STUDY QUESTION:Can a universal diagnostic test (Karyomapping) be applied for preimplantation genetic testing for multiple monogenic disorders (PGT-M) and what is the misdiagnosis rate? SUMMARY ANSWER:Among 9020 cases of PGT-M, >1000 different disorders were diagnosed by Karyomapping; independent validation of >70% of cases did not detect a misdiagnosis. WHAT IS KNOWN ALREADY:PGT-M, first performed in 1992, has been used for ∼40 000 clinical cases worldwide. A limiting factor in direct testing for disease mutations, however, is the need to design assays specific for each affected allele. Karyomapping, based on haplotype phasing using SNP microarrays, was developed in 2010 as a single, method tracing inheritance of any monogenic disorder. Karyomapping eliminates the impact of allele drop-out and DNA contamination on test accuracy and facilitates a short work-up time as the same assay platform is used for every case. STUDY DESIGN, SIZE, DURATION:Here, we used Karyomapping on a large PGT-M series from one diagnostic base from January 2014 to December 2021. PARTICIPANTS/MATERIALS, SETTING, METHODS:The 9020 individual Karyomapping cases were performed in three CooperSurgical genetic testing laboratories, in Livingston NJ, Michigan, or London (UK). All cases involved trophectoderm biopsy with embryo vitrification. DNA from cheek brush samples was obtained from both parents and an affected reference family member where possible. Genomic DNAs and that of whole genome amplified DNA from embryo biopsies were subjected to SNP microarray. Karyomapping was performed according to manufacturer's instructions by first importing into BlueFuse Multi software. Inheritance was determined as to where at-risk allele(s) were inherited, with 10 supporting 5' and 3' Key SNPs in a 2 Mbp flanking window. Wherever possible, direct mutation testing was performed using Sanger sequencing. MAIN RESULTS AND THE ROLE OF CHANCE:A total of 1017 unique disorders were detected from mutations in 912 genes. Validation of 4120 mutations was possible in 73% of cases by direct sequencing, which confirmed that all diagnoses that could be assayed were accurate. LIMITATIONS, REASONS FOR CAUTION:Karyomapping can be limited by the availability of a reference, as well as parental genomic DNA, and some loci near the telomere may be more difficult to detect because of the limitations of the SNP array rather than the Karyomapping algorithm. Of the 27% of cases where we could not confirm the findings, we cannot comment on the misdiagnosis rate. WIDER IMPLICATIONS OF THE FINDINGS:Karyomapping is now the single most used approach for PGT-M. As new approaches increasingly involve DNA sequencing, PGT for all genetic disease becomes possible by encapsulating the principles of Karyomapping and incorporating chromosome copy number analysis. TRIAL REGISTRATION NUMBER:N/A. STUDY FUNDING/COMPETING INTEREST(S):This research was funded by CooperSurgical. The PhD programs of A.S. and O.W. were supported by CooperSurgical (paid to institution). A.S. has received travel support and IT equipment from CooperSurgical. O.W. has received travel support and provision of a company laptop from CooperSurgical. L.X., P.C., E.B., and T.G. are employees of and hold stock/share ownership in CooperSurgical. N.-N.G. is an employee of, has received meeting registration fees from, and holds stock/share ownership in CooperSurgical. L.R. is an employee of CooperSurgical. D.K.G. has received consulting fees and travel support from CooperSurgical. P.E. has nothing to declare.

To demonstrate the universality of Karyomapping, through a large series of preimplantation genetic testing for monogenic diseases (PGT-M) cases.

To identify factors associated with variable PGT-A outcomes, particularly failure to produce a result, between IVF clinics

To assess the accuracy of Karyomapping for PGT-M through a retrospective analysis involving comparison with direct mutation and linkage analysis.

No test is 100% accurate. PGT laboratories should investigate reported concerns regarding specific PGT results. Outcomes of these investigations should be shared to communicate potentially preventable causes of discrepant PGT-A results.

