Oral Abstracts of the ISPD 20th International Conference on Prenatal Diagnosis and Therapy, Berlin, Germany, 10–13 July 2016

OBJECTIVES: During the past several years, clinical whole-exome sequencing (WES) has been utilized for patients with complex clinical presentations. With a molecular diagnosis rate of about 30%, clinical WES is an effective way of ending the expensive and time-consuming diagnostic odyssey. With the continuous improvement ofWESmethodology includingupdated sequencing instrumentation, shorter turnaround time (TAT) and improved analysis, clinical exome is now considered for prenatal diagnosis. Here, we report the first 52 consecutive prenatal cases. Prenatal cases are defined as a fetal sample obtained through either an invasive procedure or a product of conception (POC). METHODS: This cohort includes 37 cases with WES performed for the proband only followed by Sanger sequencing studies of parental samples and 15 trio WES. For the final report, all contributing changes are confirmed by Sanger sequencing for trio cases. The exome sequencingwas performedon the IlluminaHiSeq2500with an average of ~11.4Gb data per exome, and 97% of the targeted exome regions are sequenced at a depth of 20X. Exome sequencing data were analysed for small nucleotide changes and large CNVs. Individual DNA samples were analysed by the Illumina HumanExome-12v1 array for quality control and large CNVs detection. RESULTS: The proband WES yielded a diagnosis of 32% (12/37) cases, whereas trio WES provided diagnosis in 40% (6/15) cases. For the five patients with a brain anomaly on ultrasound, we were provided diagnosis for three individuals (60%), for the 21 patients with a brain anomaly and other organ system involved, we provided diagnosis for seven patients (33%) and for the 26 patients with an anomaly not including the brain, we provided a molecular diagnosis for eight patients (31%). For individualswith cardiac anomaly andother organ systemanomaly (six patients), we were able to provide diagnosis for three individual (50%). CONCLUSIONS: With rapid turnaround time and comprehensive analysis, the prenatal trio WES is a valuable tool in prenatal diagnosis. The next step in improving the current test will include analysis of DNA from direct chorionic villus sampling or amniotic fluid to further reduce the TAT.

OBJECTIVES: Through the NIPSIGEN project (Non-Invasive Prenatal diagnosis for Single Gene disorders; funded by the Health Innovation Challenge Fund; HICF-R6-381), we have recently developed a clinical test for the non-invasive prenatal diagnosis (NIPD) of Duchenne and Becker muscular dystrophies (DMD/ BMD) in at-risk pregnancies. Building on this experience, we are now validating an improved NIPD test for multiple single gene disorders, namely, DMD/BMD, spinal muscular atrophy (SMA), congenital adrenal hyperplasia (CAH) and cystic fibrosis (CF), which we intend to implement into clinical practice within the UK National Health Service (NHS). METHODS: Cell-free DNA (cfDNA) was extracted from maternal plasma; maternal, paternal and proband DNA was extracted from leukocytes. DNA sample libraries were prepared for massively parallel sequencing on an Illumina MiSeq using a custom built capture probe library. Over 12 000 SNPs with >40% probability of being heterozygous were targeted across multiple genomic regions containing the genes of interest on chromosomes X, 5, 6 and 7. Sequencing data were analysed by relative haplotype dosage (RHDO) on informative SNPs to determine fetal inheritance of the maternal and paternal mutant alleles using the proband haplotypes as reference. RESULTS: The test achieved 100% sensitivity and specificity on patients tested so far, which included four pregnancies at risk of DMD/BMD and five at risk of SMA. All results concurred with expected outcomes apart from one DMD patient for whom an inconclusive result was observed. Test validation was also performed on 11 healthy control pregnancies, where the fetal DNA obtained from the CVS was used to identify the reference haplotypes. As a result, test simulation of nine DMD/BMD, four SMA, two CAH and two CF cases was successfully achieved. CONCLUSIONS: To date, the implementation of NIPD testing for single gene disorders into clinical practice has been hindered by the elevated costs incurred. In this respect, our method allows for a multiplexing capacity of two to three patients per MiSeq sequencing run, while maintaining a high testing accuracy and allowing for multiple disorders to be tested simultaneously. Together, these characteristics render the test feasible from a clinical perspective, potentially opening the path for implementation of NIPD for single gene disorder testing into clinical practice in the UK and elsewhere.
