Impact of genetic structural variants in factor XI deficiency: identification, accurate characterization, and inferred mechanism by long-read sequencing

• DOI: 10.1016/j.jtha.2023.03.009
• Publication Date: 2023-03-20T23:08:15Z
• Authors: Belén de la Moren...; Ángeles Palomo; José Padilla; Laura Martín-Fern...; Juan José Rojo-Ca...; Rosa Cifuentes; Carlos Bravo-Pérez; Pedro Garrido-Rod...; Antonia Miñano; Ana María Rubio; Javier Pagán; María Llamas; Vicente Vicente; Francisco Vidal; María Luisa Lozano; Javier Corral; María Eugenia de ...
• Keywords: Identification

Congenital factor XI (FXI) deficiency is a probably underestimated coagulopathy that confers antithrombotic protection. Characterization of genetic defects in F11 is mainly focused on the identification of single-nucleotide variants and small insertion/deletions because they represent up to 99% of the alterations accounting for factor deficiency, with only 3 gross gene defects of structural variants (SVs) having been described.To identify and characterize the SVs affecting F11.The study was performed in 93 unrelated subjects with FXI deficiency recruited in Spanish hospitals over a period of 25 years (1997-2022). F11 was analyzed by next-generation sequencing, multiplex ligand probe amplification, and long-read sequencing.Our study identified 30 different genetic variants. Interestingly, we found 3 SVs, all heterozygous: a complex duplication affecting exons 8 and 9, a tandem duplication of exon 14, and a large deletion affecting the whole gene. Nucleotide resolution obtained by long-read sequencing revealed Alu repetitive elements involved in all breakpoints. The large deletion was probably generated de novo in the paternal allele during gametogenesis, and despite affecting 30 additional genes, no syndromic features were described.SVs may account for a high proportion of F11 genetic defects implicated in the molecular pathology of congenital FXI deficiency. These SVs, likely caused by a nonallelic homologous recombination involving repetitive elements, are heterogeneous in both type and length and may be de novo. These data support the inclusion of methods to detect SVs in this disorder, with long-read-based methods being the most appropriate because they detect all SVs and achieve adequate nucleotide resolution.