Long-read sequencing opens new possibilities for diagnosing rare diseases
Background and Challenges: The Labyrinth of Rare Disease Diagnosis
'Rare diseases affect less than 1% of the total patient population, but many cases experience delayed diagnosis due to the difficulty in identifying the exact genetic cause. In particular, traditional single nucleotide variant (SNV) testing alone struggles to detect complex structural variants or repeat expansions. Many patients remain without a confirmed causative gene, and are limited to symptom management, which poses a significant barrier to the development of therapeutics. Existing short-read sequencing has limitations in accurately detecting splicing errors at exon boundaries, which reduces the diagnostic rate. In this context, an imbalance in DNA non-homologous end joining (NHEJ) and homologous recombination (HR) pathways further complicates variant detection.'
Research Methods and Findings: The Power of Long-read Sequencing
'The research team utilized PacBio HiFi and Oxford Nanopore long-read sequencing to analyze the whole genomes of 500 patients with rare diseases. Long-read sequencing reads an average of 15-20kb, allowing for the simultaneous detection of complex repeat regions and large insertions/deletions, and accurately determining the copy number of genes like SMN2. Furthermore, structural variants such as large translocations in the COL1A1 gene were detected with 30% higher sensitivity compared to existing methods. In this process, DNA methylation patterns were analyzed concurrently to reveal the association between variants and transcriptional repression. As a result, the diagnostic rate increased significantly from 45% to 78%, and 12 additional causative genes were identified.'
Future Implications or Prospects: Application and Expansion to Clinical Practice
'Long-read sequencing is now ready to be directly integrated into clinical genomic diagnostic pipelines. With the cost reduced to approximately $1,200 per year, insurance coverage is also gradually expanding. In fact, 12 major medical institutions in the United States have launched a pilot program, which is greatly assisting in the development of patient-specific treatment plans. Furthermore, this technology can be applied to complex diseases such as neurodegenerative diseases or immunodeficiency, and is expected to increase the number of diagnosable rare diseases by more than double within the next five years. Ultimately, it will be possible to provide rapid and accurate genetic information to rare disease patients worldwide, significantly improving their quality of life.'
New England Journal of Medicine, Ahead of Print.
This research aims to address the diagnostic delays and difficulties in treatment selection experienced by rare disease patients who lack a confirmed genetic cause. Previously, short-read sequencing failed to detect structural variants and repeat expansions, resulting in a diagnostic success rate of only 45%, leaving many patients without a definitive diagnosis. By introducing a new approach, long-read sequencing, which reads the entire genome at once, complex variants are detected with more than 30% higher sensitivity. As a result, the rare disease diagnostic market in the United States is expected to grow from $1 billion to $1.5 billion annually, and insurance coverage is also expanding, providing tangible benefits to both patients and healthcare institutions. In the future, the combination of more genomic data and AI-based interpretation is expected to increase diagnostic accuracy to over 90% and accelerate the development of personalized treatments.