OBJECTIVES: Since 2012, haplotype-assisted noninvasive prenatal diagnosis (NIPD) has been developed in our laboratory, and validated in a number of monogenic diseases using clinical samples from trio families. The purpose of this study was to summarize the performance of this method in different monogenic diseases. METHOD: Customized panels were designed to capture the coding region and flanking area of disease-associated genes. Genomic DNA from each trio family (father, pregnant mother and proband child) and plasma cell-free DNA from each pregnant mother were targetedsequenced using the customized panels. Fetal fraction was calculated with the differently homozygous SNPs in both parents. Parental and fetal haplotype were deduced based on genetic information from the trio family and hidden Markov model. Amniocentesis was performed to confirm NIPD results. RESULTS: A total of 97 trio families covering 12 monogenic diseases have been tested before receiving prenatal diagnosis (Table). The mutations consisted of point mutations and copy number variants. Mean gestational age of pregnant women while receiving NIPD was 18 weeks. Fetal fraction was between 2.8% and 22.6%, with the mean of 9.1%. Sixty pregnancies were identified as disease-affected. So far, amniotic fluid was tested for prenatal diagnosis in 70 pregnancies, all consistent to NIPD results. CONCLUSIONS: Haplotype-assisted NIPD showed high accuracy in predicting the existence of fetal monogenic disease and can be used in different diseases and mutation types. 1-5  Table. OBJECTIVES: The Management of Myelomeningocele Study (MOMS) provided level-I evidence that fetal spina bifida aperta (SBA) repair by open approach, as opposed to postnatal repair, reduces the need for ventriculo-peritoneal shunting and improves motor outcomes at 30 months. Since less invasive techniques are contemplated, we aimed to add sensitive and more comprehensive assessment tools for evaluating outcomes in the fetal lamb model. METHODS: Spina bifida aperta was surgically induced without or with myelotomy at 75 days of gestation (term = 145 days). Myelotomy-SBA was repaired at 100 days using a standard open approach. Lambs were delivered by cesarean section just prior to term (median: 142 days). At 1 to 2 days postnatally, they underwent gross examination, neurologic assessment, whole-body magnetic resonance neuroimaging (MRI), somatosensory and motor evoked potentials (SEP and MEP) and histology of the central nervous system. RESULTS: Spina bifida aperta was successfully induced in 35 fetuses (nine without and 26 with myelotomy), without intra-operative losses. In seven fetuses, it was repaired. Currently 13 ewes have been delivered of 31 lambs: eight were dead prenatally (4/13 SBA, 4/18 controls), two SBA had lumbar evisceration. The seven SBA survivors were compared to the 14 controls: 4/7 had lumbar leakage, 1/7 had spontaneous complete skin closure; 7/7 had hindlimb somatosensory deficits; MRI confirmed 4/7 lumbar cerebral-spinal fluid leakage and 2/7 hindbrain herniation; 4/4 and 2/2 had respectively absent hindlimb SEP and MEP; 5/5 had disrupted spinal cord covered with fibrotic tissue. CONCLUSIONS: Comprehensive structural and functional assessment of the SBA fetal lamb model is feasible and reproducible, which can be used to measure effects of in utero induction and repair.
