the Conserved Domains Database (CDD) from the National Center for Biotechnology Information (NCBI). The conservation status of each variant is summarized in Supplementary Tables S1–S3. In the CM group, 60% (9/15) of the variants were located in highly conserved regions, while 20% (3/15) were in moderately conserved regions, and 13.3% (2/15) were in poorly conserved regions. One variant (OFD1 c.2del) could not be assessed for conservation as it affects the start codon. In the MD group, 50% (6/12) of the variants were in highly conserved regions, 41.7% (5/12) were in moderately conserved regions, and 8.3% (1/12) were in poorly conserved regions. All three variants identified in the MLPA-negative DMD group were located in highly conserved regions of the DMD gene. Overall, 60% (18/30) of the identified variants were found in highly conserved regions, suggesting that they may have a significant impact on protein function and disease pathogenesis.

To provide context for our findings, we reviewed the literature for previously reported mutations in the genes identified in our study (Supplementary Table S4). We found that several of the variants we detected had been described in earlier studies, providing further evidence for their pathogenicity. For example, the DMD gene variants c.7354G>T, c.7993A>G, and c.5190G>C, which we found in the MLPA-negative DMD group, have been reported in patients with Duchenne muscular dystrophy (Flanigan et al., 2009; Tuffery-Giraud et al., 2009; Okubo et al., 2016). Similarly, the PLEC gene variants c.9343C>T and c.13192G>A, identified in the MD group, have been associated with epidermolysis bullosa simplex with muscular dystrophies (Natsuga et al., 2010; Winter and Wiche, 2013). In the CM group, we identified the COL12A1 variant c.5894G>A and the COL6A1 variant c.850G>A, which have been previously reported in patients with Bethlem myopathy or Ullrich congenital muscular dystrophy (Lampe and Bushby, 2005; Butterfield et al., 2013; Naghipoor et al., 2023).