Key points are not available for this paper at this time.
Abstract Study question This study aims to validate a PGT-A platform to simultaneously sequence and genotype embryos to detect triploidy as well as high and low frequency mosaicism. Summary answer Previda Plus platform validation establishes the capability to concurrently identify aneuploidy, triploidy and mosaicism with high and low frequencies at a significantly reduced cost. What is known already Low pass sequencing platforms can identify with great accuracy monosomy and trisomy karyotypes from embryo biopsies by comparing DNA quantities across chromosomes, however, these platforms exhibit limitations in identifying triploidy since all chromosomes exhibit uniform concentration. Triploidy detection necessitates genotyping involving the detection of highly variable SNPs to distinguish different haplotypes. Intermediate copy number may represent putative mosaicism. Recent publications show that embryos with low-level mosaicism demonstrate comparable implantation rates to euploid embryos, therefore validating cut offs to differentiate between low-level and high-level mosaics from euploid and aneuploid embryos is necessary. Study design, size, duration To validate triploidy detection, we measured allele frequencies (AF) using a high frequency SNP panel. The SNP frequency profile of a diploid exhibits a mix of AF of 100% and 50%, while a triploid exhibits AF of 100%, 66%, and 33%. To validate mosaicism, cell lines that were diploid, trisomic, or monosomic for a specific chromosome were mixed in pairs in increments of 10% (From 0%:100% line-1:line-2, to 100%:0%) with multiple iterations of each combination. Participants/materials, setting, methods Fifteen cell lines (triploid or diploid) underwent amplification using a 384 highly variable SNP panel. Pooled SNP and PGT-A products underwent processing in a single sequencing workflow. Ploidy Scores (PS) were derived from the total number of SNPs, 66% and 33%, and cutoff values were established to discriminate between diploid and triploid. For mosaicism validation, thirteen different trisomic cell lines were mixed in 10% ratio increments with 15 different monosomic cell lines. Main results and the role of chance Amplified samples exhibited distinct separation in ploidy scores between diploidy (PS of 0.91 to 1.85), and triploidy samples (PS of 0.28 to 0.61) with no overlap of boxplots. Specificity, sensitivity, and accuracy were 100%. For mosaicism validation, mixing different ratios of cell lines featuring monosomy and cell lines for a trisomy for different chromosomes resulted in ratios of monosomy: disomy for one chromosome and trisomy: disomy for the other. We assessed 32 combinations of cell lines, each one in 10% increments of each cell line from 0:100% to 100%:0%. Mixes with 40% trisomy mosaicism exhibited a 2.35 median ploidy, while 50%, 60%, 70% and 80% mixes show 2.45, 2.50, 2.60, and 2.70, respectively. Mixes with 40% monosomy mosaicism exhibited a median ploidy of 1.6, while 50%, 60%, 70% and 80% mixes showed 1.50, 1.45, 1.35, and 1.25, respectively. The literature recommends considering 40% mosaicism as euploid and 80-100% as aneuploid, therefore, ploidy scores below 2.35 are deemed euploid, ploidy scores at or above 2.35, but below 2.6 (40%-60%), are classified as low level mosaic, and ploidy scores equal to or above 2.6, but below 2.7, are classified as high mosaic. Embryos equal to or above 2.8 (80%+) are considered aneuploid. Limitations, reasons for caution This test is not designed to screen for tetraploidy (92 XXYY) and has not undergone validation for it nor haploidy. Validation procedures solely utilized cell lines. Reanalysis of the remainder of embryos deemed triploid or high-rate mosaic using this validation results are currently underway. Wider implications of the findings This Previda Plus platform combined with proprietary chemistry provides a rapid, cost effective, and a streamlined workflow for aneuploidy, triploidy, and mosaicism for use in PGT-A of blastocysts biopsies. Trial registration number not applicable
Kim et al. (Mon,) studied this question.