Abstract Study question Is morphological grade or presumed low-mosaicism status the most impactful parameter on live birth rate (LBR) when selecting the embryo to transfer in PGT-A cycles? Summary answer Among euploid/presumed low-mosaic embryos, morphology provides a better selection parameter for improving live birth in PGT-A cycles. What is known already NGS-based technologies enable the detection of intermediate chromosomal copy number (CN) values, which are commonly interpreted as embryonic chromosomal mosaicism. So far, studies showed that low morphological grade blastocysts are associated with poorer prognoses. Prospective blinded trials also showed that embryos displaying putative mosaic profiles produce comparable clinical outcomes to uniformly euploid ones. Nonetheless, scientific societies suggest different approaches on embryo prioritisation: PGDIS and COGEN advise a selection based on presence of mosaicism, while ESHRE suggests parallel evaluation of mosaicism and morphology. However, the combined effect of morphology and mosaicism on live birth rate (LBR) hasn’t yet been thoroughly investigated. Study design, size, duration This retrospective multisite cohort analysis is based on a euploid/low-mosaic non-selection study, where the presumed mosaicism was not disclosed to clinics and did not affect the embryo selection procedure. It includes 2,339 consecutive IVF/PGT-A treatments (January2018-December2021) that led to 3,999 transferable blastocysts (euploid or presumed mosaic <50%) and resulted in 2,621 frozen single embryo transfers (SETs). Morphological (i.e., good/poor quality) and chromosomal (i.e., euploid/mosaic) features were assessed. Clinical outcomes were followed up for each subgroup. Participants/materials, setting, methods Morphological grading was assessed using Gardner’s criteria, <BB for poor-quality and >BB for good-quality embryos. PGT-A analysis and CN values calculations were performed on IonTorrentS5 and Ion-Reporter software (ThermoFisher). Based on raw NGS data, CN variations <30% and 30%-50% were internally categorised euploidy and mosaicism, respectively. Nevertheless, embryos belonging to either category were reported to clinics as euploid. The association between morphology, chromosomal status, and LBR was evaluated using Fisher’s exact test and multivariate analysis. Main results and the role of chance In this study population, low-mosaicism had an incidence of 15.8% (N = 630/3,999; 95%CI:14.7-16.9). Overall, LBR and miscarriage rate did not differ between euploid and presumed mosaic SETs: 45.0% vs. 44.8% (P=NS), and 11.2% vs. 13.2% (P=NS), respectively. Also, euploid and presumed mosaic embryos showed similar incidence of poor morphology (7.4% vs. 9.7%; P=NS). In the good-quality category, LBR and miscarriage rate were 46.2% vs. 44.8% (P=NS), and 11.3% vs. 13.1% (P=NS) for euploid and presumed mosaic embryos, respectively. In the poor-quality category, LBR was 21.0% vs. 23.8% (P=NS), respectively; whilst miscarriage rates could not be compared due to insufficient number of cases across the whole morphological category (N = 4/33). Accordingly, the focused scenario that employed poor-quality uniformly euploid embryos instead of good-quality mosaic embryos showed the highest impact on LBR (21.0% vs. 44.8%; P < 10-5). Finally, the proportion of cases where both selection strategies (i.e., mosaicism vs. morphology) could be applied accounted for 6.1% (N = 44/717; 95%CI:4.6-8.1) of all cycles where >1 transferable embryo was available (N = 717/2,339). Limitations, reasons for caution The results were obtained through the analysis of raw NGS data independent from any proprietary diagnostic algorithm or chromosome-specific consideration commonly used by PGT-A laboratories. These findings concern putative whole-chromosome mosaic aneuploidies only and cannot be extended to segmental mosaic configuration. Follow-up studies in prenatal diagnosis are needed. Wider implications of the findings This study suggests that, when selecting embryos for transfer, morphological grading should be considered as a higher criterion than low-mosaic conformation (i.e., mosaicism <50% of copy-number). Reporting mosaicism based on intermediate copy-number categorization ranges <50% appears to provide no clinical utility in the current technological landscape. Trial registration number not applicable
Read full abstract