OBJECTIVES: Preterm premature rupture of the fetal membrane (PPROM) remains a major complication after fetoscopic interventions because the defect in the amniotic membrane (AM) does not heal spontaneously. We examined whether surgically induced membrane defects elevate connexin 43 (Cx43) expression in the AM wound edge and drives structural changes in collagen architecture that affects healing after fetoscopic surgery. METHODS: Fetal membranes were collected at cesarean delivery from women who underwent fetal surgery (fetoscopic surgery n = 12; hysterotomy n = 2) between 18 and 29 weeks before birth. Collagen microstructure in the amniochorion was investigated by scanning electron microscopy (SEM). Collagen alignment in the wound edge of the AM was examined by Second Harmonic generation (SHG). Immunofluoresence was used to examine cell morphology and compare Cx43 expression in the epithelial and fibroblast layer of the AM. Cx43 gene expression was quantified by RT-qPCR in wound edge and intact control AM collected from the same patient. RESULTS: Scanning electron microscopy showed densely packed collagen in the wound edge of the amniochorion. This collagen arrangement changes in the fibroblast layer with evidence of highly aligned collagen fibers along the wound edge AM but not in control membranes. Cx43 expression was increased by 112.9% in wound edge AM compared to control membranes (p < 0.001), with preferential distribution in the fibroblast layer compared to the epithelial layer (p < 0.01). Mesenchymal cells had a flattened morphology, and there was evidence of poor epithelial cell migration across the fetal defect. Cx43 gene expression was significantly increased in wound edge AM compared to patient matched controls (p < 0.001). CONCLUSIONS: We propose that overexpression of Cx43 in the fetal AM after fetal surgery induces morphological and structural changes in the collagenous matrix that interferes with normal remodeling mechanisms. These processes may help to explain the increased tissue weakening and incidence of preterm birth post fetoscopic surgery. OBJECTIVES: Evaluation of the impact of visualization of the retronasal triangle view (RNTV) on the diagnosis of cranio-facial anomalies in dystrophic viable and non-viable fetuses in the first and early second trimester via 3D multiplanar reconstuction using complete volume data sets. METHODS: A total of 121 complete three-dimensional volume data sets of fetuses with genetic and/or structural anomalies were analysed. After standardized acquisition of complete fetal head volume with the fetus facing the transducer the volume was adjusted to visualize the correct cranio-facial anatomy including the integrity of bony structure of the RNTV, symmetry, presence of facial clefts and midfacial hypoplasia. RESULTS: In 112/121 cases, the multiplanar reconstruction of all three orthogonal planes including a true midsagittal section was feasible. Correct configuration of the RNTV was verified via navigating in the corresponding coronal plane and could be achieved in 29/48 non-viable fetuses and 56/64 viable fetuses. Using the same plane, an absent mandibular gap indicating a micrognathia was detected in 28 fetuses (seven non-viable and 21 viable fetuses). In 16 fetuses (4 vs 12), facial clefts were diagnosed. An absent nasal bone was identified in 27 volumes (4 vs 23). Chromosomal anomalies or single gene disorders could be confirmed in 55/112 of cases. CONCLUSIONS: Anomalies of the maxilla-mandible complex may be frequently found in gentically altered and syndromic fetuses. 3D multiplanar evaluation of affected fetuses offers the opportunity of a reliable and reproducible evaluation of the cranio-facial integrity using the RNTV via standardized planes at an early gestational age. It can add valuable information to the prenatal assessment of dystrophic fetsues even in the non-viable ones where image clarity is often reduced. OBJECTIVES: Antral follicle count (AFC) is a well-known marker of ovarian reserve, and our group has previously explored the role of AFC as a marker of fetal aneuploidy during the first trimester of pregnancy. The aim of our study was to compare the proportion of large and small antral follicles observed in both the chromosomally normal and in trisomic pregnancies. METHODS: The study group included a cohort of 786 women carrying a singleton pregnancy scanned before 14 weeks in our center. Autosomal trisomies were identified by cytogenetic studies performed either chromosomal in chorionic villi or postnatally (n = 28). Chromosomally normal pregnancies were ascertained by a normal karyotype or normal neonatal outcome (n = 758). Use of any kind of assisted reproductive techniques was an exclusion criterion; hence, all the studied pregnancies were spontaneously conceived. Proportions of small (2-5 mm) versus large (6-9 mm) antral follicles were compared in trisomic and chromosomally normal pregnancies. RESULTS: In the chromosomally normal pregnancies, mean proportions for large and small antral follicles were 8% (IC 95% 7-9) and 