Improving genetic diagnosis by disease-specific, ACMG/AMP variant interpretation guidelines for hearing loss

Abstract

The 2018 Hearing Loss Expert Panel (HL-EP)-specific guidelines specified from the universal 2015 ACMG/AMP guidelines are proposed to be used in genetic HL, which prompted this study. A genetic HL cohort comprising 135 unrelated probands with available exome sequencing data was established. Overall, 169 variants were prioritized as candidates and interpreted using the 2015 ACMG/AMP and 2018 HL-EP guidelines. Changes in rule application and variant classification between the guidelines were compared. The concordance rate of variant classification of each variant between the guidelines was 71.60%, with significant difference. The proportion of pathogenic variants increased from 13.02% (2015) to 29.59% (2018). Variant classifications of autosomal recessive (AR) variants that previously belonged to VUS or likely pathogenic in the 2015 guidelines were changed toward pathogenic in the 2018 guidelines more frequently than those of autosomal dominant variants (29.17% vs. 6.38%, P = 0.005). Stratification of the PM3 and PP1 rules in the 2018 guidelines led to more substantial escalation than that in the 2015 guidelines. We compared the disease-specific guidelines (2018) with the universal guidelines (2015) using real-world data. Owing to the sophistication of case-level data, the HL-specific guidelines have more explicitly classified AR variants toward "likely pathogenic" or "pathogenic", serving as potential references for other recessive genetic diseases.

Figure 1

The study design of the present study. The 169 variants of 135 genetically unrelated sensorineural hearing loss probands were classified based on the 2015 ACMG/AMP and 2018 HL-EP guidelines. The variants classification between the 2015 ACMG/AMP and 2018 HL-EP guidelines were compared.

Figure 2

Comparison of Bayesian values between the 2015 ACMG/AMP and 2018 HL-EP guidelines.

Figure 3

Frequency of criteria applied in the 2015 ACMG/AMP and 2018 HL-EP guidelines. The most commonly applied rule in the two guidelines was PM2 and the criterion of PM3 yielded the most drastic difference between the 2015 ACMG/AMP and 2018 HL-EP guidelines.