O-100 Incidence, morphology, ploidy and developmental arrest in direct cleavage embryos: what are they teaching us? Morphokinetic analysis of 15.081 embryos

Study question

What is the impact of direct cleavage (DC) occurrence during embryo/blastocyst development regarding morphology grade, embryo arrest and ploidy status?

Summary answer

Blastocyst formation and morphology grade are significantly reduced in DC when compared to non-DC embryos; however, there is no difference in blastocyst ploidy status.

What is known already

Chromosome instability leading to aneuploidy during early cleavage is well known in humans. DC is defined as the abrupt cleavage of one into three or more daughter blastomeres as the result from an irregular mitosis. Three or more mitosis poles can impair early embryo development; however, how some embryos with these irregular divisions might be able to develop into expanded blastocysts is still unknown. Time-lapse system provides an uninterrupted evaluation of embryo morphological and dynamic parameters, allowing our understanding in such events and if those embryos may be associated with a self-correction mechanism to reach blastocyst stage.

Study design, size, duration

Retrospective cohort study from 15.081 embryos, 3.962 cycles and 3.106 patients that were undergoing in vitro fertilization (IVF) treatment according to medical referral in a single private ART center from December 2018 to November 2021. Only two-pronuclear (2PN) embryos cultured in time-lapse incubators (EmbryoScope, Vitrolife, Sweden) were included in the study. Direct cleavage occurance was annotated when a single blastomere directly cleaved into three or more daughter blastomeres.

Participants/materials, setting, methods

DC embryos were also classified in DC 1-3: abnormal cleavage occurred in zygote (1-cell) resulting in 3–4 blastomeres and in DC 2+: abnormal cleavage occurred at the 2-cell stage resulting in 5 or 6 blastomeres. Embryo arrest, blastocyst formation rate and morphology grade (Gardner and Schoolcraft, 1999) were annotated. Blastocyst biopsy reports for ploidy status were also considered for analysis. Mann-Whitney, chi-squared and Fisher tests were applied for statistical analysis. p < 0,05 was considered significant.

Main results and the role of chance

There were 1168 DC embryos (7,7%) from 896 cycles/841 patients; 502 DC1-3 (3,3%) from 391 cycles/366 patients and 666 DC2+ embryos (4,4%) from 475 patients/505 cycles. In DC group, around 21% of patients have the incidence of more than 1 DC embryo per cycle. Maternal age was lower in DC groups: 38,0±3,61 (DC1-3) and 38,60±3,21 (DC2+) years old versus not-DC (39,02±3,01, p < 0,0001). Embryo arrest was higher (DC1-3:85,5% and DC2+:56,8% versus not-DC:35,4%, p < 0,0001) and blastocyst formation rate was lower (14,5% and 43,2% versus 64,6%, p < 0,0001) in both DC groups and also between DC groups (p < 0,0001). There were 4124 biopsied not-DC blastocysts, 24 in DC1-3 and 154 in DC2+. There was no difference in ploidy between not-DC (57,9%, 40,2%, 1,9%, aneuploid, euploid and mosaic respectively) and DC groups (54,2%, 41,7%, 4,2% and 61%, 37% and 1,9% respectively, p = 0.86) and neither between DC-groups (p = 0.69). Inner cell mass morphology grades were distinct in all groups (not-DC A:45,8%, B:45,2% C:9,0%, DC1-3 A:35,6%, B:33,9%, C:30,5% and DC2+ A:23,1%, B:57,0% C:19,9%, p < 0,0001) and between DC-groups (p < 0,0001). Trophoectoderm morphology grades were also higher for not-DC group (not-DC A:30,5%, B:48,9% C:20,6%, DC1-3 A:15,3%, B:47,5%, C:37,3% and DC2+ A:13,6%, B:44,4% C:42,0%, p < 0,0001), however there was no difference between DC-groups (p = 0.80).

Limitations, reasons for caution

Limited data are available on the implantation potential of embryos with DC, as they are not prioritize for transfer, although our study did not demonstrated a higher aneyploidy rate. Patient parameters, with the exception of maternal age, were not considered in this study.

Wider implications of the findings

Although the morphology grades are lower and embryo arrest is higher in DC embryos, probably due to unequal cleavage, if they reach blastocyst stage, the ploidy status of these embryos are similar to not-DC embryos, speculative for a self-correction mechanism. Future steps should focus on their performance in pregnancy/live-birth.

Trial registration number

Not Applicable

O-173 Artificial intelligence blastocyst ploidy distinction through morphokinetics data

Study question

Is it possible to estimate blastocyst ploidy with an artificial intelligence (AI) algorithm built using embryo morphokinetics data?

Summary answer

AI was able to estimate blastocysts status in euploid or aneuploid with an accuracy of 0.99 (training) and 0.71 and 0.70 AUC (blind-test), respectively.

What is known already

Morphokinetic parameters are well associated with implantation rates, thus ensuring time-lapse as a useful tool to enhance embryo ranking and selection. However, regarding embryo ploidy status, clinics still rely exclusively on molecular/genetic testing. Our previous studies has indicated that euploid blastocysts are faster at time of pronucleous fading (tPNf) and time to blastulation (tB). Considering the amount of data derived from the morphokinetic annotations and the promising results that AI approach is bringing to the field, this study aims to verify AI accuracy to distinguish euploid and aneuploidy blastocysts using 17 morphokinetics parameters and prospectively compared to embryo biopsy results. 

Study design, size, duration

This is a prospective cohort study including 402 embryos (cultured in a time-lapse incubator, EmbryoscopePlus, Vitrolife) from 140 patients undergoing IVF treatment with preimplantation genetic testing for aneuploidy (NGS platform) after inform consent form signature between July 2019 and September 2021. Morphokinetics annotations were analyzed by AI in an association of the technique of artificial neural networks (ANN) and genetic algorithms (GA) for ploidy assessment.

Participants/materials, setting, methods

Morphokinetic parameters were manually annotated up to full-expanded blastocyst time. Time intervals and time ratios were calculated, resulting in 17 variables for the IA analysis. Of the 402 embryos data, 252 were randomly divided into training, validating and testing (70%, 15% and 15%, respectively) of the ANN. The remaining 150 data were used for the blind test. The area under the curve (AUC) of the receiver operating characteristic curve was measured to obtain predictive power. 

Main results and the role of chance

From 402 blastocysts biopsied, 185 were euploid and 217 aneuploid. The AI algorithm was trained with 176 embryos (AUC for both euploid and aneuploid= 0.99), tested with 38 embryos (AUC for euploid=0.62 and aneuploid=0.61), validated with 38 embryos (AUC for euploid=0.82 and aneuploid=0.83). For the blind-test 150 embryos were used (AUC for euploid=0.70 and aneuploid=0.71). Blind-test database was checked only after AI algorithm was tested.

Limitations, reasons for caution

The development of this AI algorithm was built from a single IVF center database. The absence of an extent dataset for the blind-test does not allow us to transpose this algorithm for clinical use at this moment.

Wider implications of the findings

The use of artificial intelligence for embryo assessment is a promising tool in IVF laboratories. In our model, using only morphokinetics, a considerable predictive power to evaluate euploid (0.71 AUC) and aneuploid (0.70 AUC) embryos was achieved, indicating a potential use as a non-invasive approach for embryo ranking and selection.

Trial registration number

Not applicable