This article is a preprint and has not been peer-reviewed. It reports new medical research that has yet to be evaluated and so should not be used to guide clinical practice.
Background Diagnostic evaluation of rare genetic disorders continues to rely on multiple test modalities, despite the increasing use of short-read exome or genome sequencing as first-tier tests. Long-read genome sequencing (lrGS) has the potential to consolidate current standard-of-care (SoC) diagnostics into a single assay, but its accuracy and clinical utility in routine practice have not been established at large scale.
Methods We studied 1000 clinical samples, including 832 index cases, representative of one year of germline diagnostic testing at two tertiary centers. Samples underwent lrGS at approximately 30× coverage, followed by variant detection and interpretation. Diagnostic outcomes were compared with those obtained through SoC testing. Using these results, we modeled the effect of implementing lrGS as a first-tier test in a cohort of 15,150 index cases who underwent diagnostic testing during a single year.
Results The overall concordance between SoC and lrGS testing was 96.4%. lrGS identified clinically relevant findings that improved or refined genetic diagnoses in 3.4% of cases, largely through phasing for recessive disorders and the detection of novel disease-causing variants. In 0.2% of cases, SoC testing identified variants not detected by lrGS. Modeling of a generic lrGS-first diagnostic strategy demonstrated an estimated absolute increase in diagnostic yield of 2.5% across the full annual cohort.
Conclusions LrGS demonstrated high concordance with current diagnostic approaches and provided incremental diagnostic benefits. These findings support lrGS as a feasible and effective first-tier test for rare disease diagnostics.