Introduction It is relatively common for human preimplantation embryos produced during the course of IVF treatments to contain two or more cytogenetically distinct cell lines. This phenomenon, known as chromosomal mosaicism, can result in either a mixture of euploid and abnormal cells, or abnormal cells bearing distinct chromosomal gains and losses. NGS allows mosaicism to be detected with much greater sensitivity than earlier preimplantation genetic testing (PGT-A) methods. The application of NGS to trophectoderm biopsies, taken from embryos before transfer to the uterus, has provided insight into the clinical impact of mosaicism. However, because some mosaic embryos are have produced successful pregnancies, it may be appropriate to consider transfer of mosaic embryos in the absence of fully euploid embryos and following patient counseling. Materials and Methods Samples were amplified by SurePlex, sequenced with VeriSeq PGS assay (Illumina) on MiSeq (Illumina) and analyzed with BlueFuse Multi analysis (Illumina). Embryos with at least one fully aneuploid chromosome or one fully aneuploid chromosome and one mosaic chromosome were called aneuploid. Embryos with 3 chromosome abnormalities were designated complex abnormal. Mosaicism was reported when embryos had 20-80% abnormal sequencing profiles. Embryos with low degrees of mosaicism in sequencing profiles were designated as ‘Low –Level Mosaics’. When embryos had high degree of mosaicism, they were reported as ‘High- Level Mosaics’. Design Retrospective analysis of PGS procedures involving TE biopsy and NGS performed by laboratories serving over 250 fertility clinics over three years. A total of 3951 embryos resulting from 952 IVF cycles were analyzed. Results Gross aneuploidy (includes complex abnormals and polyploidies) rates in egg donors (n=402) was 25.62%. Euploid rate in egg donors was 51.00%. 20.90% embryos from egg donors were mosaic, with 11.94% being classified as High-level mosaic and 8.96% being classified as Low-level mosaic. In the non-egg donor group (n=3,545 embryos) aneuploidy (includes complex abnormals and polyploidies) ranged at 44.23%. Euploid rate in non-egg donors was about 36.47%. About 15.94% embryos were mosaic with no trend observed in advanced maternal age and mosaicism. About 8% of all embryos in the non-egg donor group were High-level mosaics while and equal percentage of embryos were Low- level mosaics. A total of 33% cycles (n=317 cycles) resulted in cohorts which had zero euploid embryos. Of these 317 cycles however, 112 cycles (35%), had at least one mosaic embryo to transfer. Of the zero-euploid group a total of 52 cycles (16%) resulted in a Low-level mosaic for patients who otherwise do not have any normal/euploid embryos to transfer. Only 7 cycles from the donor group resulted in cohorts with zero euploid embryos. Conclusion The ability of PGT-A by NGS to delineate between high-level and low-level mosaic embryos allows patients without euploid embryos for transfer the ability to identify mosaic embryos with the greatest likelihood of producing a successful pregnancy. Approximately 8% of all embryos tested by PGT-A are low-level mosaic, with 16% of cycles without identified euploid embryos producing at least one low-level mosaic embryo. The results herein can be used to counsel patients and guide expectations.

RESEARCH QUESTION:What are the incidence and patterns of meiotic trisomies and recombination separately and in relation to each other at the blastocyst stage via single nucleotide polymorphism genotyping combined with array comparative genomic hybridization.DESIGN:Single nucleotide polymorphism microarrays were carried out on a total of 1442 blastocyst stage embryos derived from 268 fertile couples undergoing preimplantation genetic diagnosis for the purposes of avoiding transmittance of known single gene disorders to their offspring; 24-chromosome aneuploidy screening via array comparative genomic hybridization was carried out in parallel.RESULTS:One hundred per cent of meiotic trisomies identified in these embryos were of maternal origin and their incidence increased significantly with advancing maternal age (P < 0.0001). A total of 55.8% of meiotic trisomies were meiosis I-type and 44.2% were meiosis II-type. Certain chromosomes were affected more by meiosis I-type errors, whereas others experienced more meiosis II-type errors. A detailed recombination analysis was carried out for 11,476 chromosomes and 17,763 recombination events were recorded. The average number of recombination sites was 24.0 ± 0.3 for male meiosis and 41.2 ± 0.6 for female meiosis (autosomes only). Sex-specific differences were observed in the locations of recombination sites. Comparative analysis conducted between 190 euploid embryos and 69 embryos presenting maternal meiotic trisomies showed similar recombination rates (P = 0.425) and non-recombinant chromatid rates (P = 0.435) between the two categories; differences, however, were observed when analysing embryos affected with specific maternal meiotic trisomies.CONCLUSIONS:This study yielded unique data concerning recombination and the origin of aneuploidies observed during the first few days of life and provides a novel insight into these important biological processes.

