Timing matters: Morphokinetic trajectories of 11,915 embryos provide new insights for blastocyst ploidy prediction using time-lapse monitoring

Abstract

Study question How do different chromosomal constitutions affect embryo developmental trajectories and the timing of morphokinetic events up to the blastocyst stage?

Summary answer Embryos with chromosomal gains follow similar developmental timelines to euploid embryos, while chromosomal losses and multiple abnormalities cause significant delays, especially during the cleavage-to-blastocyst transition.

What is known already Numerous studies have explored the association between morphokinetic events and embryo ploidy. However, the predictive value of these time-lapse parameters remains unclear. Variations in scoring algorithms, embryo assessment criteria, and the inherent complexities of human embryo development all contribute to inconsistent findings. Moreover, current studies have exclusively focused on comparing euploid (normal) and aneuploid (abnormal) embryos, without considering the impact of specific chromosomal aberrations on development. Aneuploidies are known to affect preimplantation development, potentially leading to embryo arrest and altered cell division patterns. However, specific effects of varying chromosomal constitutions on developmental trajectories up to the blastocyst stage remain unexplored.

Study design, size, duration Retrospective, international, multicenter study of 3,044 preimplantation genetic testing for aneuploidies (PGT-A) cycles, performed between January 2019 and July 2024, across four clinics and two continents. The study included 11,915 blastocysts from 2,917 patients, monitored using time-lapse imaging to track key morphokinetic timepoints, including pronuclei fading (tPNf), cleavage stages (t2 to t8), blastulation (tSB), and time to full blastocyst (tB). All parameters were annotated relative to time (t)0, defined as the time of ICSI completion.

Participants/materials, setting, methods The association between PGT-A results and individual morphokinetic timepoints were tested with mixed-effect linear regression models using interaction terms between PGT-A results and morphokinetic stages. Developmental trajectories were identified with an unsupervised learning algorithm, k-means clustering. The association between trajectory groups and ploidy status was tested with multivariate mixed-effects logistic regression analysis. P-values <0.05 were considered significant.

Main results and the role of chance Mean maternal age (±SD) was 38.9 (±3.9) years, with an overall euploidy rate of 40.1%. Compared to euploid embryos, aneuploid embryos were delayed throughout preimplantation development, from t2 to tB. These delays became increasingly pronounced as embryos approached blastulation, with a 1.3 hour delay observed at t8, 2.4 hour delay at tSB, and a 3.7 hour delay at tB (p < 0.001 for all). However, different aneuploidies exerted varying effects on developmental progression. Embryos with chromosomal gains, including clinically viable trisomies, were comparable to euploid embryos. Conversely, single and double chromosomal losses caused significant delays proportionate to their severity, starting from t4. On average, single losses reached tB 3.5 hours later than euploid embryos (p < 0.001), while double losses formed blastocysts 5.5 hours later (p < 0.001). Embryos with multiple aneuploidies were delayed across all morphokinetic timepoints (p < 0.05). We identified four distinct embryo developmental trajectories: consistently rapid, steady, inconsistent, and consistently delayed, each significantly and independently associated with ploidy status after adjusting for maternal age and embryo quality (p < 0.05). Euploidy rates for these trajectories were 50.4%, 41.6%, 38.7%, and 31.8%, while chromosomal loss rates were 14.9%, 17.7%, 21.5% and 21.2%, respectively (p < 0.001). Notably, chromosomal gain rates were comparable across all developmental trajectories (p > 0.05).

Limitations, reasons for caution The retrospective design limits generalizability. Only embryos that successfully developed to the blastocyst stage and underwent PGT-A could be analyzed, potentially introducing selection bias by excluding embryos that arrested earlier. Certain variables were not available in the dataset, therefore the full elucidation of all potential confounders may not be possible.

Wider implications of the findings This comprehensive study offers novel insights into the morphokinetics of embryos based on their chromosomal status, highlighting the importance of considering ploidy in human embryo research. These findings also have significant clinical implications, offering the potential to refine embryo selection algorithms and support more precise, biologically informed clinical decision-making.