92% (IC 95% 91-93), respectively. The corresponding mean proportions in the in trisomic group (n = 28) were 16% (IC95% 7-25) and 84% (IC95% 75-93%), significantly. The mean difference was 0.069 (IC 95% 0.04-0.134; p-value = 0.0389). After adjusting AFC for maternal age as a confounding factor in a logistic regression model, the estimate OR for the trisomic group was OR 5.10 (IC95% 1.64-15.83; p-value 0.005). CONCLUSIONS: Fetal aneuploidy appears to be associated with a higher proportion of large antral follicles. Given that a decreased antral follicle count is observed in women carrying a trisomic pregnancy, the large follicles count may remain stable, and this decline be particularly confined to small follicles. Pre/postnatal diagnosis predicted PIQ (p = 0.023), reaction time (p = 0.022), stability of reaction time (p = 0.01), and face memory (p = 0.01). SES was a significant predictor of VIQ (p = 0.002) and PIQ (p = 0.006) in the combined cohort and the US cohort, though not in the NL cohort. Testosterone was a (borderline) predictor for PIQ (p = 0.051). In the NL group testosterone was predictive of decreased total (p = 0.008) and internalizing problems (p = 0.044). CONCLUSIONS: This is the first international collaborative investigation of the neurocognitive capabilities of boys with XXY. This study further supports the need for early detection and hormonal treatment. The fact that nationality was a strong predictor of VIQ and PIQ implicates other unidentified variables, such as timing and frequency of services, which needs further investigation. SES and testosterone were associated with stronger cognitive profiles. Prenatal diagnosis was also shown to be highly predictive of improved neurodevelopmental outcome. There was a significant positive association between improved neurodevelopmental outcome and EHT across multiple domains. Congenital Muscular Torticollis occurred in 15% of the infants with 40% to the less common left side preference regardless EHT. CONCLUSIONS: This is the first well-described study on a large cohort of prenatally diagnosed boys with XXY, their response to EHT and the impact on neurodevelopment. These results are further documentation that EHT may have a positive outcome on developmental progression as well as characterizing the demographic, educational, and perinatal aspects of 158 infants with XXY and prenatally diagnosed. With proactive care, the academic and behavioral differences associated with XXY can be minimized and possibly ameliorated.

5-3
Randomized controlled trial comparing preimplantation genetic screening utilizing next generation sequencing to standard morphologic assessment for embryo selection and single embryo transfer after vitrification

Kathryn Kurtzman, Theraysa Gapasin
Illumina, Inc., San Francisco, CA, United States OBJECTIVES: The objective of this trial was to evaluate the efficacy of PGS in improving pregnancy outcomes by comparing ongoing pregnancy rates at 20 weeks of gestation in women undergoing in vitro fertilization (IVF) and single embryo transfer (SET) after vitrification in two groups. In the intervention group, a single embryo is selected for transfer using preimplantation genetic screening (PGS) with next generation sequencing (NGS), and in the control group, a single embryo is selected by standard morphologic assessment. METHOD: This is a single blinded, multicenter, randomized, controlled trial in women undergoing elective in vitro fertilization (IVF) cycles. Trophectoderm (TE) biopsy on day 5 or 6 of embryo development followed by PGS with NGS for embryo selection in the intervention group is compared to embryo selection based on standard morphology in the control group. Randomization is one to one stratified by three maternal age classes. All embryos groups undergo vitrification followed by SET in a subsequent unstimulated cycle. Primary outcome is ongoing pregnancy rates at 20 weeks of gestation between the two groups. RESULTS: Our recruitment strategy, including data based site selection, aggressive study start up with focused and ongoing collaborative face to face site visits, and appropriate resource allocation, has resulted in enrollment rates exceeding our timelines with full enrollment accomplished in 18 months. Enrollment commenced September 2014 and an adequate number of subjects have been enrolled to ensure 600 embryo transfers with at least 300 transfers in each arm. CONCLUSIONS: The regionally diverse settings and broad eligibility criteria included in this trial are applicable to the general IVF population. Large-scale trials such as this can support the use of PGS with SET to improve IVF outcomes, thus reducing the need for multiembryo transfer and its higher incidence of multiple gestations. While this study was designed to address scalability and performance of PGS in a variety of patient populations and settings utilizing individualized practices, all sites met strict qualification criteria. The results of this study cannot be extrapolated to clinics that do not perform at this level with or without PGS.

