The role of CTNNA1 truncating variants in hereditary diffuse gastric cancer (HDGC)

BackgroundRare CTNNA1/αE-catenin truncating variants were found in HDGC patients, however, disease spectrum and variant-type causality remains understudied. We aim to explore genotype-phenotype associations and understand the molecular pathways causing CTNNA1-driven diffuse gastric cancer (DGC).MethodsUsing a clinical database of 364 CTNNA1 variant carrier families (1388 individuals), we analyzed genotype–phenotype associations with multivariable logistic regression. Variants functional impact was assessed with in vitro/in vivo models. Transcriptomic profile of 11 DGC and normal stomachs was analyzed.ResultsFrom 71 European CTNNA1 carrier families, 26 carried truncating variants (21/26 met HDGC criteria). Early onset DGC was significantly more likely to occur in truncating carrier families than in those with missense variants (OR=8.33; 95%CI [3.125-25]; p<0.001). Lobular breast cancer (LBC) followed the same trend in truncating carriers (OR=4.76; 95%CI [0.98-50]; p=0.053), although non significant. In an extended series of 271 families carrying truncating variants, 32 (12%) families fulfilled HDGC criteria, and LBC was more frequent than DGC (20 vs 19 cases). Our CRISPR/Cas9 edited gastric cancer cell line bearing a CTNNA1 truncating showed complete CTNNA1/αE-catenin loss. Nonsense Mediated mRNA Decay (NMD) blockade increased CTNNA1 mRNA expression ∼13-fold, recovering to wild-type (WT) levels. In our Drosophila α-cat knockout (KO), organ development/lethality was rescued with overexpression of human WT/missense αE-catenin, but not with a truncated αE-catenin. DGC transcriptomic analysis from carriers’ revealed 67 upregulated genes in tumors vs normal, including HIF1α and PIK3R3, two known cancer therapy targets and candidates for drug repurposing in CTNNA1-driven DGC.ConclusionsDGC is 8-fold more likely to occur in carriers of truncating than missense variants, and data on LBC is still scarce. We highlight NMD as a prime mechanism for CTNNA1 truncated transcripts degradation and created an in vivo model to assess variants functional impact. CTNNA1-associated DGC overexpress a set of molecules worth exploring as therapy targets.Legal entity responsible for the studyC. Oliveira.FundingPortuguese Foundation for Science and Technology (funded S. Lobo PhD scholarship 2020.05773.BD and the LEGOH Project PTDC/BTM-TEC/6706/2020). ERN-GENTURIS (No. 739547).DisclosureJ. Herrera-Mullar, R. Karam: Financial Interests, Personal, Full or part-time Employment, NA: Ambrygenetics. P.R. Benusiglio: Financial Interests, Institutional, Financially compensated role, NA: AstraZeneca, BMS, Merck pharmaceuticals. All other authors have declared no conflicts of interest.