An artificial intelligence model (euploid prediction algorithm) can predict embryo ploidy status based on time-lapse data

Background For the association between time-lapse technology (TLT) and embryo ploidy status, there has not yet been fully understood. TLT has the characteristics of large amount of data and non-invasiveness. If we want to accurately predict embryo ploidy status from TLT, artificial intelligence (AI) technology is a good choice. However, the current work of AI in this field needs to be strengthened. Methods A total of 469 preimplantation genetic testing (PGT) cycles and 1803 blastocysts from April 2018 to November 2019 were included in the study. All embryo images are captured during 5 or 6 days after fertilization before biopsy by time-lapse microscope system. All euploid embryos or aneuploid embryos are used as data sets. The data set is divided into training set, validation set and test set. The training set is mainly used for model training, the validation set is mainly used to adjust the hyperparameters of the model and the preliminary evaluation of the model, and the test set is used to evaluate the generalization ability of the model. For better verification, we used data other than the training data for external verification. A total of 155 PGT cycles from December 2019 to December 2020 and 523 blastocysts were included in the verification process. Results The euploid prediction algorithm (EPA) was able to predict euploid on the testing dataset with an area under curve (AUC) of 0.80. Conclusions The TLT incubator has gradually become the choice of reproductive centers. Our AI model named EPA that can predict embryo ploidy well based on TLT data. We hope that this system can serve all in vitro fertilization and embryo transfer (IVF-ET) patients in the future, allowing embryologists to have more non-invasive aids when selecting the best embryo to transfer.

DNA Repair in Haploid Context

DNA repair is a well-covered topic as alteration of genetic integrity underlies many pathological conditions and important transgenerational consequences. Surprisingly, the ploidy status is rarely considered although the presence of homologous chromosomes dramatically impacts the repair capacities of cells. This is especially important for the haploid gametes as they must transfer genetic information to the offspring. An understanding of the different mechanisms monitoring genetic integrity in this context is, therefore, essential as differences in repair pathways exist that differentiate the gamete’s role in transgenerational inheritance. Hence, the oocyte must have the most reliable repair capacity while sperm, produced in large numbers and from many differentiation steps, are expected to carry de novo variations. This review describes the main DNA repair pathways with a special emphasis on ploidy. Differences between Saccharomyces cerevisiae and Schizosaccharomyces pombe are especially useful to this aim as they can maintain a diploid and haploid life cycle respectively.

Is day 5 blastocyst better than day 6 blastocyst? Evidence from NGS-based PGT-A results

Background Aneuploidy is the principal genetic factor leading to the failure of embryo implantation. For most patients who accept the non-preimplantation genetic testing (PGT) cycle, non-invasive methods to select euploid embryos with the best pregnancy potential are desirable Methods This retrospective study recruited women undergoing PGT for aneuploidy (PGT-A) with trophectoderm biopsy from January 2019 to December 2020. The ploidy status of embryos was determined by next generation sequencing (NGS). Results Altogether 2531 blastocysts from 839 PGT-A cycles were evaluated. The euploid rate of day 5 blastocysts seemed to be significantly higher than that of day 6 blastocysts, either from the same ovarian stimulation (OS) cycles (49.9% vs 35.7%, P < 0.001) or from different OS cycles (48.2% vs 27.8%, P < 0.001). Both the younger maternal age (adjusted OR = 0.917, 95% CI: 0.892–0.944, P < 0.001) and day 5 stage (adjusted OR = 1.735, 95% CI: 1.415–2.127, P < 0.001) were independently associated with the greater euploid rate of blastocysts. However, after single euploid embryo transfer, the clinical outcomes of day 5 blastocysts were comparable to those of day 6 blastocysts, no matter whether they were from the same OS cycles or not. Conclusions Our results revealed that day 5 blastocysts possess a higher euploid rate than day 6 blastocysts independent of the OS cycles. Giving priority to a day 5 blastocyst over day 6 blastocyst will increase the likelihood to select single euploid embryo for transfer in non-PGT cycles.

PTEN and DNA Ploidy Status by Machine Learning in Prostate Cancer

Machine learning (ML) is expected to improve biomarker assessment. Using convolution neural networks, we developed a fully-automated method for assessing PTEN protein status in immunohistochemically-stained slides using a radical prostatectomy (RP) cohort (n = 253). It was validated according to a predefined protocol in an independent RP cohort (n = 259), alone and by measuring its prognostic value in combination with DNA ploidy status determined by ML-based image cytometry. In the primary analysis, automatically assessed dichotomized PTEN status was associated with time to biochemical recurrence (TTBCR) (hazard ratio (HR) = 3.32, 95% CI 2.05 to 5.38). Patients with both non-diploid tumors and PTEN-low had an HR of 4.63 (95% CI 2.50 to 8.57), while patients with one of these characteristics had an HR of 1.94 (95% CI 1.15 to 3.30), compared to patients with diploid tumors and PTEN-high, in univariable analysis of TTBCR in the validation cohort. Automatic PTEN scoring was strongly predictive of the PTEN status assessed by human experts (area under the curve 0.987 (95% CI 0.968 to 0.994)). This suggests that PTEN status can be accurately assessed using ML, and that the combined marker of automatically assessed PTEN and DNA ploidy status may provide an objective supplement to the existing risk stratification factors in prostate cancer.

