Skip to main content

Conference Program

Subpage Hero

Loading

Truncating Variants in APOB Are Not VUS When Screening for Familial Hypercholesterolemia

Clinical Genetics and Therapeutics
  • Primary Categories:
    • Clinical Genetics
  • Secondary Categories:
    • Clinical Genetics
Introduction:
Predicted truncating variants in APOB account for more than 10% of the Variants of Uncertain Significance (VUS) when screening the population for Familial Hypercholesterolemia. APOB codes for two isoforms: apoB48 and apoB100. ApoB100 is a core structural protein of VLDL, LDL and Lp(a) lipoproteins. Missense variants that disrupt apoB100’s binding capacity to LDLR cause Familial Hypercholesterolemia, whereas haploinsufficiency of APOB usually results in hypobetalipoproteinemia. 



ClinVar entries for APOB are difficult to parse because of these two phenotypes. This is also why clinical labs spend a lot of resources interpreting APOB truncating variants, and often return an uncertain result. The aim of this study was to investigate the clinical impact of APOB truncating variants in large unselected populations in the United States to streamline the interpretation of these variants in the context of Familial Hypercholesterolemia screening.

Methods:
We studied 135,733 participants from 7 health systems in the US that are part of the Helix Research Network. Phenotypic information was extracted from electronic health records and standardized in OMOP format. Three different phenotypes were evaluated for each individual: (i) the first diagnosis of elevated LDL-C, (ii) the first prescription of statin, and (iii) a  phenotype score developed by fitting a logistic regression model based on 264 phenotypes with known relevance to atherosclerotic disease. Positive controls were individuals with the established APOB pathogenic variants (p.Arg3527Gln and p.Arg3527Trp) as well as those with pathogenic variants in LDLR.

Results:
93 participants (0.07%) had a truncating variant in APOB, and 127 had one of the two APOB p.Arg3527 pathogenic variants. Participants with a truncating variant in APOB had fewer and later onset documented elevations in LDL-C compared to those without a variant (Hazard Ratio (HR) = 0.09 (0.01-0.6)), or to those with pathogenic variants. The same result was observed for the date of first statin prescription (HR = 0.14 (0.05-0.4). Aligning with these clinical observations, we find that truncating variants are associated with lower than average LDL-C levels, up through the last exon of the APOB gene by using the Power Window method (an optimized sliding window approach to identify key regions of a gene associated with a phenotype).

Conclusion:
This population study provides strong evidence that truncating variants in APOB do not cause Familial Hypercholesterolemia. For variant interpretation, these variants should not be returned as VUS, and we suggest to automatically call them as benign in the context of Familial Hypercholesterolemia. A similar approach can be used for all genes where haploinsufficiency is not the mechanism of disease. Clear evidence that pLOF variants are not pathogenic in these scenarios will reduce the burden of variant interpretation without decreasing the accuracy of the screening.

Agenda

Sponsors