P–210 Abnormal cleavage patterns during embryo preimplantation development and their effect on blastulation: an overview from IVF patients with multiple IVF cycles in a time-lapse incubator

Study question How common abnormal cleavage patterns (ACP) are in IVF and what are their consequences on embryo developmental competence? Summary answer ACP might affect up to 25% of the 2PN-zygotes, independently from patients’/cycles’ characteristics, and mostly cause embryo developmental arrest around the 4-to–8-cell transition. What is known already Since its implementation in IVF, time-lapse-microscopy (TLM) allowed the standardization of embryo culture within undisturbed incubators, but it has not improved embryo selection especially if blastocyst transfer is performed. Nevertheless, TLM holds the potential for boosting our knowledge of embryo preimplantation development. In particular, a continuous observation of embryo morpho-dynamics unveiled peculiar blastomere cleavage patterns previously unidentifiable with a static morphological assessment. These events are possibly associated with massive mitotic errors, affecting both chromosomes and cytoskeletal components, as well as downstream metabolic imbalances. Still, the causes of ACP and their consequences on embryo developmental/reproductive competence require further investigation. Study design, size, duration Observational study including 75 patients (age:38.6±3.7yr, FSH:8.8±3.6IU/l, AMH:1.7±1.3ng/ml; BMI:21.4±2.4) who conducted multiple IVF cycles (N = 160; 8.7±5.0 cumulus-oocyte-complexes and 6.3±3.6 metaphase-II collected; 201±245 days between first and second cycles) in a time-lapse incubator between 2014–2020. All annotations were performed blindly by two operators and confirmed by a third in case of discordance. The outcomes were the blastulation rate after any ACP, their association between each other and with patients’/cycles’ characteristics. Participants/materials, setting, methods We included only ICSI-cycles after ovarian-stimulation with blastocyst culture conducted in the Embryoscope. Overall, 981 metaphase-II were inseminated and 677 2PN-zygotes annotated. The ACP investigated were: (i)cytokinesis-failure, formation of cytoplasmic septa without cell division; (ii)Chaotic-cleavage, disordered and uneven cleavages; (iii)Direct-unequal-cleavage (DUC), cleavage of zygotes or single blastomeres directly into 3; (iv)Rapid-cleavage, t3-t2<5hr; (v)Reverse-cleavage, fusion of 2 blastomeres into 1; (vi)Fragmentation, presence of numerous non-nucleated fragments; (vii)Blastomeres’ exclusion/extrusion, nucleated cells excluded/extruded from the morula. Main results and the role of chance Among the 2PN-zygotes, the prevalence of cytokinesis-failure was 5.9% (N = 40/677), 15.7% for chaotic-cleavage (N = 106/677), 18.6% for DUC (N = 126/677), 4.1% for rapid-cleavage (N = 28/677), 3.5% for reverse-cleavage (N = 24/677) and 24.1% for fragmentation (N = 163/677). Among the morulae, the prevalence of blastomere exclusion/extrusion was 27% (N = 109/410;1.5±1.2 excluded/extruded cells,range:1–7). The risk for reverse-cleavage was higher among 2PN-zygotes facing failed-cytokinesis (N = 8/40,20% versus N = 16/637,2.5%, OR:9.7,95%CI:3.9–24.3,p<0.01). Fragmentation was instead higher among 2PN-zygotes undergoing chaotic cleavage (N = 47/106,44.3% versus N = 116/571,20.3%, OR:3.1,95%CI:2–4.8,p<0.01) or DUC (N = 46/126,36.5% versus N = 117/551,21.2%, OR:2.1,95%CI:1.4–3.2,p<0.01). Lastly, higher prevalence of blastomeres’ exclusion/extrusion were reported among morulae obtained after chaotic-cleavage (N = 17/29,58.6% versus N = 92/381,24.1%, OR:4.4,95%CI:2–9.7,p<0.01), DUC (N = 26/37,70.3% versus N = 83/373,22.3%, OR:8.3,95%CI:3.9–17.4,p<0.01) and in presence of fragmentation (N = 79/195,75.2% versus N = 30/305,9.8%, OR:27.8,95%CI:15.6–49.8,p<0.01); only a higher trend after rapid-/reverse-cleavage. No predictive factor of ACP was identified among patients’ and cycles’ characteristics, except for higher risks of fragmentation (OR:2.6,95%CI:1.1–6.3,p= 0.04) and blastomeres’ exclusion/extrusion (OR:2.7,95%CI:1.1–7.2,p=0.04) among patients with previous experience with these events. The viable-blastocyst rate per 2PN-zygote was 45.1% (N = 305/677). It was lower in case of failed-cytokinesis (N = 12/40,30% versus N = 293/637,46%, OR:0.5,95%CI:0.25–0.99,p=0.05), chaotic cleavage (N = 20/106,18.9% versus N = 285/571,49.9%, OR:0.23,95%CI:0.14–0.39,p<0.01), DUC (N = 27/126,21.4% versus N = 278/551,50.5%, OR:0.27,95%CI:0.17–0.42,p<0.01), rapid-cleavage (N = 6/22,21.4% versus N = 299/649,46.1%, OR:0.32,95%CI:0.13–0.8,p=0.02), and reverse-cleavage (N = 5/19,20.8% versus N = 300/653,45.9%, OR:0.31, 95%CI:0.11–0.84,p=0.02). No difference was instead shown in case of fragmentation and/or blastomeres’ exclusion/extrusion. Limitations, reasons for caution The patients included were poor-prognosis women undergoing ≥2 cycles. We are expanding the sample size to account for all cycles conducted in time-lapse incubators. Larger sample size will provide also statistical-power to investigate the effect of ACP on blastocysts’ chromosomal and implantation competence, and more visualizations of rapid-/reverse-cleavage events. Wider implications of the findings: After ACP,developmental-arrest mostly occurs around the 4-to–8-cell transition (50–70% versus ∼30%), when embryonic-genome-activation takes place. Surviving embryos often fragment and/or exclude/extrude blastomeres at morulation, without further impact on blastulation-rates. Moreover, ACP seem independent from patients’/cycles’ characteristics. These evidence incite future Research on the biological/genetic mechanisms triggering ACP and their consequences. Trial registration number None