The role of disease-associated short tandem repeats in amyotrophic lateral sclerosis

Short tandem repeats (STRs) are recognized contributors to various neurodegenerative disorders, with evidence supporting genetic pleiotropy among these STRs. Multiple STRs have been associated with amyotrophic lateral sclerosis (ALS), although the strength of evidence supporting each association varies. To establish the role of disease-associated repeat expansions as pleiotropic risk factors in ALS susceptibility and progression, we genotyped a panel of 39 STRs, known to cause neurological diseases, within Project MinE in 6519 patients and 2412 controls, utilizing 100 and 150 bp short-read sequencing technology. Pathogenic allele frequencies were compared to those in a control cohort comprising 4930 Genome Aggregation Database (gnomAD) genomes. Repeat sizes and motif changes were detected using ExpansionHunter and ExpansionHunter Denovo. We developed a model to predict genotyping failures in STRs and established a best-practice protocol for assessing the accuracy of STR genotyping in short-read sequencing data. Following our genotyping assessment, 11 out of the 39 STRs exhibited insufficient genotyping accuracy, warranting caution in studying these STRs using these tools in combination with short-read sequencing. Furthermore, the observed differences in STR genotyping accuracy across studies applying different sequencing technologies and genotyping tools in control cohorts highlight the importance of a carefully designed experimental setup when interpreting potential disease-associated STR findings. Pathogenic C9orf72 and premutated ATXN2 expansions were confirmed to be significantly associated with ALS susceptibility. Additionally, pathogenic C9orf72 expansions were significantly associated with reduced mean ALS survival by 11.5 months and an earlier mean age at onset by 2.4 years. Premutation expansions in ATXN1 showed a nominally significant association with ALS susceptibility, while pathogenic expansions in NIPA1 displayed a nominally significant association with ALS survival. Previously reported ALS-associated pleiotropy in HTT and STMN2 could not be confirmed. Motif changes were identified in BEAN1, RFC1, ATXN8, C9orf72, DAB1, FXN and SAMD12; however, none of the motif changes were linked to ALS. Re-evaluation of clinical data from patients with ALS and a repeat expansion typically associated with another disease revealed that 7% of these patients’ diagnoses had to be reclassified to the disease associated with the repeat expansion (e.g. Kennedy’s disease or spinocerebellar ataxia). This underscores the value of broad STR screening in neurodegenerative cases. Pathogenic and premutation STRs were also found in controls in unexpected high frequencies, suggesting reduced penetrance or underdiagnosis, and highlighting the need for caution when interpreting genetic associations with disease without a proper control cohort.