Abstract Study question Can morphokinetic analysis be used as a tool to predict the ploidy status of blastocysts? Summary answer A blastocyst is more likely to be aneuploid when it divides slowly to two cells and eight cells but quickly to three cells. What is known already Existing data suggests morphokinetic analysis and PGT-A both have advantages and limitations when used independently. However, the hope is that if strong associations can be found linking morphokinetic qualities of embryos and their ploidy status, then this can be used as a predictive tool. Currently, morphokinetic analysis cannot be used as a predictive tool for euploidy. However, there is some existing data that suggests that clinics can correlate morphokinetics and ploidy status using their in-house data. Study design, size, duration The study was retrospective and involved the assessment of blastocysts that had undergone PGT-A with next generation sequencing (NGS) and had a confirmed outcome of aneuploid or euploid between January 2021 and June 2022. Only patients who obtained genetic results showing they had at least one aneuploid embryo and at least one euploid embryo were included to allow for a matched case-control study – this resulted in 89 aneuploid and 89 euploid embryos being analysed. Participants/materials, setting, methods All embryos were cultured in the same conditions, including the same incubators, media (both from Vitrolife, SWE), and standard operating procedures. Morphokinetic data was exported and analysed from annotations made on the EmbryoScope annotation software. Embryos that could not be fully annotated were excluded from analysis. A conditional logistic regression was performed in order to ascertain the effects of each morphokinetic parameter on the likelihood that the embryo will be aneuploid or euploid. Main results and the role of chance From 89 patients, 178 embryos were analysed, of which half were euploid and half were aneuploid. The morphokinetic time points assessed were time to pronuclei fading (tPNF), time to 2-cell (t2), time to 3-cell (t3), time to 4-cell (t4), time to 5-cell (t5), time to 8-cell (t8), time to start of blastulation (tSB) and time to blastulation (tB). The duration of the second cell cycle (cc2) and its synchrony (s2) were also analysed. When euploid and aneuploid case-control matched embryos were compared, t2, t3 and t8 were the best predictors of aneuploidy when analysed together (P = 0.06, P = 0.074, P = 0.071 respectively). There are higher odds of a blastocyst being aneuploid if a later t2 (OR = 1.361), an earlier t3 (OR = 0.784) and a later t8 (OR = 1.047) have occurred. A later t2 (P = 0.044, OR = 1.361) is statistically significant when alongside an earlier t3 (P = 0.122, OR = 0.828). A later t8 is statistically significant, irrespective of t2 and t3 (P = 0.05, OR = 1.043). In addition to these results, there is a general trend for aneuploid embryos to develop to each time point slower than euploid embryos. Furthermore, there is more variation in the range of values for each time point in aneuploid embryos compared to less variation in euploid embryos. Limitations, reasons for caution There was a small number of embryos included in the study, so a larger cohort would increase the statistical power. A multi-centre study would also be beneficial to incorporate factors such as larger patient demographics, multiple embryo culture environments and variable clinical protocols. Wider implications of the findings Utilising morphokinetic data to predict aneuploidy can help us to counsel the patient on selecting embryos for biopsy, and to make a predictive tool that would be applicable within our laboratory and patient settings. A secondary impact could be supporting the use of non-invasive methods to predict embryo ploidy. Trial registration number Not applicable
Read full abstract