Study question Do external factors affect euploidy in egg donor cycles? Summary answer The study demonstrates that during human assisted reproduction, embryonic chromosome abnormalities may be partly iatrogenic. What is known already Chromosome abnormalities have been linked in the past to culture conditions such as temperature and Ph variations, as well as hormonal stimulation. Those reports were performed with older screening techniques (FISH), or ART methods no longer in use, and the subjects studied were not a homogeneous group. Study design, size, duration A total of 1645 donor oocyte cycles and 13 282 blastocyst biopsies from 42 fertility clinics were included in this retrospective cohort study. Samples from donor cycles with PGS attempted between September 2011 and July 2015 were included. Participants/materials, setting, methods PGS cycles from multiple fertility clinics referred to Reprogenetics (Livingston, NJ) that involved only oocyte donation were included in this study. Testing was performed by array comparative genomic hybridization (aCGH). Ploidy data were analyzed using Generalized Linear Mixed Models with logistic regression using a logit link function considering a number of variables that represent fixed and random effects. Main results and the role of chance Euploidy rate was associated with the referring center and independent of almost all the parameters examined except donor age and testing technology. Average euploidy rate per center ranged from 39.5 to 82.5%. The mean expected rate of euploidy was 68.4%, but there are variations in this rate associated with the center effect. Limitations, reasons for caution Data set does not include details of the donor selection process, donor race or ethnic origin, ovarian reserve or ovarian responsiveness. Due to the retrospective nature of the study, associations are apparent, however, causality cannot be established. Discrepancies in regard to completeness and homogeneity of data exist due to data collection from over 40 different clinics. Wider implications of the findings This is the first study to show a strong association between center-specific ART treatment practices and the incidence of chromosome abnormality in human embryos, although the meiotic or mitotic origin of these abnormalities could not be determined using these technologies. Given the widespread applications of ART in both subfertile and fertile populations, our findings should be of interest to the medical community in general as well as the ART community in particular. Study funding/competing interest(s) No external funds were used for this study. S. Munne is a founding principle of Reprogenetics/current employee of Cooper Genomics. M Alikani's spouse is a founding principle of Reprogenetics/current consultant for Cooper Genomics. The remaining authors have no conflicts to declare.

aCGH is an established chromosome screening technique in use since 2010. All chromosomes are studied, with resolution as low as 6 MB. Over 91679 embryos and 44801 cycles have been screened using this method. In this study we present follow-up of these cycles. Follow-up data was requested from 202 centers for embryos tested via aCGH from 2011-2016. Day 3 and day 5 were reported separately. From 11/2010-4/2016, aCGH was performed on 2141 blastomere cycles (16862 embryos) and 42660 blastocyst cycles (74817 embryos) for a total of 91679 embryos. Each sample was amplified and tested via aCGH (24sure, Illumina). Follow-up was requested from IVF centers on 18,046 cases for patients with at least one euploid embryo. For this study, a pregnancy includes biochemical, ectopic, and pregnancies that ended in spontaneous abortion, as well ongoing or delivered. aCGH data was obtained on 1353 blastomere cycles (1434 embryos) and 6010 blastocyst cycles (6944 embryos). Overall implantation rate for blastomere and blastocyst was 41.7% and 63.9% respectively. Pregnancy rate per cycle was 43.2% and 49.6% respectively, with pregnancy rate per transfer at 59.1% for blastomere and 60.5% for blastocyst. Miscarriage rate was 9.7% and 10.7% respectively, and ongoing pregnancy rate was 51.7% and 55.9% for blastomere and blastocyst samples. For blastocyst biopsy we further stratify the data by age in the following table: For blastocyst biopsy cases, implantation rate across entire maternal age (MA) range was constant. This indicates that once an euploid embryo is replaced it implants well at any MA and further solidifies that loss of implantation with advancing MA is due to chromosome abnormalities. There are patients with euploid embryos that have not yet had a transfer, diluting the pregnancy rate per cycle, however, the pregnancy rate per transfer was calculated based on patients who have had transfers and is a better representation of overall success. For ongoing pregnancy rate, there is a significant difference (p.<001) between blastomere and blastocyst samples reflecting the expected improved outcome of a blastocyst biopsy.Tabled 1Egg Donor< 3535-3738-4041-42>42Implantation42%41%47%38%43%41%Pregnancy Rate/Transfer66%62%50%58%58%51%Miscarriage Rate9%8%9%13%12%16%Ongoing Pregnancy Rate67%57%51%53%53%43% Open table in a new tab

