Lessons Learned from a Decade of Parallel Sequencing in Neuromuscular Disorders: RNA splicing variant detection boosts diagnostic rate from 40% to 62%.
Sandra Cooper1
1Kids Neuroscience Centre, University of Sydney; Children's Hospital at Westmead; Children's Medical Research Institute
Objective:

To retrospectively review the diagnostic odyssey for 247 families ascertained over 2012-2022 with inherited neuromuscular conditions and triaged into research-led genomic testing.  To quantify diagnostic yield from parallel DNA sequencing modalities (targeted-capture gene panel, exome, genome), additional diagnoses secured through auxiliary RNA and protein investigations, to define an optimised diagnostic algorithm to maximise diagnostic yield in NMD.

Background:

Adoption of massively parallel sequencing (MPS) has transformed the diagnostic landscape for inherited neuromuscular disorders (NMDs). As genomic and transcriptomic technologies accelerate, there is a widening gap in clinical guidelines mapping their optimal integration into the diagnostic pathway.

Design/Methods:
Parallel sequencing of DNA via targeted-capture gene panel, exome, or genome. RNA studies and/or protein studies when indicated to enable provision of a molecular diagnosis. 
Results:

Integration of MPS and a suite of functional genomics investigations has diagnosed 62% (152/247) involving 67 known NMD genes, 6 novel disease genes and one novel class of pathogenic splicing variant.

45% (69/152) of total diagnoses required post-exome investigations.  Arrestingly, novel disease genes account for only 4% (6/152) of diagnoses whereas 39% of diagnoses (59/152) involve at least one splice-altering or structural variant. Protein studies quantifying a measurable decrement in the gene product supported 18% (27/152) diagnoses.

Conclusions:

The key lesson is simple: “go back to where you think the problem (gene) is and have a really good look.”  Expert clinical phenotyping remains of profound importance to target scrutiny of all rare, coding and non-coding variation in phenotypically consistent genes.  We emphasise diagnoses due to splice-altering or structural variants in known NMD genes outnumber diagnoses involving novel genes or phenotypes by 7-fold. With contemporary NMD gene lists, modern-day exomic sequencing (+/- 50 intronic bases) would identify 85% of causal variants.   We provide a diagnostic algorithm recommending optimised deployment of different investigations to most effectively and expeditiously reach the same diagnoses.

10.1212/WNL.0000000000202266