5-4
Clinical experience with multigene carrier panels OBJECTIVES: The goal of carrier screening is to identify individuals at-risk for having a child with genetic disease. Professional guidelines recommend screening for select genetic conditions based on ethnicity. Expanded carrier screening includes diseases beyond what is currently recommended. A recent joint statement from professional societies acknowledges the potential benefits of this testing. However, limited data have investigated the positivity rates of expanded carrier screens. These data are critical to providers when deciding what testing to offer, and assessing implications for their patients and practices. We describe our experience with three screening panels varying in size from 3 to >200 disorders. METHODS; We reviewed outcomes for three multigene carrier screening panels: Trio (three diseases), Standard (23 diseases), and Global (>200 diseases). All panels utilized targeted genotype analysis of pre-selected mutations via next-generation sequencing (NGS). Standard and Global panels also included hemoglobinopathy evaluation (MCV and electrophoresis) and hexosaminidase-A enzyme analysis for Tay-Sachs disease screening. We calculated the frequency of positive results for each panel. RESULTS: Positivity rates were 5.37% for the Trio panel, 12.10% for the Standard panel, and 34.14% for the Global panel. The most frequent positive results in the Global panel were for (in descending order) hemoglobinopathies, glucose-6-phosphate dehydrogenase deficiency, cystic fibrosis, familial Mediterranean fever, pseudocholinesterase deficiency, spinal muscular atrophy, and primary congenital glaucoma. CONCLUSIONS: Two hundred diseases, the likelihood of identifying a carrier can be as high as 34%. Understanding panel positivity rates is important for providers to choose the right test for their practice, set appropriate expectations for patients, and plan for follow-up counseling and partner testing.

5-5
Aneuploidy screening in polarbodies is superior to trophectoderm analysis as it detects the meiotic errors and circumvents decision making of mosaic embryo results Angelika Daser 1 , Cornelia Vogt 1 , Elisabeth Müller 1 , Melanie Schranz 1 , Dagmar Seifert 1 , Thomas Hahn 2 , Martin Schorsch 2 1 SH-Gen Research, Wiesbaden, Germany 2 Fertility Center Wiesbaden, Wiesbaden, Germany OBJECTIVES: Lately mosaic embryos have been transferred and resulted in life births. Mosaicism in early embryos develops through postzygotic mitotic errors to different degrees and varying persistence. Therefore, decisions which embryos to transferwhich degree of mosaicism in which affected chromosomesare notoriously difficult, and there are no criteria which would assist. Alternatively, there are polar bodies as products of the two meiotic divisions providing clear information of the chromosome content of the maternal part (~90% of all meiotic errors) of the zygote. Therefore, aneuploidy screening in eggs gives relevant and robust information about which embryos to transfer. METHODS: As both polar bodies have to be analysed, a fast and cost-efficient method is needed. We have developed a simple PCR method (Molecular Copy number Counting -MCC) to count chromatids directly with DNA at limiting dilution and digital readout. After dispensation of the polar body DNA into eight PCR reaction units, two rounds of specific PCR amplification are requireda first round multiplex PCR for all markers on all chromosomes and a second specific marker monoplex PCR. The read out is digitalthe number of PCR products equals the number of chromatids. RESULTS: We analysed the oocytes of 375 cycles; the largest demand for MCC was in the age group of above 38 years with 234 cycles (68%). In the younger groups, almost all cycles ended in an embryo transfer, but 35% (81 cycles) of the highest age group had only aneuploid embryos, and hence, no transfer was made. As expected, pregnancy rates decreased with age across the groups. However, the age effect is almost extinguished if pregnancy rates are evaluated after embryo transfer. CONCLUSIONS: Polar body analysis with MCC has proven a direct and robust method with beneficial biological outcome, especially in couples with poor prognosis. The method allows a strategy that is fast and reduces costs to the minimum necessary to identify euploid oocytes, exclude aneuploid oocytes, and avoid futile cycles with aneuploid embryos.

