Next-generation sequencing and comprehensive data reassessment in 263 adult patients with neuromuscular disorders: insights into the gray zone of molecular diagnoses

Abstract

Background: Neuromuscular disorders (NMDs) are heterogeneous conditions with a considerable fraction attributed to monogenic defects. Despite the advancements in genomic medicine, many patients remain without a diagnosis. Here, we investigate whether a comprehensive reassessment strategy improves the diagnostic outcomes.

Methods: We analyzed 263 patients with NMD phenotypes that underwent diagnostic exome or genome sequencing at our tertiary referral center between 2015 and 2023. We applied a comprehensive reassessment encompassing variant reclassification, re-phenotyping and NGS data reanalysis. Multivariable logistic regression was performed to identify predictive factors associated with a molecular diagnosis.

Results: Initially, a molecular diagnosis was identified in 53 cases (20%), while an additional 23 (9%) had findings of uncertain significance. Following comprehensive reassessment, the diagnostic yield increased to 23%, revealing 44 distinct monogenic etiologies. Reasons for newly obtained molecular diagnoses were variant reclassifications in 7 and NGS data reanalysis in 3 cases including one recently described disease-gene association (DNAJB4). Male sex reduced the odds of receiving a molecular diagnosis (OR 0.42; 95%CI 0.21-0.82), while a positive family history (OR 5.46; 95%CI 2.60-11.76) and a myopathy phenotype (OR 2.72; 95%CI 1.11-7.14) increased the likelihood. 7% were resolved through targeted genetic testing or classified as acquired etiologies.

Conclusion: Our findings reinforce the use of NGS in NMDs of suspected monogenic origin. We show that a comprehensive reassessment enhances diagnostic accuracy. However, one needs to be aware that genetic diagnoses are often made with uncertainty and can even be downgraded based on new evidence.

Keywords: Neuromuscular disease; Next-generation sequencing; Reanalysis.

Fig. 1

Overview of results after reassessment of NGS data. Panel A shows the change in classification of all patients after extensive reassessment including reclassification of all variants, reanalysis of all negative cases and extensive re-phenotyping and evaluation of all available clinical data including external genetic reports. Cases were regarded as solved if they had a (likely) pathogenic variant in a gene with an established gene-disease relationship compatible with the phenotype of the patient. In case of compound heterozygous variants both variants needed to be classified as (likely) pathogenic. Uncertain cases were patients where at least one of the variants in question was classified as a variant of uncertain significance (VUS). Cases were classified as having an acquired disease if NGS was negative and clinical and other diagnostic findings were compatible with a distinct non-monogenic disorder (see supplementary file 1 for a clinical description of all cases). Three patients were solved by external genetic testing after negative NGS. Panel B shows the inheritance pattern of all yielded diagnosis (excluding patients with VUS) and the type of all variants classified as (likely) pathogenic. Other variants included 1 intronic variant, 1 splice region (near-splice) variant and a GCG repeat expansion detected by WES in ABPN1. Panel C shows the classification of cases by disease (phenotype) groups after reassessment as described for panel A. Panel D shows a list of all genes with the number of detected (likely) pathogenic variants and VUS for each gene. ACMG denotes American College of Medical Genetics and Genomics, CNV copy number variation, HSP hereditary spastic paraplegia, Mito mitochondrial disease and MND motor neuron disease

Fig. 2

Forest plot. Odds for receiving a genetic diagnosis after reclassification and reanalysis of NGS data. Odds ratios are shown for per 1-unit increase for covariates included in the multiple logistic regression model. NGS denotes next generation sequencing, WES whole exome sequencing and WGS whole genome sequencing