Blastulation and ploidy prediction using morphology assessment in 33,999 day-3 embryos

Area of Science:

• Reproductive Medicine
• Embryology
• Genetics

Background:

• Contemporary in vitro fertilization (IVF) often favors blastocyst-stage embryo transfer.
• However, some centers continue Day-3 embryo transfers, missing advanced morphological markers and blastulation potential.
• Predicting blastocyst development and euploidy from Day-3 morphology is crucial for these centers.

Purpose of the Study:

• To model the association between Day-3 embryo morphology and subsequent blastulation and euploidy.
• To develop and validate a predictive model for blastocyst development and euploidy using Day-3 characteristics.
• To aid fertility centers performing Day-3 embryo transfers in selecting viable embryos.

Main Methods:

• Retrospective cohort study of 33,999 Day-3 embryos from 5,702 IVF cycles.
• Day-3 embryos assessed for cell number and fragmentation.
• Trophectoderm (TE) biopsy for preimplantation genetic testing for aneuploidy (PGT-A) using next-generation sequencing (NGS) on expanded blastocysts.

Main Results:

• Developed prediction models for blastulation and euploidy incorporating Day-3 cell count, fragmentation, and female age.
• Models demonstrated good calibration in validation samples.
• Machine learning models achieved higher accuracy in predicting blastulation and euploidy compared to conventional selection criteria.

Conclusions:

• The developed prediction models can effectively prioritize Day-3 embryos with higher potential for blastocyst development and euploidy.
• These models offer a valuable tool for centers performing Day-3 transfers, requiring no specialized equipment.
• The models are accessible for further validation and clinical application in assisted reproductive technology.