5-6
The timing of reproductive genetic counseling is the most important factor influencing acceptance of expanded carrier screening OBJECTIVES: Expanded carrier screening (ECS) for more than 100 genetic diseases is now available to women who are pregnant or considering pregnancy and their reproductive partners. The factors that influence uptake of ECS have not been extensively studied. Our objective was to retrospectively examine which factors influence a woman's decision to undergo ECS. METHODS: We reviewed 500 consecutive medical records from women with documented prenatal or preconception genetic counseling for various indications at a large prenatal genetic counseling service between April 2012 and June 2013. No patients were excluded, and they were offered different preconception or prenatal screening and/or diagnostic tests. We tabulated acceptance rates for ECS by women and their reproductive partners along with various factors, including demographic information, personal and family history of genetic disease, timing, and indication for counseling. Data were analysed using descriptive statistics, t-tests, and chi-square tests. RESULTS; Expanded carrier screening was offered to 483 of 500 women and 192 (39.8%) total accepted. Of the subset (n = 67) seen preconceptionally 46 (68.7%) accepted ECS. This was significantly more than those seen during pregnancy (n = 416), of whom 146 (35.1%) accepted screening (p ≤ 0.001). For pregnant patients, the mean gestational age of those accepting carrier screening (12 weeks 3 days; n = 146) was significantly lower than for those declining (13 weeks 4 days; n = 270; p ≤ 0.001). Maternal and paternal ethnicity did not significantly affect ECS acceptance, although some trends were identified. CONCLUSIONS: These results suggest that women who receive genetic counseling prior to or earlier in the first trimester of pregnancy are more likely to accept ECS. This supports the importance of presenting reproductive age patients and their partners early with options for genetic carrier screening. Study of additional medical records is ongoing, and overall results may help delineate how other factors such as maternal ethnicity and family history influence acceptance of ECS. Other areas of further research may include the effects of socio-economic status and insurance coverage on ECS acceptance rates. These data can guide individualized counseling about carrier screening options. 5-6 Image.

6-1
Confined placental mosaicism: Re-evaluation of pregnancy characteristics and influence on fetal growth Jérôme Toutain 1 , Laurence Taine 1 , Damien Goutte-Gattat 1 , Jacques Horovitz 1 , Robert Saura 2 1 CHU de Bordeaux, Bordeaux, France 2 Institut Européen de Chimie et Biologie, Pessac, France OBJECTIVES: An association between increased nuchal translucency (NT) and confined placental mosaicism (CPM) of trisomy 16 has recently been reported. Very low levels of serum pregnancy-associated plasma protein A (PAPP-A) have also been demonstrated for CPM of trisomy 16. Furthermore, the influence of CPM subtypes on fetal growth, whatever the trisomy confined to placenta, still remains controversial. We wanted to re-evaluate the link between NT, PAPP-A levels and CPM, as well as the influence of CPM subtypes (types 2 and 3) on fetal growth and adverse pregnancy outcomes. METHODS: From July 2009 to December 2015, 5512 chorionic villus samplings were performed in our Center. Conventional karyotype after long-term cultured villi (LTC-villi) was systematically established. In case of suspicion of CPM, karyotype after short-term cultured villi was performed to define type 2 CPM (chromosomal abnormality limited to the mesenchymal core) or type 3 CPM (chromosomal abnormality found both in the cytotrophoblast and