P-546 The structure of genome variations at CNV level differs in recurrent and sporadic pregnancy loss

Abstract

Abstract Study question Whether the CNV load differs in fetal tissues of miscarriages from families with recurrent and sporadic pregnancy loss? Summary answer In RPL group duplications are more common than deletions, de novo variants than inherited ones. Pathogenic CNVs (dup16p11.2, del22q13, dupXq28) were found in both groups. What is known already Data on CNVs in tissues of miscarriages can be found in more than 24 articles. Altogether 7732 samples were analyzed by aCGH and NGS; 988 CNVs of uncertain clinical significance were found in 636 samples (8.2%). The origin was determined for 130 variants: 85 maternal (65.4%), 34 paternal (26.2%), 11 de novo (8.4%) CNVs. No aberration unambiguously associated with miscarriage was found so far. Comparative analysis of CNVs in family trios with recurrent and sporadic miscarriage may help to characterize structural genome variations associated with embryo lethality both from embryo and parental side. Study design, size, duration Identification of structural genome variations (SGV) at the CNV level was performed by aCGH-analysis for family trios: mother, father and fetal. Characteristics of SGV associated with embryo lethality included such parameters as CNV type (deletion/duplication) and its origin. It was assumed that the structure of genomic variability may differ in recurrent pregnancy loss (RPL) and sporadic pregnancy loss (SPL) both in embryonic tissues and in parents. Participants/materials, setting, methods Eleven family trios (7 RPL and 4 SPL) were included in the research. Standard G-banding was performed for all abortion samples to exclude aneuploidy. STR-haplotyping for the SRY gene was used to exclude maternal cell contamination. Finally all samples were analyzed for CNVs using SurePrint G3 Human CGH+SNP 4 × 180K microarrays (Agilent Technologies). Average gestation age was 7.9 weeks. Average maternal and paternal age was 30.8 and 30.5 years, respectively. Main results and the role of chance In seven and four embryos from RPL and SPL groups 110 and 60 CNVs were identified, respectively. In RPL, there was a tendency towards an increase in the frequency of duplications over deletions (72:38) in comparison to SPL (36:24) (p = 0.5). Duplications prevailed over deletions in five of seven embryos from RPL group; while there were two cases of each combination in SPL. One can suggest that some SPL cases can move into RPL group over time. De novo variants prevailed over inherited in the RPL group (76:43) and inherited prevailed in the SPL (32:28) (p < 0.05). This result allows to suggest that in cases with RPL there may be some common genetic/epigenetic factor predisposing to CNV generation either in primordial germ cells or during early blastomere cleavage. Among the inherited variants in the RPL group there were 24 maternal and 8 paternal CNVs, while in the SPL – 18 and 14 (p = 0.1). In both groups pathogenic CNVs associated with syndromes characterized by neurodevelopmental disorders (NDD) in children were found (del22q13, dupXq28 – in RPL group, dup16p11.2 – two cases in SPL). Presumably, such aberrations may explain a high risk of NDD in a child of a woman with RPL/SPL (PMID:30058166, PMID:33580539). Limitations, reasons for caution No one CNV unambiguously associated with pregnancy loss has been described so far. Possibly, total structural genome variations level should be considered as pathogenic factor for embryo development. There is no adequate algorithm of clinical significance CNV interpretation for the embryonic period. Wider implications of the findings Prenatal or preimplantation genetic testing should be considered if one of the parents is a carrier of CNV associated with microdeletion/microduplication syndrome. This study was supported by the Russian Science Foundation № 21-65-00017, https://rscf.ru/project/21-65-00017/. Trial registration number Russian Science Foundation № 21-65-00017