Population-scale repeat expansions elucidate disease risk and brain atrophy
Article Meta
Article Date: 08 April 2026
Article URL: https://www.nature.com/articles/s41586-026-10345-6
Article Title: Population-scale repeat expansions elucidate disease risk and brain atrophy
Article Image: None provided
Summary
This Nature study performs a population-scale survey of pathogenic short tandem repeat (STR) expansions by analysing repeat length at 37 disease-associated loci across 1,020,833 samples from seven diverse cohorts using short-read whole-exome (WES) and whole-genome (WGS) sequencing. Calls were made with GangSTR (WES) and ExpansionHunter (WGS), followed by strict quality control and validation (including PCR for HTT).
Key findings: carrier frequencies of pathogenic expansions at many loci (for example HTT, DMPK, C9orf72, CACNA1A) are higher than reported disease prevalences; there are clear ancestry-specific frequency differences; disease risk and penetrance increase progressively with repeat length and age; and crucially, carriers of expansions who were not diagnosed yet already showed disease-specific brain atrophy and, for HTT, markedly higher plasma neurofilament light chain (NfL).
Notable quantitative results: undiagnosed carriers of HTT expansions had ~22.1% lower putamen volume (and similar caudate/pallidum losses); CACNA1A pathogenic carriers showed ~24.6% cerebellar grey-matter loss; C9orf72 pathogenic carriers had ~9% thalamic volume loss. HTT pathogenic carriers also had ≈1.9-fold higher NfL.
Key Points
- Large-scale genotype-first survey: 1,020,833 WES samples plus 465,021 WGS samples analysed across 37 reliable disease-associated STR loci.
- Pathogenic-repeat carrier frequencies often exceed known disease prevalences (example: HTT carrier frequency higher than Huntington’s disease prevalence), implying lower population penetrance, later onset, underdiagnosis or methodological factors.
- Clear ancestry-specific enrichments and depletions for several loci (eg AR, ATXN2, CACNA1A, C9orf72, DMPK), matching known epidemiology in some cases and revealing novel patterns in others.
- Phenome-wide association (7,671 ICD-10 traits) replicated known locus–disease links (HTT–HD, DMPK–myotonic disorders, C9orf72–MND) and revealed repeat-length dependent increases in disease risk and penetrance.
- Brain imaging (n≈66k) and plasma proteomics (n≈49k) show locus-specific atrophy and biomarker changes before clinical diagnosis — strong evidence of a prediagnostic window for intervention.
- Short-read calling limitations remain (difficulty with very long repeats, complex motifs, UTR/intronic regions and interruptions), so results were QC-filtered and partly validated by PCR and manual review.
- Implication: population sequencing can inform epidemiology, penetrance estimates, biomarker development and earlier identification of people at risk for repeat-expansion diseases.
Context and relevance
STR expansions drive more than 70 neurological and neuromuscular disorders. Until recently, studies were mostly locus- or disease-centric and limited by assay cost or cohort size. This paper leverages biobank-scale WES/WGS data and modern STR callers to deliver a cross-locus, multi-ancestry view. The results matter for genetic epidemiology (carrier frequencies and ancestry differences), clinical genetics (penetrance, interruptions, somatic expansion effects) and clinical neurology (preclinical brain atrophy and NfL elevation). They also reinforce the need for improved genotyping (long-read validation, interruption-aware methods) and careful interpretation when translating population-level carrier rates into clinical risk.
Why should I read this?
Short version: they scanned over a million sequences and found repeat expansions are more common than we thought — and, crucially, that people carrying them often already show the same brain changes seen in disease before any diagnosis. If you care about neurodegeneration, genetic risk, biomarker development or population screening, this paper saves you hours of digging: big cohort, neat validation, clear implications for earlier detection and trial design.
Author note (punchy)
Punchy: this is high-impact, field-moving work. It changes how we think about penetrance and the preclinical window for repeat-expansion diseases — worth reading in full if you work on genetics, neurology or translational biomarker research.