Confirmation and pathogenicity of small copy number variations incidentally detected via a targeted next-generation sequencing-based preimplantation genetic testing for aneuploidy platform.

OBJECTIVE

To evaluate the technical accuracy, inheritance, and pathogenicity of small copy number variants (CNVs) detected by a targeted next-generation sequencing-based preimplantation genetic testing for aneuploidy (PGT-A) platform.

DESIGN

Retrospective observational study performed between 2020 and 2022.

SETTING

Clinic.

PATIENT(S): A total of 12,157 patients who underwent clinical PGT-A performed by targeted next-generation sequencing for whole chromosome and large segmental aneuploidies.

INTERVENTION(S): An incidental finding was reported when a CNV gain/loss of at least 3 consecutive amplicons appeared in at least 2 embryos from the same in vitro fertilization cycle.

MAIN OUTCOME MEASURE(S): The primary outcome measures were the specificity, incidence, inheritance, and pathogenicity of small CNVs detected by the PGT-A platform. Accuracy of the PGT-A platform CNV calls was assessed via concordance with the CNV calls (size and genomic location) on chromosomal microarray of the gamete provider(s). Parental origin of the CNV and pathogenicity classifications were also reported.

RESULT(S): An incidental finding that met reporting criteria was identified in 75 (0.62%; 95% confidence interval, 0.5%-0.8%) of 12,157 unique PGT-A patients. Chromosomal microarray follow-up was requested for all cases, and results were received for 1 or both members of 65 reproductive couples. In all cases, 1 of the gamete providers was confirmed to have the CNV identified in the embryos (100.0%, N = 65/65; 95% confidence interval, 94.5-100). The identified CNV was of maternal origin in 34 cases (52.3%) and of paternal origin in 31 cases (47.7%). A significant correlation was identified between PGT-A-predicted CNV sizes and chromosomal microarray detected sizes (r = 0.81) and genomic coordinates on parental deoxyribonucleic acid. Twenty-six (40%) of the CNVs were classified as benign/likely benign, 30 (46.2%) as a variant of uncertain significance, and 9 (13.8%) as pathogenic/likely pathogenic.

CONCLUSION(S): Certain PGT-A platforms may enable the detection of inherited, small CNVs with extremely high specificity without prior knowledge of parental status. Most CNVs in this data set were confirmed to be benign/likely benign or a variant of uncertain significance. Pathogenic/likely pathogenic CNVs associated with a broad range of phenotypic features may also be detected, although a reliable negative predictive value for small CNVs with current PGT-A technologies is unknown because of the many technical challenges.