Lisa Boucret

Time-lapse imaging and deep-learning algorithms are promising tools to assess the most viable embryos and improve embryo selection in IVF laboratories. Here, we developed and validated a deep learning model based on self-supervised contrastive learning. The model was developed with a new approach based on matched KID (Known Implantation Data) embryos derived from the same cohort of a stimulation cycle, both judged to be of good quality according to classical morphological criteria and morphokinetics, transferred fresh or frozen, but with a different implantation fate (clinical pregnancy vs. failure of implantation). We used self-supervised contrastive learning to train convolutional neural networks to ensure an unbiased and comprehensive learning of the morphokinetics features of the embryos, followed by a Siamese neural network fine-tuning and an XGBoost final prediction model to prevent overfitting. 1580 embryo videos of 460 patients were included between January 2020 and February 2023. With the knowledge of the implantation outcome of a previous transfer of an embryo derived from the same stimulation cycle, this model could predict the pregnancy outcome of the subsequent transfer with an AUC of 0.57. Without any knowledge of transfer history, the model achieved a satisfactory performance in predicting implantation (AUC = 0.64). This model could be considered as an adjunct tool for biologists to better select embryos and reduce the number of useless transfers per patient, when a cohort with several embryos classified as good quality by classical criteria is obtained.

There is great controversy as to whether women with Diminished Ovarian Reserve (DOR) exhibit only a quantitative decrease in ovarian reserve or also impaired oocyte and embryo quality. In this retrospective study, we aimed to evaluate the impact of DOR on embryo morphokinetic parameters with a time-lapse system. 1314 embryos were obtained from 256 couples undergoing IVF or ICSI cycles, with 242 embryos in the DOR group as classified by the Bologna and POSEIDON criteria and 1072 embryos derived from the Normal Ovarian Reserve (NOR) group. For each morphokinetic parameter (t2, t3, t4, t5, t8, tB, ECC2, cc2a, ECC3, s2, s3), a generalized linear mixed model was created to control for female age, BMI, smoking status, method of insemination and correlation between oocytes from a same cohort. No significant association was found between DOR and any of the morphokinetic parameters studied. In a secondary analysis, we evaluated the influence of maternal aging, comparing morphokinetic characteristics between two age groups (<37 and ≥37 years). In the univariate analysis, we found that embryos from older women displayed a slower embryo development (in particular for t3, t4, t5, tB, and ECC2), although without statistical significance in the multivariate analysis. In conclusion, our study did not reveal any substantial impact of ovarian aging on early morphokinetic parameters and suggested potential biases that may be a source of controversy in the literature.

Time-lapse systems (TLS) and associated algorithms are interesting tools to improve embryo selection. This study aimed to evaluate how TLS and KIDScore™ algorithm changed our practices of embryo selection, as compared to a conventional morphological evaluation, and improved clinical pregnancy rates (CPR). In the study group (year 2020, n = 303 transfers), embryos were cultured in an EmbryoScope+ time-lapse incubator. A first team observed embryos conventionally once a day, while a second team selected the embryos for transfer based on time-lapse recordings. In the control group (year 2019, n = 279 transfers), embryos were selected using the conventional method, and CPR were recorded. In 2020, disagreement between TLS and the conventional method occurred in 32.1% of transfers, more often for early embryos (34.7%) than for blastocysts (20.5%). Irregular morphokinetic events (direct or reverse cleavage, multinucleation, abnormal pronuclei) were detected in 54.9% of the discordant embryos. When it was available, KIDScore™ was decreased for 73.2% of the deselected embryos. Discordant blastocysts mainly corresponded with a decrease in KIDScore™ (90.9%), whereas discordant Day 3 embryos resulted from a decreased KIDScore™ and/or an irregular morphokinetic event. CPR was significantly improved in the TLS group (2020), as compared to the conventional group (2019) (32.3% vs. 21.9%, p = 0.005), even after multivariate analysis. In conclusion, TLS is useful to highlight some embryo development abnormalities and identify embryos with the highest potential for pregnancy.

