P-132 Does the automated blastocyst assessment system, iDAScore®, provide valuable data for predicting the results of preimplantation genetic testing for aneuploidy (PGT-A)?

Abstract

Abstract Study question What iDAScore® could indicate the presence of euploidy in PGT-A? Is it a valid predictor of the result of PGT-A? Summary answer PGT-A results with an iDAScore® of 8.3 or higher can indicate euploidy. However, patient age and trophectoderm (TE) grade can be effective predictors as well. What is known already PGT-A was approved as a general test in Japan in 2022. Currently, only patients who meet the criteria of Japan Society of Obstetrics and Gynecology can undergo the test. It is expected that the test will improve the live birth rate and reduce the miscarriage rate. The iDAScore® uses artificial intelligence and deep learning to score blastocysts. It is anticipated to be more objective than is morphological evaluation, which varies among individuals. Unlike PGT-A, the iDAScore® can assess embryos without cellular invasion. However, only a few reports have investigated the association between PGT-A and iDAScore® results. Study design, size, duration The study was performed retrospectively on 3,781 cycles of in vitro fertilization (IVF) conducted at our clinic between August 2020 and October 2022. Of the total of 5,684 embryos derived from conventional IVF or intracytoplasmic sperm injection that were cultured in a time lapse incubator, 293 blastocysts were used in the study. Participants/materials, setting, methods On the basis of their PGT-A results, the blastocysts were divided into euploidy and aneuploidy groups. The mean age of patients in each group, the methods of insemination, and the median iDAScore® were evaluated. The area under the curve (AUC) and the cut-off value were calculated from the PGT-A and iDAScore® results. A logistic regression analysis of the predictive parameters of PGT-A results, including the iDAScore®, was also performed. Main results and the role of chance Patients’ mean ages in the euploidy and aneuploidy groups were 38.5 and 41.5 years, respectively. This was significantly higher in the aneuploidy group (P<0.001). There was no significant difference in the stage of maturity at oocyte retrieval (P = 0.302). Similarly, no significant differences were associated with method of insemination (P = 0.961). The median iDAScore® for the euploidy group was 8.7; for the aneuploidy group, it was 7.8. This result was significantly higher in the euploidy group (P<0.001). The AUC, calculated from PGT-A results and the iDAScore® with receiver operating characteristic curves, was 0.685 (95% CI [0.617−0.754]). The cut-off value was 8.3. The logistic regression analysis of age, stage of maturity, insemination method, timing of blastocyst freezing, diameter before blastocyst freezing, inner cell mass, TE grade in the Gardner classification, and iDAScore® as parameters for the PGT-A results indicated significant differences for age and TE grade (age: P<0.001, odds ratio [95%CI]: 1.320 [1.19-1.45]; TE grade 2: P = 0.03, odds ratio [95%CI]: 2.480 [1.11-5.57]; TE grade 3: P = 0.02, odds ratio [95%CI]: 3.740 [1.27-11.00]), while no significant differences were associated with the iDAScore®. Limitations, reasons for caution The iDAScore® is likely to depend on the timing of blastocyst freezing. Furthermore, the iDAScore® utilised in this study was the score of blastocysts that fulfilled our freezing criteria; the blastocyst that reaches a size of at least 150-160 µm in diameter on Day 5,6, or 7 is frozen. Wider implications of the findings An iDAScore® of 8.3 or above could indicate a euploidy PGT-A result. However, the logistic regression analysis indicated that age and TE grade were more significant predictors than the iDAScore® was. Therefore, it is difficult to predict PGT-A results from the iDAScore® itself. Trial registration number Not applicable