ChromInst: a multicentre evaluation of robustness in aneuploidy and structural rearrangement testing.

Background: Preimplantation genetic testing for aneuploidy and for chromosomal structural rearrangement (PGT-A/-SR) can improve clinical pregnancy rates and live birth rates, and shorten the time to pregnancy. The large-scale statistics on their efficacy and accuracy across different centres, as well as the frequency of abnormalities for each chromosome, will provide a valuable supplement to previous research.

Methods: Patients who had PGT-A or -SR procedures at five reproductive centres from 2018 to 2022 were recruited based on PGT-A/-SR indications. ChromInst and next-generation sequencing (NGS)-based PGT technology were utilised to detect copy number variations in embryos. Sequencing data metrics such as median absolute pairwise difference (MAPD) and detection success rate were analysed to evaluate the robustness of ChromInst. To assess ChromInst's accuracy, the chromosomal results from amniocentesis, abortions, and neonatal blood was as the gold standard for negative PGT results; the fluorescence in situ hybridisation (FISH), which was performed on embryos that identified as aneuploid through PGT was as the gold standard for positive PGT results. The frequency of abnormalities in each chromosome was also explored in aneuploid embryos.

Results: A total of 5,730 embryos were tested from 1,015 patients in the study, 391 of whom had PGT-A and 624 of whom had PGT-SR. 99.5% (5,699/5,730) of the embryos had an NGS sequencing MAPD value < 0.25, and 99.3% (5,689/5,730) of the embryos achieved successful PGT-A/-SR detection. Compared with the gold standard, the concordance of negative PGT-A/-SR results was 99.8% (506/507), and that of positive results was 99.8% (1,123/1,125). The euploidy rate in the PGT-A population was 45.9% (981/2,135). The proportion of euploid + balanced embryos was highest among couples with non-polymorphic inversions (44.6%, 152/341), followed by those with Robertsonian translocations (39.0%, 293/752), and lowest among those with reciprocal translocations (22.5%, 483/2,143). Chromosomes 16, 22, and 15 had the highest frequency of autosomal trisomies among the embryos from PGT-A patients, while chromosomes 16, 22, and 21 had the highest frequency of monosomies. High-frequency chromosomes with de novo chromosomal abnormalities for trisomies and monosomies were similar in the PGT-SR patients to those in the PGT-A patients.

Conclusions: ChromInst-based PGT-A/-SR could accommodate operational variations among different clinical centres, ensuring accurate results through robust and stable detection performance. Prior to PGT-A/-SR, more trustworthy data could be provided to support the genetic counselling.