the mesenchymal core). Amniocentesis was performed to exclude fetal mosaicism, and uniparental disomy testing was carried out, if appropriate. Results were compared to a control population of 93 patients in whom karyotype was strictly normal after LTC-villi. RESULTS: Thirty-six CPM were observed (0.65%, 13 type 2, 23 type 3). The mean NT was not increased for types 2 and 3 CPM. For type 3 CPM, the median serum PAPP-A was statistically lower than for the control population. Incidence of intrauterine growth restriction (IUGR), neonatal hypotrophy, and stillbirth was comparable between type 2 CPM and the control population. In type 3 CPM, IUGR was noticed in 77.3%, neonatal hypotrophy in 72.2%, and intrauterine fetal death or stillbirth was deplored in 17.4% of cases (Table 1). The percentage of abnormal cells after LTC was negatively associated with birth weight. CONCLUSIONS: Increased NT did not appear to be associated with type 2 or type 3 CPM. Interestingly, median serum PAPP-A level was decreased in type 3 CPM, both for trisomy 16 and other trisomies. Regarding fetal growth, our study confirmed that when a CPM is suspected, CPM subtypes need to be carefully established. Although type 2 CPM has no effect on fetal development, type 3 CPM is associated with IUGR, intrauterine fetal death, neonatal hypotrophy, and stillbirth. When a type 3 CPM is diagnosed, we recommend therefore a close ultrasonographic monitoring in order to manage the fetal growth restriction. 6-1 Table. OBJECTIVES: Children intrauterine exposed to chemotherapy are more prone to be born with a birth weight below the 10th percentile, defined as 'Small for Gestational Age (SGA)'. Pathways for angiogenesis, proliferation, and inflammation are known to be related to SGA. The mechanism underlying SGA following in utero exposure to cancer treatment is so far unexplored, and models to explore the effect of chemotherapy on the placental growth and function are not available. This study investigates the development of a suitable murine xenograft model to examine the possible effect of chemotherapeutic agents or other drugs on the placental tissue. METHODS: The test population consisted of five examination groups of six immunodeficient nude mice. After ovariectomy and hormonal stimulation with 17b-estradiol and/or progesterone pellets, placental tissue derived from first or third trimester pregnancies was engrafted subcutaneously bilateral in the flanks. General morphology was evaluated by H&E and immunohistochemistry (hCG, CD31). hCG secretion was measured in serum and urine by ELISA. The expression level of selected genes (IGF-1, PGF, IGF-2, eNOS, Flk-1, and Flt-1) was compared by using RT-qPCR. WTSS and IPA (Ingenuity Pathway Analysis) are ongoing to investigate the stability of selected pathways relevant for the angiogenesis, inflammation, and proliferation. RESULTS: Comparing the placental tissue before and after engraftment, we found preserved structure and histological features. hCG secretion, evaluated by ELISA in the urine and blood, was present for 3 weeks. (Fig. 1) RT-qPCR analysis showed preserved genetic characteristics of selected genes relevant for angiogenesis, proliferation, and inflammation of the primary engrafted placental tissue. Results on WTSS and IPA are expected to be ready for presentation in July 2016. CONCLUSIONS: We created a murine xenograft model with proven stability and preserved structure of the engrafted placental tissue for 3 weeks. This established murine xenograft model can be used to examine the effect of cancer treatment on the placental growth and function, in order to identify risk factors related to SGA and to develop preventive measures. Moreover, it could serve as a model for other fetal toxicity studies. 6-2 Image.

