Reproducibility of different aneuploid findings at second trophectoderm biopsy evaluated by independent NGS platforms

Abstract

Introduction Preimplantation genetic testing of aneuploidies (PGT-A) enables the selection of euploid embryos in order to increase implantation rates and reduce miscarriage rates after embryo transfer. Next Generation Sequencing (NGS) platforms provide higher dynamic range than other methods and are validated for the detection of copy number variations starting at 20% mosaicism. Sub-chromosomal segmental aneuploidies above 15 MBp are reported in routine PGT-A, and the resolution can be increased to 1-2 MBp when PGT targeted to a previously described structural rearrangement is carried out (PGT-SR). While calling whole chromosomal aneuploidy leads to the elimination of clearly non- viable embryos from transfer to the uterus, numerous healthy live-births have been reported after transferring mosaic embryos. Clinical impact of segmental aneuploidies detected in trophectoderm biopsies remains unclear. Material and methods All embryos were biopsied at day 5/6 and trophectoderm samples were analyzed by VeriSeq™ PGS Kit (Illumina). Discarded embryos dedicated to research with the consent of patients were de-vitrified and after blastocyst re-expansion a second trophectoderm sample of the same size was biopsied for analysis using the PG-Seq™ platform (RHS). In total, 95 embryos were re- analyzed and the results of both replicates were compared. After the first PGT-A, 73 embryos were classified aneuploid, 12 mosaic, and 10 euploid. Assessing the same set of samples by aberration type, 63 whole chromosomal aneuploidies were originally called using VeriSeq™, 28 whole chromosomal mosaics, 36 segmental aneuploidies and 9 segmental aneuploidy mosaics. Reproducibility of various aneuploidy types was evaluated after second biopsy analysis using PG-Seq™. Results In the group of 10 euploid embryos, the original status was confirmed in 9 embryos with one rebiopsied sample reporting a mosaic loss of chromosome 17. Aneuploidy was concordant in 63 out of 73 samples with 9/10 of the non-concordant results changing from segmental aneuploidy to euploid. Only 3 of 12 mosaic embryos remained mosaic with 6 rebiopsied mosaic embryos turning to euploid and 3 aneuploid. When evaluating each chromosomal change separately, reproducibility of 96.8% was observed for whole chromosomal aneuploidies (61/63), 50% for segmentals (18/36) and only 10.8% for mosaics (4/37). Particularly, none of the 9 segmental mosaics were detected in the second biopsy. In contrast to this, second biopsy testing revealed 2 whole chromosomal aneuploidies, 8 whole chromosomal mosaics and 5 segmental mosaics which were not detected in the previously analysed sample. Performance of both NGS platforms appeared to be comparable in terms of resolution and dynamic range for mosaic detection. Conclusions Our comparative study showed that concordance between two trophectoderm samples is about 97% when whole chromosomal aneuploidies are reported. This reflects the numbers described previously in studies comparing trophectoderm and inner cell mass of the embryo. In striking contrast, reproducibility of segmental aneuploidies and mosaic findings remains very poor even in the blastocyst stage and disables us to predict real developmental capacity of the embryo. Non-concordance is to be expected in mosaic embryos. Further clinical studies are needed to unravel whether segmental aneuploidies have similar potential to result in healthy live-births as already reported for mosaic embryos.