The Association of Kinetic Variables with Blastocyst Development and Ploidy Status

Background: Despite a plethora of studies conducted so far, a debate is still unresolved as to whether TLM can identify predictive kinetic biomarkers or algorithms universally applicable. Therefore, this study aimed to elucidate if there is a relationship between kinetic variables and ploidy status of human embryos or blastocyst developmental potential. Methods: For conducting this retrospective cohort study, the normal distribution of data was verified using Kolmogorov-Smirnov test with the Lilliefors’ amendment and the Shapiro-Wilk test. Kinetic variables were expressed as median and quartiles (Q1, Q2, Q3, Q4). Mann-Whitney U-test was used to compare the median values of parameters. Univariate and multiple logistic regression models were used to assess relationship between blastocyst developmental potential or ploidy status and kinetics. Several confounding factors were also assessed. Results: Blastocyst developmental potential was positively correlated with the t4-t3 interval (s2) (OR=1.417, 95% CI of 1.288-1.560). s2 median value was significantly different between high- and low-quality blastocysts (0.50 and 1.33 hours post-insemination, hpi, respectively; p=0.003). In addition, timing of pronuclear appearance (tPNa) (OR=1.287; 95% CI of 1.131-1.463) had a significant relationship with ploidy changes. The median value of tPNa was statistically different (p=0.03) between euploid and aneuploid blastocysts (Euploid blastocysts=8.9 hpi; aneuploid blastocysts=10.3 hpi). Conclusion: The present findings are in line with the study hypothesis that kinetic analysis may reveal associations between cleavage patterns and embryo development to the blastocyst stage and ploidy status.

P–214 Effect of artificial activation of oocytes (AOA-ICSI) on the ploidy status of the resultant blastocysts. A sibling-oocytes pilot study

Study question Will artificial activation of oocytes alter the ploidy status of the resultant blastocysts? A sibling-oocytes pilot study Summary answer AOA-ICSI does not increase the risk of having aneuploidy blastocysts and can improve the fertilization rate in patients with sperm factor deficiency. What is known already Despite introducing ICSI as an aid to improve chances of fertilization, fertilization failure can still occur in 2–3% of ICSI cycles. Fertilization is a complex process triggered by a cascade of events following calcium (Ca2+) oscillations. Evidence suggests that the deficiency, localization or altered structure of the sperm-derived protein PLCζ in oocyte activation may be a reason for meiotic II arrest in the oocyte. Artificial oocyte activation has been proposed to compensate for the lack of calcium oscillation and resumes meiotic progression. There are however insufficient studies to determine its effect on the chromosomal status of the resultant blastocysts. Study design, size, duration This is a prospective, randomized study conducted at our Center from August-October 2020. A total of 20 couples intended for ICSI + Preimplatation Genetic Testing for Aneuploidy (PGT-A) cycles were recruited based on fulfilling one of the following criteria: 1) previous total fertilization failure (TFF), 2) history of low fertilization rate (&lt;30%), 3) more than 2 cycles of failed IVF cycles (no implantation) 4) poor embryo development (no blastocysts formed) and 5) severe male factor. Participants/materials, setting, methods A total of 231 MII oocytes underwent randomization in a 1:1 ratio between AOA-ICSI and control group. All oocytes are subjected to ICSI treatment. Oocytes in the AOA-ICSI group are treated in 25μl droplets 10μM ready to use bicarbonate buffered calcium ionophore (Kitazato, Japan) for 15 minutes post-ICSI. The blastocysts were biopsied and subjected to PGT-A. Primary outcome was the aneuploidy rate and secondary outcomes were fertilization rate and blastocyst rate. Main results and the role of chance There were 11 out of 40 (27.5%) aneuploid blastocysts in the AOA-ICSI group and 7 out of 23 aneuploid blastocysts (30.4%) in the control group [odds ratio (OR) = 0.87; 95% confidence interval (CI) 0.28–2.68, p = 0.8040). There was no statistically significant difference between both groups. However, fertilization rate of the AOA- ICSI group was significantly higher than the fertilization rate in the control group (68.6% vs 49.6% respectively, OR = 2.22; 95% CI, 1.31–3.81, p = 0.0034). There were 40 blastocysts formed in the AOA-ICSI group and 23 blastocysts formed in the control group. It was found that the AOA-ICSI group yielded a higher blastocyst rate (49.4%) compared to the control group (41.1%) (OR = 1.40; 95% CI, 0.71 to 2.78, p = 0.3379) but the difference was not statistically significant. Limitations, reasons for caution The possibility of TE cells biopsied may not be representative of the whole blastocyst makes it possible to have false clinical data. The dosage and time were also not evaluated in this study as exposure time was found to be a critical factor of fertilization rate in a previous study. Wider implications of the findings: This study showed that AOA-ICSI does not increase the risk of having aneuploidy blastocysts and can improve the fertilization rate in patients with sperm factor deficiency. Additional studies involving a larger number of patients with more specific indication can further justify the benefits of AOA as a therapeutic application. Trial registration number NA