BackgroundRecent randomized trials have indicated the benefit of using comprehensive chromosome screening (CCS) methodologies and their potential of enhancing in vitro fertilization (IVF) success rates. Most recently a new methodology, known as next generation sequencing (NGS), was introduced into preimplantation genetic diagnosis (PGD) for CCS. NGS has been shown to be advantageous over other CCS methodologies, potentially offering higher precision of diagnosis [e.g. detection of abnormalities present only in some cells of the trophectoderm biopsied (mosaicism); a feature that is not available using other CCS techniques].ObjectiveTo clinically apply a methodology that combines PGD for single gene disorders with CCS-NGS of blastocysts through utilization of a single trophectoderm biopsy.Materials and methodsA total of 52 embryos derived from 11 different patients were assessed in this study. A single trophectoderm biopsy was carried out on day 5/6 of preimplantation development. Multiple displacement amplification was performed on the samples and aliquots from each amplified product were used to perform Karyomapping (Illumina, USA) for PGD of single gene disorders and NGS for CCS. The VeriSeq PGS assay and a MiSeq desktop sequencer (Illumina) were used for NGS. The BlueFuse Multi analysis software (Illumina) was utilized for interpretation of results.Results60% of the embryos were found to be either unaffected or carriers of the disorder being tested while, 40% of the embryos assessed were found to be euploid via NGS. 31.1% of the overall embryos assessed were seen to have at least one mosaic abnormality, while 8.9% of the embryos were determined to have only a single mosaic abnormality. Furthermore, segmental abnormalities were detected in 7 of the embryos tested and ranged from 7.8 to 104.6 megabases in size. A total of 16 embryos were found to be at low risk for the single gene disorder assessed and aneuploidy and could therefore be considered for transfer.ConclusionsThis study documents the successful clinical application of PGD for single gene disorders in combination with CCS-NGS using a single trophectoderm biopsy. The combination of powerful and highly accurate methodologies such as Karyomapping for single gene disorders and NGS for CCS, is expected to benefit patients undergoing IVF/PGD and can enhance their chances for a successful cycle leading to a healthy pregnancy.Financial supportInstitutional support.ReferencesNon applicable. BackgroundRecent randomized trials have indicated the benefit of using comprehensive chromosome screening (CCS) methodologies and their potential of enhancing in vitro fertilization (IVF) success rates. Most recently a new methodology, known as next generation sequencing (NGS), was introduced into preimplantation genetic diagnosis (PGD) for CCS. NGS has been shown to be advantageous over other CCS methodologies, potentially offering higher precision of diagnosis [e.g. detection of abnormalities present only in some cells of the trophectoderm biopsied (mosaicism); a feature that is not available using other CCS techniques]. Recent randomized trials have indicated the benefit of using comprehensive chromosome screening (CCS) methodologies and their potential of enhancing in vitro fertilization (IVF) success rates. Most recently a new methodology, known as next generation sequencing (NGS), was introduced into preimplantation genetic diagnosis (PGD) for CCS. NGS has been shown to be advantageous over other CCS methodologies, potentially offering higher precision of diagnosis [e.g. detection of abnormalities present only in some cells of the trophectoderm biopsied (mosaicism); a feature that is not available using other CCS techniques]. ObjectiveTo clinically apply a methodology that combines PGD for single gene disorders with CCS-NGS of blastocysts through utilization of a single trophectoderm biopsy. To clinically apply a methodology that combines PGD for single gene disorders with CCS-NGS of blastocysts through utilization of a single trophectoderm biopsy. Materials and methodsA total of 52 embryos derived from 11 different patients were assessed in this study. A single trophectoderm biopsy was carried out on day 5/6 of preimplantation development. Multiple displacement amplification was performed on the samples and aliquots from each amplified product were used to perform Karyomapping (Illumina, USA) for PGD of single gene disorders and NGS for CCS. The VeriSeq PGS assay and a MiSeq desktop sequencer (Illumina) were used for NGS. The BlueFuse Multi analysis software (Illumina) was utilized for interpretation of results. A total of 52 embryos derived from 11 different patients were assessed in this study. A single trophectoderm biopsy was carried out on day 5/6 of preimplantation development. Multiple displacement amplification was performed on the samples and aliquots from each amplified product were used to perform Karyomapping (Illumina, USA) for PGD of single gene disorders and NGS for CCS. The VeriSeq PGS assay and a MiSeq desktop sequencer (Illumina) were used for NGS. The BlueFuse Multi analysis software (Illumina) was utilized for interpretation of results. Results60% of the embryos were found to be either unaffected or carriers of the disorder being tested while, 40% of the embryos assessed were found to be euploid via NGS. 31.1% of the overall embryos assessed were seen to have at least one mosaic abnormality, while 8.9% of the embryos were determined to have only a single mosaic abnormality. Furthermore, segmental abnormalities were detected in 7 of the embryos tested and ranged from 7.8 to 104.6 megabases in size. A total of 16 embryos were found to be at low risk for the single gene disorder assessed and aneuploidy and could therefore be considered for transfer. 60% of the embryos were found to be either unaffected or carriers of the disorder being tested while, 40% of the embryos assessed were found to be euploid via NGS. 31.1% of the overall embryos assessed were seen to have at least one mosaic abnormality, while 8.9% of the embryos were determined to have only a single mosaic abnormality. Furthermore, segmental abnormalities were detected in 7 of the embryos tested and ranged from 7.8 to 104.6 megabases in size. A total of 16 embryos were found to be at low risk for the single gene disorder assessed and aneuploidy and could therefore be considered for transfer. ConclusionsThis study documents the successful clinical application of PGD for single gene disorders in combination with CCS-NGS using a single trophectoderm biopsy. The combination of powerful and highly accurate methodologies such as Karyomapping for single gene disorders and NGS for CCS, is expected to benefit patients undergoing IVF/PGD and can enhance their chances for a successful cycle leading to a healthy pregnancy. This study documents the successful clinical application of PGD for single gene disorders in combination with CCS-NGS using a single trophectoderm biopsy. The combination of powerful and highly accurate methodologies such as Karyomapping for single gene disorders and NGS for CCS, is expected to benefit patients undergoing IVF/PGD and can enhance their chances for a successful cycle leading to a healthy pregnancy.

