Re-classification of hypertrophic cardiomyopathy genetic variants in patients from the sarcomeric human cardiomyopathy registry (SHaRe)

Background Genetic testing is a powerful tool for identifying family members at risk of developing disease, however it depends on accurate variant interpretation. Classifications of genetic variants reflect our knowledge at a point in time, influenced by many factors such as public availability of case data, evolving variant classification specifications and gene validity. Purpose We report re-classification of hypertrophic cardiomyopathy (HCM) variants from a large multi-centre international registry of genetically characterised HCM patients (Sarcomeric Human Cardiomyopathy Registry; SHaRe), to improve validity of variant classifications. Methods SHaRe participants were clinically evaluated and enrolled at international sites. Data were entered from the genetic report performed between 1989 and 2020, including the laboratory classification. All variants were systematically re-evaluated. 68% underwent an expedited curation after computational re-annotation confirmed the absence of new evidence sufficient to change classification, while 21% underwent full manual curation using American College of Medical Genetics and Genomics (ACMG) criteria. Results Of 10,413 HCM patients, 4700 (45%) had a variant identified in a clinically significant HCM gene based on testing done at the time, 3314 (71%) with a P/LP variant and 870 (19%) with a variant of uncertain significance (VUS). There were 1274 unique variants identified: 867 (68%) underwent expedited curation, 322 (25%) retained their classification of P/LP, 89 (7%) as B/LB and 456 (36%) as VUS. 270 (21%) were manually curated and 116 are pending. 125/1274 (10% of total variants) had a change in classification. There were 101 (8%) downgrades, including 84 classifications from LP/P to VUS (majority were MYH7 missense variants with insufficient evidence); 16 (1%) VUS to LB/B; and 1 LP to LB. There were 24 upgrades from VUS to LP/P (2%) for 69 patients (mostly missense variants in MYBPC3 and MYH7 with additional published cases and functional studies). Ten (<1%) variants had incongruent classifications within SHaRe. Four (0.3%) variants–MYBPC3 R817G, R820W, C1266Y, and MYH7 I511T–reached LP with inclusion of internal SHaRe probands, which would otherwise remain VUS with publicly available proband counts. Conclusion 125 (10%) variants seen in 203 individuals had a clinically significant change in classification upon re-evaluation of genetic variants found in a population of HCM patients with deep genetic and clinical phenotyping. Evidence informing variant classification evolves over time and periodic re-evaluation is critical. Public sharing of case data remains a key factor in reclassification.