Retrospective analysis of outcomes after IVF using an aneuploidy risk model derived from time-lapse imaging without PGS

Time-lapse imaging of human preimplantation IVF embryos has enabled objective algorithms based on novel observations of development (morphokinetics) to be used for clinical selection of embryos. Embryo aneuploidy, a major cause of IVF failure, has been correlated with specific morphokinetic variables used previously to develop an aneuploidy risk classification model. The purpose of this study was to evaluate the effectiveness and potential impact of this model for unselected IVF patients without biopsy and preimplantation genetic screening (PGS). Embryo outcomes – no implantation, fetal heart beat (FHB) and live birth (LB) – of 88 transferred blastocysts were compared according to calculated aneuploidy risk classes (low, medium, high). A significant difference was seen for FHB (P < 0.0001) and LB (P < 0.01) rates between embryos classified as low and medium risk. Within the low-risk class, relative increases of 74% and 56%, compared with rates for all blastocysts, were observed for FHB and LB respectively. The area under the receiver operating characteristic curve was 0.75 for FHB and 0.74 for LB. This study demonstrates the clinical relevance of the aneuploidy risk classification model and introduces a novel, non-invasive method of embryo selection to yield higher implantation and live birth rates without PGS. The largest single cause of failure of the human embryo to implant, and of early miscarriage, is aneuploidy (errors in numbers of chromosomes of the early (preimplantation) human embryo). More than half of human embryos are believed to be affected by aneuploidy and many must be unwittingly transferred following IVF, resulting in failed implantation, miscarriage or the birth of a baby with a related disorder. It is not possible for embryologists in IVF laboratories to identify aneuploid embryos under the microscope; however, our studies with time-lapse incubation in conjunction with preimplantation genetic screening (PGS) of IVF embryos have allowed us to develop and publish a model that rates each embryo based on its developmental patterns (morphokinetics) as being at low, medium or high risk of aneuploidy. PGS, whilst very effective, requires expensive technology and expertise, is not widely available and requires the embryo to undergo biopsy (removal of cell(s)). In this study, we tested the aneuploidy risk classification model on embryos with a known outcome and, importantly, on those embryos resulting in a live birth. This demonstrated how the model has the potential to greatly enhance IVF outcome without biopsy and PGS. By using such unique, non-invasive and specifically designed embryo selection models, we can now make more informed choices in order to select the most viable embryo to transfer, with the lowest risk of aneuploidy. Selection of an embryo classified as low risk has improved the relative chance of a live birth by 56% over conventional embryo selection