Some embryos determined to be mosaic have been shown to be capable of normal implantation and development, albeit with reduced implantation and increased miscarriage rates. This study correlates the original biopsy findings with details of the localization of aneuploid cells in mosaic blastocysts. Comparison of PGS results to inner cell mass (ICM) and several trophectoderm samples of the same embryo. Embryos determined to be fully (90-100%) aneuploid or instead to be mosaic (10-90% abnormal cells) during routine PGS cycles were reanalyzed with Next Generation Sequencing (NGS). An ICM sample, usually consisting of 5-10 cells, was initially isolated. Subsequently, 2-4 trophectodermal samples ranging in size from 10 to ∼50 cells were taken from each embryo. The NGS result from each specimen was compared to the original biopsy. Mosaics were sub-classified as complex mosaics (3 or more chromosome abnormalities), aneuploid mosaics (1-2 chromosomes being mosaic) or partial aneuploid mosaic (normal / partial aneuploidy). The results are shown in the table. Five embryos were normal in all retested samples (5/43, 11.6%). All partial mosaic embryos with normal ICMs had normal or mostly normal retested trophectoderm samples (8/8), compared to 2/8 with full aneuploid mosaicism.Tabled 1Distribution of Normal Biopsy Samples in Mosaic EmbryosInitial NGS DiagnosisTotal EmbryosNormal ICMNormal Trophectoderm(Total of all samples)Full Aneuploidy190/19 (0%)4/70 (6%)Full Partial Aneuploidy30/3 (0%)2/10 (20%)Complex Mosaic122/12 (17%)a14/44 (32%)Partial Aneuploid Mosaic158/15 (53%)b30/59 (51%)dMosaic Full Aneuploidy168/16 (50%)c10/55 (18%)eAll Mosaics4310/43 (23%)54/158 (34%)a vs. b+c: p<0.05d vs. e: p<0.01 Open table in a new tab A diagnostic biopsy of ∼5 trophectoderm cells may detect mosaicism that does not exist in the ICM and may be highly localized in the trophectoderm. In this group of 43 mosaic embryos, there were 12 (28%) with normal ICMs that also tested normal for all or several trophectoderm specimens. The presence of aneuploid cell lines only in limited areas of the trophectoderm suggests that the initial chromosomal malsegregation event occurred relatively close to the time of biopsy. The implantation potential of a blastocyst with a normal ICM and a mildly or highly mosaic trophectoderm remains unknown. Embryos with an initial diagnosis of complex mosaic are less likely to have a normal ICM. Embryos diagnosed as partial aneuploid mosaic may be more likely to possess both a normal ICM and substantially normal trophectoderm, compared to full aneuploid mosaics. These findings may aid the determination of mosaic embryos that are suitable for transfer. Depending on patient age, a mosaic embryo may have the same or lower chance of being compromised by aneuploidy as an undiagnosed embryo, and should be given the appropriate replacement priority if no euploid embryos are available.

