P-194 Initial experience of using iDAScore as a tool to predict euploid blastocysts

Abstract

Study question Can the annotationfree embryo scoring system (iDAScore) predict the likelihood of euploidy in embryos? Summary answer Our results have shown that there is a significant positive correlation between iDAScore and blastocyst euploidy. What is known already iDAScore (Vitrolife, Sweden) is an embryo ranking model that was developed by using artificial intelligence (AI) and deep learning. It does not require any user-dependent annotation which largely eliminates the subjectivity of manual annotation by embryologists. It has also been reported to be a robust predictor in likelihood of embryo implantation. However, the correlation between iDAScore and euploidy has not been extensively studied. Therefore, this study is to determine if iDAScore can be utilized as a predictor for embryo euploidy. Study design, size, duration A total of 860 blastocysts (mean age: 30.0; age range: 19.0-35.0) were assessed with iDAScore on day-5 and/or day-6 of embryo development from January 2022 to October 2022 at Alpha IVF & Women’s Specialists. The euploidy rates were analyzed based on the stratified iDAScores (Group A:0.0-4.9; Group B: 5.0-7.4; Group C:7.5-8.9; Group D:9.0-10.0). Participants/materials, setting, methods Scores were computed by the iDAScore software at the Vitrolife Technology Hub (Vitrolife, Gothenburg, Sweden) prior to trophectoderm biopsy for Preimplantation Genetic Testing for Aneuploidies (PGT-A) (Ion Torrent, USA). The ploidy status (euploid or aneuploid) based on PGT-A results were recorded and assessed with the corresponding iDAScore. Main results and the role of chance The euploidy rates of Group A, B, C and D were 56.7% (17/30), 60.0% (105/175), 65.9% (178/270) and 74.8% (288/385) respectively. Group D shows significantly higher euploidy rate compared to Group B and Group C (p = 0.0005, p = 0.0143). A weak significance was observed between Group D and Group A (p = 0.0505). This is probably due to smaller sample size in Group A. No significant differences were displayed among Group A, Group B and Group C (p = > 0.005). Our results demonstrated an increasing trend between iDAScore and the proportion of euploid embryos. Limitations, reasons for caution Few of the blastocysts in Group A could be selected for biopsy so the sample size was only a fraction of the other groups. Further studies and larger sample sizes would allow the findings to be stated more confidently. Wider implications of the findings Although the iDAScore algorithm scores the chance of embryo implantation, we have demonstrated its potential to assist in predicting the embryo ploidy status in a non-invasive manner. This may be used to inform the decision of which and how many blastocysts to biopsy and transfer, thus reducing treatment costs. Trial registration number not applicable