Abstract 14451: High Agreement Between a Cardiology Specific Scoring System and the ACMG Criteria for Genetic Variant Interpretation and Classification
Introduction: The assessment of pathogenicity of genetic sequence variants is probabilistic and remains a challenge. Recently a consensus recommendation for genetic variant interpretation and classification has been published by the American college of medical genetics (ACMG). We evaluated the agreement between these recommendations and a previously published scoring system that was developed specifically for cardiomyopathies and primary electrical diseases by the Amsterdam group (Amsterdam criteria = AC).
Methods: We included rare genetic variants (prevalence <0.1% in ExAC) that were identified in the first 524 probands who were sequenced using a cardiology specific gene panel (75 genes) at our institution. These variants were classified by both the AC and ACMG criteria using a 5-tier-classification system: class 1 (benign), class 2 (likely benign), class 3 (variant of unknown significance), class 4 (likely pathogenic) or class 5 (pathogenic). The agreement between the two classification systems was evaluated by Krippendorffs α.
Results: In total, 253 variants were scored including different types of variation: 196 missense, 19 frameshift, 19 splice, 11 nonsense, 6 in-frame deletions or duplications, 1 synonymous and 1 stop loss variant. The figure summarizes the results of the variant assessment of the scoring systems. The α value was 0.81 (95% CI = 0.77-0.85), indicating good agreement. The classifications fully matched for 142 out of 253 variants (56%). The ACMG criteria resulted in more variants of unknown significance (49% vs 34%; p=0.0006). A transition from class 4 or 5 to class 1,2 or 3 or vice versa, which would influence medical decision making, was observed for 36 variants (14%).
Conclusions: In conclusion, we observed a good agreement between the AC and the ACMG criteria. In the absence of a real golden standard, we conclude that the ACMG criteria seem to be applicable in cardiogenetic diseases, but resulted in more variants of unknown significance.
Author Disclosures: T. Robyns: None. C. Kuiperi: None. R. Willems: None. J. Van Cleemput: None. D. Nuyens: None. J. Breckpot: None. G. Matthijs: None. A. Corveleyn: None.
- © 2016 by American Heart Association, Inc.