P-284 Correlation between a new day 4 embryo scoring system and blastocyst development and pregnancy rate

Study question

Does a new day 4 embryo scoring system predict blastocyst development and aneuploidies?

Summary answer

The new day 4 embryo scoring system correlates with blastocyst development, blastocyst morphology, day of blastulation, and pregnancy rate.

What is known already

A small number of studies propose different day 4 embryo grading systems. The majority of them report no differences in implantation and clinical pregnancy outcome between day 4 and blastocyst transfer. Moreover, only limited data exist on the association between day 4 grading systems, embryo development, and PGT-A results. In addition, none of the previous studies analyzed the totality of the different day 4 embryo morphology and their subsequent development.

Study design, size, duration

This is a single-center retrospective analysis from June 2019 to August 2021 including 762 embryos from 129 cycles. Day 4 embryos were graded A to J considering A the best morphology and were classified into 1, 2, 3, or 4 categories (1 being the more advanced development). Categories were compared among them to assess blastocyst development and aneuploidy rate.

Participants/materials, setting, methods

Data were obtained from medical records with a previously given consent. Patient characteristics such as age, body mass index (BMI), and anti-Mullerian hormone (AMH) were compared among the four categories using ANOVA with Multiple Comparisons. Blastocyst development, blastocyst morphology, day of blastulation, and pregnancy rate were compared among the categories using Chi-square contingency. All analyses considered a significant p-value of 0.05.

Main results and the role of chance

Younger women presented significantly more category 1 of day 4 embryos when compared with the other categories (p < 0.0001). BMI and AMH were similar among the categories (p = 0.3063 and p = 0.2533, respectively). There was a high correlation between day 4 categories and PGT-A blastocyst development (p < 0.0001), blastocyst morphology (p < 0.0001), and day of blastulation (p < 0.0001). Also, there was a significant correlation between pregnancy rates and the four categories (p = 0.0192). No correlation was observed when aneuploidy rate and level of aneuploidy were analyzed (p = 0.3614 and p = 0.1490, respectively).

Limitations, reasons for caution

The limited number of embryos analyzed is a limitation of the study, and additional data are being gathered. Multiple embryos of the same patients can be a cofounder of the study and more robust statistical analysis may overcome it.

Wider implications of the findings

The new day 4 embryo score may be useful for embryo selection to PGT-A biopsy. Moreover, the proposed score may help to add valuable information to select the best embryos for transfer.

Trial registration number

not applicable

P–165 Using Artificial Intelligence to Classify Embryo Shape: An International Perspective

Study question

Is a pre-trained machine learning algorithm able to accurately detect cellular arrangement in 4-cell embryos from a different continent?

Summary answer

Artificial Intelligence (AI) analysis of 4-cell embryo classification is transferable across clinics globally with 79% accuracy.

What is known already

Previous studies observing four-cell human embryo configurations have demonstrated that non-tetrahedral embryos (embryos in which cells make contact with fewer than 3 other cells) are associated with compromised blastulation and implantation potential. Previous research by this study group has indicated the efficacy of AI models in classification of tetrahedral and non-tetrahedral embryos with 87% accuracy, with a database comprising 2 clinics both from the same country (Brazil). This study aims to evaluate the transferability and robustness of this model on blind test data from a different country (France).

Study design, size, duration

The study was a retrospective cohort analysis in which 909 4-cell embryo images (“tetrahedral”, n = 749; “non-tetrahedral”, n = 160) were collected from 3 clinics (2 Brazilian, 1 French). All embryos were captured at the central focal plane using Embryoscope™ time-lapse incubators. The training data consisted solely of embryo images captured in Brazil (586 tetrahedral; 87 non-tetrahedral) and the test data consisted exclusively of embryo images captured in France (163 tetrahedral; 72 non-tetrahedral).

Participants/materials, setting, methods

The embryo images were labelled as either “tetrahedral” or “non-tetrahedral” at their respective clinics. Annotations were then validated by three operators. A ResNet–50 neural network model pretrained on ImageNet was fine-tuned on the training dataset to predict the correct annotation for each image. We used the cross entropy loss function and the RMSprop optimiser (lr = 1e–5). Simple data augmentations (flips and rotations) were used during the training process to help counteract class imbalances.

Main results and the role of chance

Our model was capable of classifying embryos in the blind French test set with 79% accuracy when trained with the Brazilian data. The model had sensitivity of 91% and 51% for tetrahedral and non-tetrahedral embryos respectively; precision was 81% and 73%; F1 score was 86% and 60%; and AUC was 0.61 and 0.64. This represents a 10% decrease in accuracy compared to when the model both trained and tested on different data from the same clinics.

Limitations, reasons for caution

Although strict inclusion and exclusion criteria were used, inter-operator variability may affect the pre-processing stage of the algorithm. Moreover, as only one focal plane was used, ambiguous cases were interpoloated and further annotated. Analysing embryos at multiple focal planes may prove crucial in improving the accuracy of the model.

Wider implications of the findings: Though the use of machine learning models in the analysis of embryo imagery has grown in recent years, there has been concern over their robustness and transferability. While previous results have demonstrated the utility of locally-trained models, our results highlight the potential for models to be implemented across different clinics.

Trial registration number

Not applicable

Mortality rate with irinotecan/cisplatin plus concomitant radiation therapy in patients with locally advanced esophageal squamous cell carcinoma

e15560

Background: The cisplatin/irinotecan doublet plus concomitant radiation therapy has been tested in phase II studies as a non-infusional alternative for the treatment of locally advanced esophageal squamous cell carcinoma (ESCC), showing good antitumoral activity and acceptable safety profile. Here, we describe our institutional experience with this combination, specially regarding its toxicity profile. Methods: A series of 26 patients with locally advanced, unresectable ESCC patients treated at our institution between September/2006-January/2008 is reported. Treatment consisted of cisplatin 30 mg/m2 and irinotecan 65 mg/m2 on weeks 1, 2, 4, 5, 8, 9, 11 and 12, and radiation therapy (50,4 Gy, 1,8 Gy/day) on weeks 8–12. Study endpoints were tumor response rate (RECIST), safety profile (CTCAE v3), overall and one year survival. Results: The mean age of patients was 61 years, with stage III disease in 88% of cases and performance status (ECOG) 0–1 in 90% of them. Twenty patients (77%) were available for response, with complete response, partial response/stable disease and progressive disease seen in 40%, 45% in 15% of cases, respectively. With a median follow up of 20 months, the overall survival was 9,2 months and the one-year survival 34,6%. All patients were evaluable for toxicity, with severe (grade 3–4) nausea/vomiting, diarrhea, neutropenia and thrombocytopenia documented in 45%, 33%, 18% and 13% of cases, respectively. Notably, treatment related deaths were observed in 5 cases (19%), mainly due to infectious complications, which led to the closured of this protocol due to excessive toxicity. Conclusions: In spite of its promising antitumor activity, this treatment regimen cannot be recommended for routine use due to its unacceptable treatment-related mortality.

No significant financial relationships to disclose.