Validation of array comparative genome hybridization for diagnosis of translocations in preimplantation human embryos

Fluorescent in-situ hybridization (FISH) for preimplantation genetic diagnosis (PGD) of structural chromosome abnormalities has limitations, including carrier testing, inconclusive results and limited aneuploidy screening. Array comparative genome hybridization (CGH) was used in PGD cases for translocations. Unbalances could be identified if three fragments were detectable. Smallest detectable fragments were ∼6 Mbp and ∼5 Mbp for blastomeres and trophectoderm, respectively. Cases in which three or more fragments were detectable by array CGH underwent PGD by FISH and concordance was obtained in 53/54 (98.1%). The error rate for array CGH was 1.9% (1/54). Of 402 embryos analysed, 81 were normal or balanced, 92 unbalanced but euploid, 123 unbalanced and aneuploid and 106 balanced but aneuploid. FISH with additional probes to detect other aneuploidies would have missed 28 abnormal embryos in the reciprocal group and 10 in the Robertsonian group. PGD cases (926) were retrospectively reviewed for reciprocal translocations performed by FISH to identify which could have been analysed by array CGH. This study validates array CGH in PGD for translocations and shows that it can identify all embryos with unbalanced reciprocal and Robertsonian translocations. Array CGH is a better approach than FISH since it allows simultaneous screening of all chromosomes for aneuploidy.Fluorescent in-situ hybridization (FISH) has been used in preimplantation genetic diagnosis (PGD) for structural chromosome abnormalities. This approach has limitations such as the need for testing of carriers, poor fixation and no possibility for aneuploidy screening of all chromosomes. The use of array comparative genome hybridization (CGH) can solve these limitations. Array CGH was used in 47 PGD cases for Robertsonian or reciprocal translocations. The smallest detectable fragment size was ∼6 Mbp for single blastomeres (day-3) and ∼5 Mbp for trophectoderm (day-5) biopsies. Translocation cases in which three or more translocated fragments were detectable by array CGH underwent PGD. Of those, a total of 54 non-transferred embryos analysed by array CGH were reanalysed by FISH. The error rate for array CGH was 1.9% (1/54). Of the 402 embryos analysed by array CGH, 81 embryos were normal or balanced, 92 were unbalanced but euploid, 123 unbalanced and aneuploid and 106 balanced but aneuploid. We retrospectively reviewed 926 PGD cases for reciprocal translocations performed by FISH in order to identify those in which fewer than two fragments were less than ∼6 Mbp. In conclusion, this study validates the use of array CGH in PGD for translocations showing a low error rate and shows that array CGH can identify all embryos with unbalanced reciprocal and Robertsonian translocations. Array CGH is a better approach than FISH or single-nucleotide polymorphism arrays because it allows simultaneous screening of all chromosomes for aneuploidy and does not require previous genetic testing of the parents.