Endometriosis and infertility are closely linked, but the underlying mechanisms are still poorly understood. This study aimed to evaluate the impact of endometriosis on in vitro fertilization (IVF) parameters, especially on embryo quality and IVF outcomes. A total of 1124 cycles with intracytoplasmic sperm injection were retrospectively evaluated, including 155 cycles with endometriosis and 969 cycles without endometriosis. Women with endometriosis had significantly lower ovarian reserve markers (AMH and AFC), regardless of previous ovarian surgery. Despite receiving significantly higher doses of exogenous gonadotropins, they had significantly fewer oocytes, mature oocytes, embryos, and top-quality embryos than women in the control group. Multivariate analysis did not reveal any association between endometriosis and the proportion of top-quality embryo (OR = 0.87; 95% CI [0.66–1.12]; p = 0.3). The implantation rate and the live birth rate per cycle were comparable between the two groups (p = 0.05), but the cumulative live births rate was significantly lower in in the endometriosis group (32.1% versus 50.7%, p = 0.001), as a consequence of the lower number of frozen embryos. In conclusion, endometriosis lowers the cumulative live birth rates by decreasing the number of embryos available to transfer, but not their quality.

To determine if a modification of the cytokine profile occurs in the follicular fluid (FF) of women with endometriosis undergoing in vitro fertilization (IVF), we performed a prospective observational study from January 2018 to February 2019. In total, 87 women undergoing IVF were included: 43 for severe endometriosis-related infertility and 40 controls with other causes of infertility. The cytokine profile of the FF was determined by multiplex fluorescent-bead-based technology allowing the measurement of 59 cytokines. Monocyte Chemoattractant Protein 1 (MCP-1) was the only variable retained in the multivariate analysis. We identified two subgroups of patients in the endometriosis group: MCP-1-low group (n = 23), which had FF MCP-1 levels comparable to the control group, and MCP-1-high (n = 20), which had significantly higher FF levels. Only patients in the MCP-1-high group had a significantly altered cytokine profile in the FF, and had a significantly higher serum estradiol level (p = 0.002) and a significantly lower number of oocytes recovered (p = 0.01) compared to the MCP-1-low and the control group. Our study has shown an alteration of the oocyte microenvironment in women with endometriosis associated with high follicular fluid levels of MCP-1, allowing the identification of a subgroup of endometriosis patients with a potentially worse prognosis.

To date, there is no consensus on the ideal management strategy of patients with poor ovarian response (POR) to controlled ovarian stimulation (COS) for in vitro fertilization (IVF). Currently, these patients are given the choice of: (1) canceling the cycle; (2) proceeding with COS regardless of the poor response, and performing the oocyte retrieval and transfer of embryos when available; or (3) conversion to an intrauterine insemination (IUI). When the decision to proceed with the COS cycle is taken, it is not clear whether IVF or conversion to IUI is the best choice. If live birth rates were comparable between the two strategies, conversion to IUI would be the better option for poor responders, since it is less invasive and is associated with a lower cost. We designed a non-inferiority, multicentric, randomized controlled trial that will be conducted in 18 French Reproductive Medicine centers. We defined POR as the presence of only two or four mature follicles ≥ 14 mm on ovulation trigger day. Patients with POR will be randomized into two parallel arms: “IVF” and “conversion to IUI.” Our main objective is to compare the efficiency of IVF and conversion to IUI in patients with POR to COS. The primary outcome is the live birth rate, defined as the birth of a living infant after 22 weeks’ gestational age, or weighing ≥ 500 g. One of the secondary objectives is to compare the cost-efficiency of both strategies at 12 months. We will need to include 940 patients (470 in each arm), and the duration of the inclusion period is estimated to be 36 months. This is the first randomized controlled trial to compare the outcomes of IVF and embryo transfer to conversion to IUI in patients with POR to COS. If our study shows that conversion to IUI is non-inferior to IVF in terms of clinical efficiency and live birth rate, it would confirm IUI as a better alternative for patients, both individually (less invasive and more patient-friendly) and collectively (lower cost). ClinicalTrials.gov, ID: NCT03362489 . Registered on January 10th, 2018.