Parental and cell-division origin analysis to reduce false-positives in mosaic embryos for preimplantation genetic testing

Study question: Can parental and cell-division origin analysis identify the false-positive chromosomal aberrations in mosaic embryos detected during preimplantation genetic testing for aneuploidy (PGT-A)? Summary answer: Parental origin analysis reclassified over half of mosaic embryos as euploid to reduce diagnostic uncertainty, and cell-division origin analysis effectively differentiated mitotic from meiotic errors to guide evidence-based mosaic embryo transfer. What is known already: Mosaic embryos pose significant challenges in PGT-A, with reported detection rates varying widely (2-35.6%) due to biological and technical factors, but current methods lack reliable approaches to distinguish true mosaicism from technical artifacts. Decisions regarding mosaic embryo transfer remain conservative and hampered by inadequate understanding of the origins of errors (mitotic vs meiotic) and limited data on long-term neonatal outcomes. Study design size duration: In this retrospective study, we analyzed 9062 PGT-A results and 8645 amniocentesis samples from 2021 to 2024 to investigate the difference in mosaicism rates between PGT treatment and prenatal diagnosis. An analysis of parental and cell-division origins was performed on 1221 consecutive results from PGT-A and PGT for monogenic disorders (PGT-M) from 259 patients across 304 treatment cycles in 2024. Multi-site re-biopsies of 36 donated embryos and the clinical outcomes of 19 mosaic embryo transfers were analyzed in 2024 and 2025. Participants/materials setting methods: An innovative algorithm, termed parental haplotype trace (PH-trace), was developed to identify the genetic origin of chromosomal aberrations for mosaic verification. Briefly, biallelic homozygous single-nucleotide polymorphisms (SNPs) in the parental genome exhibit equal allelic frequencies in euploid embryos. When chromosomal aberrations occur, these SNPs show an allelic bias toward either the maternal or paternal genome. We defined the ratio of maternal-biased SNPs to paternal-biased SNPs as the uneven score to quantitatively assess the parental origin of chromosomal aberrations. Receiver operating characteristic (ROC) curve analysis based on uneven scores from euploid and aneuploid embryos was used to determine critical thresholds for identifying false-positive mosaic embryos. Additionally, heterozygous SNPs in the parental genome were used to determine the cell-division origin of chromosomal aberrations. To validate our findings, we performed multi-site re-biopsies of aneuploid and mosaic embryos. Main results and the role of chance: The prevalence of mosaicism differed significantly between PGT-A and prenatal diagnosis (12.2% vs 0.9%, P < 0.001). Parental origin analysis based on PH-trace reclassified 52.6% of mosaic embryos as euploid and increased the overall rate of usable embryos by 8.6%. Mitotic errors accounted for the majority of true mosaic cases, providing critical guidance for embryo transfer prioritization. Re-biopsy validation of donated embryos revealed that 94.1% of predicted false-positive copy-number variations (CNVs) were not detected in subsequent samples, whereas 71.4% of parental-biased CNVs were repeatedly detected. Among mosaic embryos resulting in live births, 66.7% were false-positive, and 22.2% originated from mitotic errors. Limitations reasons for caution: Our two-tiered analytical approach relies on a sufficient number of informative SNPs in parental and embryonic genomes (e.g. over 30 informative SNPs per chromosomal aberration). Additionally, the clinical implications of this method require further validation through long-term follow-up studies. Wider implications of the findings: This study presents an effective strategy to identify false-positive mosaic embryos, thereby improving embryo utilization in clinical treatment. By elucidating the cell-division origin of mosaicism, our findings provide embryologists with evidence-based criteria for prioritizing embryo transfers. Furthermore, this approach may reduce the number of IVF cycles and associated costs for patients with limited euploid embryo availability. Study funding/competing interests: This work was supported by the National Natural Science Foundation of China (82371728), the Chongqing's Mid-young Medical Elite Talent Project (YXGD202555), the Key Projects of the Collaborative Medical Research between Science and Health Commission of Chongqing (2026ZDXM004), the Chongqing Municipal Technological Innovation and Application Development Special Project (CSTB2022TIAD-KPX0146), and the Chongqing Nature Science Foundation (CSTB2023NSCQ-MSX0443). The authors have no competing interests to declare. Trial registration number: N/A.