L26/P-299 Uncovering the hidden impact of chromosomal anomalies on embryo developmental competence and live birth compatibility
Human Reproduction · 01.07.2026 · 41 (Supplement_1)
DOI: 10.1093/humrep/deag083.635 ↗Kurz gefasst
Fachartikel in der Zeitschrift Human Reproduction, online erschienen am 08.07.2026. Verfasst von D. Lani und 8 weiteren Autor:innen.
Abstract (Originalsprache)
Study question Can an AI model trained for predicting implantation identify differences in chromosomal anomalies when stratified by their compatibility with live birth? Summary answer Embryos with chromosomal anomalies incompatible with live birth show lower AI-driven embryo scores compared to those with anomalies potentially compatible with live birth. What is known already Conventional embryo assessment relies primarily on static morphology and morphokinetics to select competent embryos. However, it is only until invasive PGT-A is performed that we can classify embryos based on their chromosomal load. Evidently, not all chromosomally abnormal embryos exhibit impaired development, and developmental competence is not determined solely by ploidy status. It has been suggested that certain aneuploidies compatible with live birth may progress through development with morphokinetic patterns similar to euploid embryos. An AI-driven embryo quality score based on integrated time-lapse morphokinetic data has emerged as a complementary approach, enabling identification of developmental differences before PGT-A biopsy. Study design, size, duration Retrospective observational study analysing 1,225 blastocysts from 637 patients undergoing autologous preimplantation genetic testing for aneuploidy (PGT-A) cycles. Embryos were evaluated using an AI-derived embryo quality model trained for predicting implantation (AI score, ranged 1-10) based on integrated time-lapse data. The study assessed associations between chromosomal anomalies compatible and incompatible with live birth, and the AI score as a proxy for developmental competence. Data was retrospective and did not influence embryo selection or treatment decisions. Participants/materials, setting, methods Blastocysts were classified as euploid (n = 704), and aneuploid either with anomalies potentially compatible with live birth (trisomy 13, 18, 21, XXX, XXY and X0; n = 50), or with anomalies incompatible with live birth (all other chromosomal configurations) (n = 471). Beta mixed-effects regression models were applied to account for patient-level clustering and AI score distribution. Discriminative performance for chromosomal status was evaluated using mixed-effects logistic regression and area under the curve. Main results and the role of chance Overall mean maternal age was 38.8 (±4.4). As for euploid and overall aneuploid embryos, mean maternal age was 36.9 (±4.8) and 40.0 (±3.9), respectively. Mean AI embryo quality scores were 7.17 (±0.31) for euploid embryos, 6.79 (±0.35) for aneuploid embryos with anomalies incompatible with live birth and 7.52 (±0.31) for embryos with anomalies potentially compatible with live birth. Embryos with anomalies potentially compatible with live birth showed no statistically significant difference in the likelihood of achieving high AI scores relative to euploid embryos (p = 0.339). In contrast, embryos with anomalies incompatible with live birth had lower likelihood of achieving a high AI embryo quality score compared with euploid embryos (OR = 0.84; 95% CI: 0.72–0.97; p = 0. 022). The AI score is able to discriminate between euploid embryos and those with anomalies incompatible with live birth (AUC = 0.623 95% CI:0.613-0.647). Limitations, reasons for caution This was a single-centre retrospective study with a limited number of embryos carrying abnormalities potentially compatible with live birth. Developmental competence and its correlation with implantation or live birth outcomes were not assessed, limiting clinical extrapolation and generalizability. Wider implications of the findings An AI model trained for predicting implantation can discriminate between embryos compatible with live birth and aneuploid embryos incompatible with live birth. Findings suggest that AI scores may be used to prioritize biopsies in patients with increased risk of aneuploidy to increase the chances of identifying euploid embryos. Trial registration number No
Zitieren
APA
Lani, D., Fraire-Zamora, J., Barragán, M., Torra-Massana, M., Rodríguez-Aranda, A., Kelley, K., Wiemer, K., Ben-Meir, A., & Pini, P. (2026). L26/P-299 Uncovering the hidden impact of chromosomal anomalies on embryo developmental competence and live birth compatibility. Human Reproduction, 41(Supplement_1). https://doi.org/10.1093/humrep/deag083.635
Vancouver
Lani D, Fraire-Zamora J, Barragán M, Torra-Massana M, Rodríguez-Aranda A, Kelley K, et al. L26/P-299 Uncovering the hidden impact of chromosomal anomalies on embryo developmental competence and live birth compatibility. Human Reproduction. 2026;41(Supplement_1). doi:10.1093/humrep/deag083.635
BibTeX
@article{Lani2026,
author = {Lani, D. and Fraire-Zamora, J.J. and Barragán, M. and Torra-Massana, M. and Rodríguez-Aranda, A. and Kelley, K. and Wiemer, K. and Ben-Meir, A. and Pini, P.},
title = {L26/P-299 Uncovering the hidden impact of chromosomal anomalies on embryo developmental competence and live birth compatibility},
journal = {Human Reproduction},
year = {2026},
volume = {41},
number = {Supplement_1},
doi = {10.1093/humrep/deag083.635}
}
EUVIMED