Solving Undiagnosed Neuromuscular Disorders: Uncovering Causative Genotypes Beyond Known Genes and Beyond the Exome

Abstract

Neuromuscular disorders (NMDs) are a group of inherited diseases that are highly heterogeneous both clinically and genetically. The global prevalence is approximately 1 in 1,000. To date, ~700 genes have been identified to be associated with NMDs. The genetic cause of inherited NMDs in many families remains elusive even after extensive investigation, indicating the possibility of novel genes and genetic mechanisms that have yet to be explored. In addition, lack of a definitive diagnosis can interfere with disease prognosis, classification, and therapeutic options. This thesis aimed to identify the genetic cause of unsolved NMD cases by combining systematic genomic reanalysis with detailed phenotype assessment and functional studies. This would provide patients with a confirmed diagnosis, improved disease management, and access to therapy or inclusion in therapeutic trials.

As part of this work, pre-existing NGS datasets from 101 previously unsolved NMD families were systematically reanalysed using updated bioinformatic approaches and phenotype-driven variant prioritization strategies. This led to a diagnostic yield of 16.83%, with causative variants identified in 17 families. In five cases, intronic variants in known NMD genes (COL6A3, SGCA, DOK7, DYSF, CHRND) were considered causative following in silico predictions and careful correlation with the phenotype. One case had an extended phenotype (PTPN11), and one case had a dual diagnosis (MYH2, KIF21A). A novel ATP2A2 missense variant was identified in two unrelated families, establishing ATP2A2 as a new NMD gene.

This thesis also identified and characterized a distinct pathogenic mechanism underlying congenital myasthenic syndrome (CMS) caused by variants in the intracellular M3-M4 cytoplasmic loops of acetylcholine receptor (AChR) subunits encoded by CHRND and CHRNA1. Unlike classical AChR-related CMS, these variants showed preserved surface expression and largely normal channel kinetics in conventional functional assays. However, agrin-induced clustering assays in genetically modified C2C12 myotubes demonstrated markedly impaired AChR clustering, establishing defective postsynaptic receptor aggregation as the primary disease mechanism.

In addition, this work established a novel association between a heterozygous ATP2A2 missense variant and dominantly inherited recurrent rhabdomyolysis. Functional studies showed impaired SERCA2-mediated calcium reuptake and abnormal intracellular calcium homeostasis in skeletal muscle, supporting the pathogenicity of the identified variant.

Overall, this thesis showed that research-based reanalysis of pre-existing NGS data improved the diagnostic yield of previously unsolved cases consistent with previous literature, reinforcing the utility of in-depth, phenotype-driven reanalysis with expert review. The findings highlight the evolving nature of genomic reanalysis and interpretation, demonstrating that previously unsolved cases may become diagnosable as our knowledge of disease genes, variant interpretation and analytical tools continues to improve.