Diagnosing missed cases of spinal muscular atrophy in genome, exome, and panel sequencing datasets

Genet Med. 2024 Dec 9:101336. doi: 10.1016/j.gim.2024.101336. Online ahead of print.

Abstract

Purpose: We set out to develop a publicly available tool that could accurately diagnose spinal muscular atrophy (SMA) in exome, genome or panel sequencing datasets aligned to a GRCh37, GRCh38, or T2T reference genome.

Methods: The SMA Finder algorithm detects the most common genetic causes of SMA by evaluating reads that overlap the c.840 position of the SMN1 and SMN2 paralogs. It uses these reads to determine whether an individual most likely has zero functional copies of SMN1.

Results: We developed SMA Finder and evaluated it on 16,626 exomes and 3,911 genomes from the Broad Institute Center for Mendelian Genomics, 1,157 exomes and 8,762 panel samples from Tartu University Hospital, and 198,868 exomes and 198,868 genomes from the UK Biobank. SMA Finder's false positive rate was below 1 in 200,000 samples, its positive predictive value was greater than 96%, and its true positive rate was 29 out of 29. Most of these SMA diagnoses had initially been clinically misdiagnosed as Limb-girdle muscular dystrophy (LGMD).

Conclusion: Our extensive evaluation of SMA Finder on exome, genome and panel sequencing samples found it to have nearly 100% accuracy and demonstrated its ability to reduce diagnostic delays, particularly in individuals with milder subtypes of SMA. Given this accuracy, the common misdiagnoses identified here, the widespread availability of clinical confirmatory testing for SMA, as well as the existence of treatment options, we propose that it is time to add SMN1 to the ACMG list of genes with reportable secondary findings after genome and exome sequencing.

Keywords: Diagnosis; Exome; Muscle Disease; SMA; Spinal Muscular Atrophy.