Study of the impact of ClinGen Revisions on ACMG/AMP variant semi-automatic classification for Rare Diseases diagnosis

• Published in: Clinica chimica acta 2025-01, Vol.566, p.120065, Article 120065
• Main Authors: Rius, Ana, Aguirre, Nicolas, Erra, Lorenzo, Brunello, Franco Gino, Biagioli, German, Zaiat, Jonathan, Marti, Marcelo A.
• Format: Article
• Language: English
• Subjects: ACMG/AMP guidelines; Automation; ClinGen; Clinical genetics; Genetic Testing - methods; Genetic Variation; Humans; Rare Diseases; Rare Diseases - classification; Rare Diseases - diagnosis; Rare Diseases - genetics; Variant interpretation
• ISSN: 0009-8981; 1873-3492; 1873-3492
• DOI: 10.1016/j.cca.2024.120065

•Comparative analysis of several genetic variant classification using ClinGen-Rev and ACMG-2015 semi-automatic protocols.•ClinGen-Rev accuracy improved to 89.2 %, a significant increase from ACMG-2015′s 65.6 %.•ClinGen-Rev reclassification reduces VUS by 8 %, refining prioritization of actionable variants.•Key improvements driven by allele frequency and molecular effect labels, optimizing clinical decisions.•Discussion of future directions for refining non-coding variant classification and further protocol development. With the rapid development of massive sequencing technologies, the analysis of genetic variants for clinical diagnosis has exponentially escalated, particularly in the context of Rare Diseases (RDs). Diagnosing them involves identifying the genetic variants responsible for the underlying pathology development. In 2015, the American College of Medical Genetics (ACMG) established a set of recommendations to assess the evidence associated with each variant, aiming to achieve a standardized five tier classification. Over the past 5 years, ClinGen, the NIH-funded Clinical Genome Resource, has reviewed these criteria in order to make variant classification a more reproducible and rigorous process. This paper examines the impact of ClinGen-Rev modifications on variant classification, comparing them with the ACMG-2015 original recommendations. After analyzing sets of genetic variants, extracted from VCFs samples, using both criteria, we observed a change in 8.0 % of the clinical verdicts for these variants. ClinGen-Rev modifications correctly categorized 89.2 % of the curated variants, representing a significant improvement compared to the 65.6 % achieved by ACMG-2015. We also analyzed the modifications impact in a real like clinical setting, showing a significant overall reduction of VUS variants and thus potential reduction in analysis time. Finally, we discuss the underlying reasons for the most relevant changes in terms of specific labels and present their implications on the prioritization and selection process of variants, identifying some recommendations of key significant importance.