P–663 The ratio of serum progesterone (P4) to the number of follicles (P4/Follicle) is a more objective parameter for euploidy rate as compared to systemic progesterone

Study question Does the ratio of serum progesterone (P4) to the number of follicles (P4/Follicle) on the day of final oocyte maturation affect the ploidy status of the embryos? Summary answer A high P4/Follicle ratio negatively affects the euploid rate of the embryos. What is known already During ovarian stimulation, exogenous gonadotropins are administered to achieve multifollicular growth. Intense gonadotropin stimulation towards the end of the follicular phase seems to cause a premature progesterone rise in stimulated IVF cycles. The impact of serum progesterone elevation during the follicular phase has been studied intensively. Though most studies have focused on the effect of progesterone elevation on the endometrial receptivity, little is known about its possible impact on embryo development and ploidy status. The only study that investigated the effect of progesterone on the embryo ploidy status, was unable to show any significant impact. Study design, size, duration This retrospective study was performed at ART Fertility Clinics Abu Dhabi, UAE and Muscat, Oman. All stimulation cycles (n = 975) were performed between January 2015 to December 2019 with patients aged between 18–45, Body mass index (BMI) of 18–35, stimulated either with rFSH or hMG. All embryos underwent ICSI and Preimplantation Genetic Testing for Aneuploidies (PGT-A),Patients with surgical sperm extraction, warmed oocytes or natural cycle IVF were excluded. Participants/materials, setting, methods Serum P4 was measured on the last ultrasound prior triggering for final oocyte maturation. The P4/Follicle ratio was calculated as the ratio of P4 on trigger day to the number of follicles &gt; 10 mm on the last ultrasound. Serum P4 and P4/Follicle ratio were then analyzed using linear and univariate regression model to find potential correlation with the number of oocytes retrieved, number of mature oocytes, embryo quality (day 3 and 5), and euploid rate. Main results and the role of chance A total of 975 cycles were analyzed, with a mean age of 33.88±0.05 years, a mean BMI of 26.7±0.035 kg/m2. The mean number of oocytes collected was 12.53±0.058. Mean serum P4 on trigger day was 0.83±0.005 ng/ml and higher serum P4 values were observed as the number of oocytes retrieved and the number of mature oocytes increased (β = 0.026, p &lt; 0.0001 and β = 0.028, p &lt; 0.001, respectively). On the other hand, the mean P4/Follicle ratio was 0.056±0.00041 ng/ml and, unlike serum P4, the P4/Follicle ratio showed a negative correlation with the number of oocytes retrieved as well as with the number of mature oocytes (β=–0.001, p &lt; 0.001 and β=–0.001, p &lt; 0.001, respectively). While day 3 embryos were not affected by serum P4 or P4/Follicle ratio, the blastocyst quality was negatively affected by both increasing serum P4 levels and the P4/Follicle ratio (β=–0.012 p &lt; 0.05, β=–0.002, p &lt; 0.001, respectively). Euploid rates were positively correlated in cycles with increased serum P4 β = 0.18, p &lt; 0.001), while negatively correlated in cycles with a high P4/Follicle ratio (β=–0.015, p &lt; 0.001). After adjusting for potential confounders, only P4/Follicle remained as a significant negative factor for euploid rate (β=–0.004, p &lt; 0.001, 95% CI: –0.007- –0.001, p &lt; 0.001), which was not observed for serum P4 (p = 0.46). Limitations, reasons for caution This is an observational study based on retrospective data; an improved extrapolation of the results might be obtained by performing a prospective study. Wider implications of the findings: The findings of this study should encourage clinicians to optimize the ovarian stimulation protocols not only based on serum P4, but also considering the P4/Follicle ratio. Trial registration number Not applicable