Objective: To determine whether undetected aneuploidy contributes to pregnancy loss after transfer of euploid embryos that have undergone array comparative genomic hybridization (aCGH).Design: Case-control study.Setting: University-based fertility center.Patient(s): Cases included 38 patients who underwent frozen euploid ET as determined by aCGH, resulting in miscarriage. Controls included 38 patients who underwent frozen euploid ET as determined by aCGH, resulting in a live birth.Intervention(s): Next-generation sequencing (NGS) protocols were internally validated. Saved amplified DNA samples from the blastocyst trophectoderm biopsies previously diagnosed as euploid by aCGH were reanalyzed using NGS. Cytogenetic reports of the products of conception for 20 of the pregnancies resulting in miscarriage were available for comparison.Main Outcome Measure(s): The incidence of aneuploidy and mosaicismusing NGS within embryos resulting in miscarriage and live birth.Result(s): Of euploid embryos analyzed by aCGH resulting in miscarriage, 31.6% were mosaic and 5.2% were polyploid by NGS. The rate of chromosomal abnormalities was significantly higher in embryos resulting in miscarriage (36.8%) than in those resulting in live births (15.8%). The rate of mosaicism was twice as high among embryos resulting in miscarriage than those resulting in live birth, but this was not statistically significant. Next-generation sequencing detected more cases of mosaicism than cytogenetic analysis of products of conception.Conclusion(s): Undetected aneuploidy may increase the risk of first trimester pregnancy loss. Next-generation sequencing may detect mosaicism and triploidy more frequently than aCGH, which could help to identify embryos at high risk of miscarriage. Mosaic embryos, however, should not be discarded as some can result in live births. 2016 by American Society for Reproductive Medicine.

