Utilization of exome sequencing for muscular disorders in Thai pediatric patients: diagnostic yield and mutational spectrum

We performed ES in 43 Thai pediatric patients with MD or CM from 42 unrelated families. Their clinical features and WES data were analyzed simultaneously, enabling the correlation of the ES results with other diagnostic modalities. The overall diagnostic yield for ES was 70% (30/43). The most common MD disease genes were DMD (30%; 8/27), LMNA (15%; 4/27) and COL6A1 (11%; 3/27), while the most prevalent CM disease gene was RYR1. These gene profiles are like those reported by other studies in pediatric populations7.8.

Earlier research on ES in pediatric neuromuscular patients produced diagnostic yields ranging from 37 to 65%8,9,10,11. The high diagnostic rate for ES in the current investigation (70%) was probably due to several reasons. The patients referred to the 2 tertiary-care centers in this study were highly selective and strongly suspected to have genetic disorders. Most of the patients in the cohort had infantile-onset weakness. This study also included the MLPA-negative DMD subgroup, in which the diagnostic rate was high. In addition, it was performed as an early diagnostic step. Unlike the practice of many other countries, most of the patients in this cohort did not first undergo extensive serial gene or targeted panel testing. Consequently, ES would have shown a higher level of positive results than otherwise, making ES a contributing factor to the diagnostic yield.

A limitation of using ES early is that it cannot detect molecular pathologies in congenital myotonic dystrophy type 1 (repeat expansion in the DMPK gene) and FSHD (contraction of D4Z4 repeats on chromosome 4q35). The 2 disorders are the most common hereditary myopathies after DMD in adult, accounting for 10–35% and 6–10% of hereditary myopathies, respectively12:13. This study excluded patients with clinically suspected FSHD without molecular confirmation. Although none of the patients in our cohort had phenotypes specific to congenital myotonic dystrophy type I, infantile and childhood-onset congenital forms can be indistinguishable from other myopathies. By excluding congenital myotonic dystrophy type I and FSHD, mutation-targeted testing could increase the apparent diagnostic yield of ES.

ES changed the management of 73% (22/30) of the diagnosed cases (see supplementary Table S1 online). Muscle biopsies were unnecessary once the molecular diagnosis was made. The confirmation of DMD point mutations allowed for steroid treatment, while identifying the nonsense mutation made Patient DMD8 a candidate for premature stop codon read-through treatment (Ataluren)6. The other patient in which ES unravelled the diagnosis and allowed for choice of medication was the patient with COLQ– congenital myasthenic syndrome14. The findings of this study highlight the utility of ES in ending diagnostic odyssey and permitting precision treatment.

In the MPLA-negative DMD group, the vast majority of cases were solved. One patient in the MLPA-negative DMD group who had clinical findings and muscle histopathology consistent with dystrophinopathy but with no variant found was believed to have a regulatory/promoter or deep intronic alteration. In previous studies on patients whose clinical and histopathological data suggested DMD, ES led to alternative diagnoses of inherited MD15. We found no disease-causing variants in other genes. In the future, targeted NGS could be used for patients with MLPA-negative DMD in our setting.

Although our three patients with COL6A1-associated muscular disorders (patients MD1-MD3) were found to harbor only one variant, there is a possibility that they are actually compound heterozygous with the other variant not identified by our methods.

Patients MD6 and MD7 had delayed motor milestones, proximal muscle weakness, and diffuse white matter hypersignal intensity on brain MRI. Muscle biopsies showed faint and absent merosin staining in Patients MD6 and MD7, respectively. The clinical diagnosis was autosomal recessive merosin-deficient congenital MD (OMIM#607855)16. In patient MD6, trio-ES identified a previously reported heterozygous splice-site c.283+1G>C variant in LAMA2. In patient MD7, singleton-ES identified a novel heterozygous frameshift variant, c.2718delT (p.Phe906LeufsTer169). The patient’s clinical, histopathological and ES data led to a diagnosis of autosomal recessive merosin-deficient congenital MD. This diagnosis suggested that a second variant in trans was missed in each case. Mutations that have structural variants or variants in noncoding regions cannot be identified by ES. Both patients will undergo long-read genome sequencing.

The phenotypes of TTN-related myopathy in this cohort were diverse. TTN variants were found in both the CM and MD groups. Patient MD14, categorized into the MD group, had arthrogryposis multiplex congenita, finger/elbow flexion contractures and proximal muscle weakness. A muscle biopsy showed nonspecific myopathic changes and faint collagen VI staining. The clinical characteristics of patient MD14 mimicked collagenopathy. Patient CM9 had perinatal onset hypotonia with weakness, respiratory failure and cardiomyopathy, and a muscle biopsy revealed centrally located nuclei and fibre-type disproportion. Neither patient MD14 nor patient CM9 had ptosis or ophthalmoplegia. IT identified TTN compound heterozygous truncating and missense variants in both patients. Recent evidence has shown that missense mutations associated with the disease exert their effects when they occur with a truncating mutation. Previously reported cases with TTN compound heterozygous truncating and missense variants typically presented at birth. Their clinical courses were characterized by variable progression of weakness, contractures, scoliosis and respiratory symptoms but spared the extraocular muscles17.

A likely pathogenic variant was found in a gene causing the phenotype beyond the clinical features described in this study. Patient MD15 presented with progressive muscle weakness, areflexia and elevated serum CPK with a family history of a deceased male sibling with similar clinical findings. Muscle biopsies from Patient MD15 and his male sibling showed vacuolated muscle fibers with granular eosinophilic and amphophilic materials, cyclooxygenase (COX)-negative fibers and numerous abnormal mitochondria on electron microscopy. These findings suggested mitochondrial defects. Causative variants were not identified in nuclear- or mitochondrial-related genes. Trio-ES analysis revealed a homozygous missense c.616G>A (Gly206Ser) variant in the FUS gene. This variant has been reported in a heterozygous state in adult patients with familial autosomal dominant amyotrophic lateral sclerosis (OMIM#608030)18.

To our knowledge, there has been only 1 report of a homozygous FUS variant in a family with adult-onset amyotrophic lateral sclerosis with an autosomal recessive pattern of inheritance19. The missense c.616G>A (Gly206Ser) variant is rare, with a gnomAD allele frequency of 0.0000649. In addition, no homozygotes were found in the healthy population database (https://gnomad.broadinstitute.org/). The clinical presentation of patient MD15 did not support a diagnosis of motor neuron disease. Therefore, the c.616G>A (Gly206Ser) variant in FUS is classified as likely pathogenic but not playing an etiological role in Patient MD15’s phenotype. The clinical significance of discovering this variant would be to provide appropriate genetic counseling about the risk of a currently untreatable disease for asymptomatic parents in their 30s.

Increasing evidence shows that NGS technology, including targeted NGS and ES, has high clinical utility and saves time and costs7. However, there are currently no universally accepted guidelines for genetic testing in pediatric neuromuscular patients. In our tertiary-care setting, we employed ES since the clinical presentation in most patients did not point to a specific disease except in the MLPA-negative DMD group. In the DMD group, targeted NGS may be an option in the future. With other patients, limited gene testing carries the risk of missing disorders beyond provisional diagnosis. The cost-effectiveness of using ES early in our setting remains to be explored.

Leave a Reply

%d bloggers like this: