37. PREIMPLANTATION GENETIC TESTING: TO THE OPEN QUESTIONS ABOUT DATA INTERPRETATION OBTAINED BY MASSIVE PARALLEL SEQUENCING

Abstract

Introduction the accuracy results of preimplantation genetic testing of embryos for aneuploidy (PGT-A) by massive parallel sequencing directly depends on the effectiveness of the quality control. The aim of this study was to analyze the external and intralaboratory system of the quality assessment to optimize protocols, algorithms of troubleshooting and selection of the most informative objective criteria for the interpretation of the PGT-A results. Materials and methods we analyzed the quality criteria more than 2000 human bioptates of the trophectoderm cells. PGT-A was performed using VeriSeq reagent kits. Sequencing was performed with the Illumina Miseq instrument. Sequencing data analysis was done by BlueFuse Multi Software. External quality assessment of PGT-A was carried out through participation in the program of interlaboratory comparisons of the consortium UK NEQAS. Results with the aim of obtaining optimal performance of the sequencing was carried out intralaboratory validation of methods of libraries preparation for sequencing. In 10% of cases, genomic profiles, which were difficult to interpret, were obtained with optimal performance criteria of whole genome amplification (WGA) and the quality of subsequent sequencing. One of the problems is the suspicion of segmental aneuploidies in a mosaic version. Recognition of false mosaicism helps, in particular, the absence of chromosome break points, the stereotypical pattern of chromosome copy number variations (CNV), which may be associated with difficulties in reading of GC-rich DNA regions. As a result of participation in the program of external quality assessment of a consortium UK NEQAS we found that partial DNA degradation in the sample makes it impossible to detect trisomies. Furthermore, studying the PGT-A statistical results of different laboratories, attention is drawn to a significant spread in the frequency of mosaicism detected by them. This may be primarily due to the complexity of interpreting so-called noisy CNV charts. When analyzing the sequencing results using BlueFuse Multi software, the researcher is able to estimate for each sample the total overall noise score, as well as the Region Confidence value for each chromosome. However, the low information content of these parameters to estimate the number of copies of a single chromosome often generates subjectivity in the decision of the expert. In this regard, the most important task is to minimize the technical factors that can lead to the appearance of intense artifacts in CNV-charts, which complicates their interpretation. We, together with the adjacent laboratory in Moscow, have developed a research proposal, the implementation of which will help to shed light on the technical issues of all embryo and PGT-A stages currently remain elusive. Conclusions the main reason for obtaining difficult to interpret genomic profiles is the low quality of the original DNA and as a consequence – false positive or false negative results of PGT-A. To prevent false-positive and false-negative results of PGT-A, a complex analysis of objective quality criteria of different stages of PGT analysis is necessary due to the lack of the information content of each individual factor in case of moderately pronounced DNA degradation in the sample.