The accuracy of chromosomal microarray testing for identification of embryonic mosaicism in human blastocysts

BACKGROUND: Most previous studies of chromosomal mosaicism in IVF embryos were performed by fluorescence in situ hybridization (FISH) methods. While there are reports implicating chromosome aneuploidy in implantation failure following transfer and pregnancy loss by spontaneous miscarriage, the significance of mosaicism for the developmental potential of growing embryos is unknown. However, the low prevalence of chromosomal mosaicism in chorionic villus sampling and amniotic fluid specimens suggests the presence of selection against mosaic embryos for implantation and early pregnancy. The absence of evidence for selective allocation of abnormal cells to the trophectoderm (TE) of mosaic blastocysts permits these cells to be a good proxy for embryonic mosaicism detection by chromosomal microarrays (CMA). The purpose of this study was to establish the limits of detection and the prevalence of chromosome mosaicism in day 5/6 human embryos using CMA with TE biopsies. RESULTS: From reconstitution experiments we established log(2) ratio thresholds for mosaicism detection. These studies indicated that chromosomal mosaicism at levels as low as between 25-37% can be consistently identified. Follow-up studies by FISH on non-transferred abnormal embryos confirmed the diagnostic accuracy of CMA testing. The number of cells in a TE biopsy can influence mosaicism detection. CONCLUSIONS: Chromosomal microarrays can detect mosaicism in TE biopsies when present at levels as low as between 25-37% and the prevalence of day 5/6 blastocysts which were mosaic and had no other abnormalities reached 15% among a cohort of 551 embryos examined. Validated protocols for establishing detection thresholds for mosaicism are important to reduce the likelihood of transferring abnormal embryos.