Mitotic errors occur throughout preimplantation development, resulting in mosaicism. Errors that occur after the determination of the inner cell mass (ICM) will result in discordance between the trophectoderm and ICM. Whole blastocysts or isolated ICMs and trophectoderm samples were reanalyzed to determine the degree to which a trophectoderm biopsy reflects the overall status of the blastocyst. Intra-embryo comparison of NGS results of ICM and trophectoderm specimens. Embryos determined to be fully (>90%) aneuploid or mosaic (10-90% abnormal cells, including aneuploid/aneuploid mosaics) during routine NGS cycles were reanalyzed. Whole blastocysts were retested, as were 5-10 cell biopsies of ICMs and multiple 10-15 cell samples of trophectoderm of each embryo. Mosaicism was determined to have been initiated after ICM differentiation when an ICM sample displayed an abnormality not present in any trophectoderm sample, or when a trophectoderm sample displayed an abnormality not present in the ICM. Among whole mosaic blastocysts, 17/23 (74%) retested 'normal', indicating that fewer than 10% of the cells shared any single defect. Mosaicism was initiated after ICM differentiation in 10 of 61(16%) ICMs, and in 57/271 (21%) individual trophectoderm specimens (some embryos had two distinct aneuploid cell lines). Unique chromosome defects were determined to have occurred after ICM differentiation in either the ICM or trophectoderm in 39/61 embryos. Total concordance among all samples occurred in only 15 of 75 (20%) tested embryos. In the remaining 60 embryos, there were 70/303 (23%) samples that differed from the original finding. Mosaic embryos were more likely to yield trophectoderm biopsy results with additional mosaicisms (62/116, 53%) than were full and partial aneuploid embryos (14/75, p<0.01). Also, mosaicism occurred in fully aneuploid embryos at a reduced rate compared to meiotically euploid blastocysts: 7/21(33%) vs. 15/20 (75%), p<0.02. Reanalysis of whole mosaic embryos demonstrated that most blastocyst mosaicism is initiated after the 8 cell stage. These known mosaic embryos likely tested as normal because they contained aneuploid cell levels below the threshold of detection of NGS, an indication that many mosaic embryos contain a small percentage of aneuploid cells. The data show that mosaicism can occur in the ICM after it has differentiated from the trophectoderm, and that mitotic errors continue to accumulate in the trophectoderm. The impact of this dynamic of mitotic errors is that a ∼5-cell trophectoderm biopsy may not reflect the genetic state of the ICM, and cannot show the full extent of trophectodermal mosaicism. Some degree of trophectodermal mosaicism is compatible with normal development, but it is unknown whether embryos with mosaic ICMs can develop normally.

aCGH, qPCR and SNP arrays have been extensively used for preimplantation genetic screening (PGS) but a more robust and sensitive technique, Next Generation Sequencing (NGS) can detect mosaicism and polyploidy more effectively. The objective of this study was to determine if miscarriages occurring after PGS were due to abnormalities not detected by aCGH. Retrospective Analysis with prospective sample re-analysis by aCGH and NGS A total of 43 miscarriages were reported upon follow up of spontaneous pregnancy loss resulting from 2442 cycles with euploid blastocysts obtained from PGS by blastocyst biopsy and array CGH. Karyotype analysis of products of conception (POC) in 18 samples revealed aneuploidy (full aneuploidy, mosaic aneuploidy and partial aneuploidy) which was inconsistent with the aCGH analysis from the blastocyst biopsy and went undetected prior to embryo transfer. In the remaining 25 samples no POC data was available. Saved amplified DNA samples from these 43 blastocyst biopsies previously diagnosed as "euploid" by aCGH were re-analyzed by aCGH yielding the same result. A third aliquot of the same amplified DNA was then analyzed by NGS to determine if abnormalities not detectable by aCGH were present. The DNA from the embryos replaced that resulted in pregnancy and miscarriage were reanalyzed by NGS and the results are shown in the table below:Tabled 1NGS Reanalysis of TE Biopsy SpecimenNormalTriploidMosaic Whole AneuploidyMosaic Partial AneuploidyFull TrisomyNo DiagnosisPOC Not AnalyzedN=2511 (44%)2 (8%)3 (12%)7 (28%)02 (8%)POC AneuploidN=186 (33%)09 (50%)1 (6%)02 (11%)TotalN=4317 (40%)2 (5%)12 (28%)8 (19%)04 (9%) Open table in a new tab NGS is known to be able to detect mosaicism as well as triploidy, which is also well-known cause of spontaneous pregnancy loss. Whole and partial mosaicism could also play a role in spontaneous pregnancy loss depending on the chromosomes involved as well as the percentage of abnormal cells in the embryo. Of 16 embryos diagnosed euploid by aCGH that resulted in an aneuploid loss 10 (62%) were diagnosed as abnormal by NGS (2 not analyzable). Similarly, of the 23 embryos diagnosed euploid by aCGH that resulted in loss but not diagnosed by POC analysis, 12 (52%) were abnormal by NGS (2 not analyzable). The use of NGS would have avoided 56%(22/39) of the pregnancy losses resulting from aCGH tested embryos. Likely, mosaicism and triploidy accounted for most of these losses and NGS is more sensitive at detecting them. These data support the notion that NGS is the most powerful technique for PGS analysis in predicting euploid outcome but will not predict all euploid losses. It reduces the miscarriage rate by more than 50% over aCGH and will significantly improve outcomes.