P-727 New methods reveal the true incidence of DNA contamination in PGT-A samples for the first time and avoid errors that could result in serious misdiagnoses

Abstract

Abstract Study question What is the occurrence of contamination in embryo biopsy samples and could this lead to incorrect interpretation of preimplantation genetic testing for aneuploidy (PGT-A) results? Summary answer On average, contamination affects 0.4% of biopsy samples, but can be significantly more common in some clinics. Misdiagnosis can occur when contamination is not detected. What is known already Until recently, most commercially available platforms for PGT-A have utilised whole genome amplification followed by sequencing of a random selection of DNA fragments scattered across the genome using next generation sequencing (NGS). However, the simple quantitative measurements of DNA fragments derived from each chromosome, provided by such methods, cannot reveal when a biopsy sample is contaminated with non-embryonic DNA. Negative controls are seldom used during PGT-A and are inadequate as they do not evaluate contamination in the actual tube containing the biopsy specimen. The extent to which failure to detect DNA contamination is a problem for PGT-A is unknown. Study design, size, duration This was a retrospective study involving analysis of 49,287 trophectoderm biopsy samples that underwent PGT-A over a three-year period. Embryos found to have a contaminated biopsy specimen typically underwent a second biopsy. In such cases, results from the two samples were compared to ascertain whether the contaminated sample would have been misdiagnosed if the analysis had been restricted to examining only the relative chromosome copy number, as is the case for most NGS-based PGT-A methods. Participants/materials, setting, methods All trophectoderm biopsies underwent targeted DNA amplification and next generation sequencing using a highly validated PGT-A method that evaluates the relative chromosomal copy number, similar to traditional PGT-A methods, but combines this with analysis of variations in DNA sequence (single nucleotide polymorphisms - SNPs). The genotype of each SNP, and the relative quantity of DNA fragments containing each of the different alleles, allows detection of otherwise invisible states, such as triploidy, haploidy, and contamination. Main results and the role of chance From the 49,287 TE biopsies analysed, contamination with non-embryonic DNA was detected in 218 (0.44%). There was variation in the rates of contamination between the 25 clinics that provided samples, ranging between 0% and 1.5%. Additionally, one clinic had a contamination rate of 7.7%, but the number of biopsies derived from that site was considered too small for reliable evaluation (n = 26). 156 of the embryos with a contaminated biopsy specimen underwent secondary biopsy (71.5%), allowing the relative chromosome copy number result from the contaminated specimen to be compared to that obtained from an uncontaminated sample. The results were split into three categories: 1) no change in interpretation between the first (contaminated) and second biopsy specimens; 2) false positive – the contaminated sample was euploid but would have been wrongly interpreted as triploid and would have been erroneously discarded, potentially impacting the patients chances of achieving a pregnancy; 3) false negative – the contaminated sample was fully aneuploid but would have been incorrectly classified mosaic or euploid and could have been eligible for transfer, potentially leading to implantation failure or abnormal pregnancy. 19% of contaminated samples gave a false negative result, while 24% gave a false positive, appearing to be triploid. Limitations, reasons for caution It is not possible to determine the origin of contaminants with certainty without having DNA from the contamination source for comparison. Additionally, we were unable to conclude whether contamination is more likely to occur in IVF or ICSI cycles as only 3% of samples were fertilized using IVF. Wider implications of the findings Contamination detection during PGT-A is important to prevent misdiagnosis of embryos. Misclassification due to undetected contamination can lead to discard of potentially viable embryos. It can also lead to the transfer aneuploid embryos, wrongly classified as mosaic, which could lead to increased rates of implantation failure, miscarriage and aneuploid pregnancy. Trial registration number Not applicable