6-4
Increased risk for pregnancy loss, true fetal mosaicism and uniparental disomy associated with rare autosomal aneuploidies identified during non-invasive prenatal testing Mark D Pertile, Nicola Flowers 2 , David Francis 2 , Sara Cronin 2 , Grace Shi 2 , Ralph Oertel 2 , Fiona Norris 1 , Damien L Bruno 1 1 Victorian Clinical Genetic Services, Parkville, Victoria, Australia 2 Victorian Clinical Genetics Services, Melbourne, Victoria, Australia OBJECTIVES: Non-invasive prenatal testing (NIPT) using a whole genome sequencing (WGS) approach enables identification of chromosome abnormalities beyond that of standard trisomy detection. Our preliminary experience testing almost 10 000 women with high and average risk pregnancies at 10 to 13 weeks of gestational age indicates high test sensitivity (100%) and specificity (99.99%) for trisomy 21, with very low test failure (0.2%). Here, we report our experience using a WGS method to investigate a test algorithm QC failure that accounts for the majority of these test failures, and which is associated with a significantly increased risk for adverse pregnancy outcome. METHODS: We prospectively tested 9746 pregnancies using massively parallel sequencing of maternal plasma DNA. Normalized chromosome values (NCV) were calculated to identify pregnancies at high risk for standard aneuploidies. Additionally, we plotted normalized chromosome coverage values for all 22 autosomes and the X chromosome for every patient. Test failure occurred when normalized chromosome denominator (NCD) likelihood scores fell outside a pre-determined range. NCD provides an overall score for how chromosomal coverage profiles deviate from the expected. We used the normalized coverage plots to identify the chromosome(s) causing the algorithm failure and then re-analysed the sequence data using WISECONDOR for whole genome analysis. RESULTS: Fifteen cases (1/650; 0.15%) were associated with NCD scores leading to test failure. These included trisomy 2 (three cases), T7 (2), T10 (1), T15 (4), T16 (2), and T20 (1). One case deletion 10q and one of multiple aneuploidy associated with suspected maternal malignancy were also recorded. True fetal mosaicism at amniocentesis was observed for T2, T7 and T16, miscarriage for T15 (three cases) and T20, fetal demise at 17/40 for T2, confirmed or suspected uniparental disomy (UPD) for T2 (true mosaic), T15 (UPD15mat) and T16 (MCDA twins). Cases of T2, T7 and T10 had normal SNP microarray at amniocentesis. CONCLUSIONS: This study demonstrates that QC failures arising from abnormal sequence coverage affecting NCD likelihood scores are typically associated with fetal or placental trisomy involving non-target chromosomes. These trisomies are often meiotic in origin and may be associated with miscarriage early in pregnancy. However, trisomy rescue can lead to true fetal mosaicism or normal fetal karyotype, sometimes seen in association with UPD or fetal demise. The findings from this study are particularly relevant for laboratories using Illumina's proprietary analysis software and have the potential to improve patient care by expanding the clinical utility of NIPT for this specific type of test failure.

6-5
Incremetal yield of genomic microarray in early growth restricted fetuses over karyotyping

Institut D'Investigacions Biomèdiques August Pi ì Sunyer, Barcelona, Spain
OBJECTIVES: Fetal genetic disease is a well recognized cause of early and severe intra-uterine growth restriction. However, scarce data are available about the association between fetal growth restriction with microdeletion and microduplication syndromes. The aim of this multicenter study was to assess the incremental yield of genomic microarray over karyotyping in early and severe growth restriction. METHOD: This multicenter study included 137 consecutive early (<25 weeks) and severe (<3rd percentile on estimated fetal weight) growth restricted fetuses with a normal karyotype or quantaitive fluorescent polimerase chain reaction (QF-PCR), studied during a 3-year period (January 2013-December 2015) in three centers of Barcelona. BAC (Bacterial Artificial Chromosome) array-CGH (CytoChip Focus Constitutional, BlueGnome, Illumina) was performed in DNA extracted from amniotic fluid. The incremental yield was defined by the rate of fetuses presenting with a pathogenic copy number variants below 10 Mb in normal karyotype/QF-PCR results. RESULTS: In our series, the incremental yield of genomic microarray over karyotyping was 7.3% (95%CI: 2.9 to 11.7) (10/137). Among malformed fetuses, this rate was 15.0% (3/20); in non-malformed fetuses with other findings (soft markers, abnormal amniotic fluid…) was 2.6% (1/38); and in fetuses with IUGR as isolated finding 7.6% (6/79). CONCLUSIONS: Our findings support the use of genomic microarray after a normal QF-PCR/karyotype given that provides a 7.6% incremental yield in growth restricted fetuses as isolated finding, increasing up to 15.0% in malformed fetuses. OBJECTIVES: A physiological redox state is essential for placental and fetal development. Hypoxia and imbalanced oxidative stress are key factors in the pathogenesis of intra-uterine growth restriction (IUGR). However, their roles in selective intra-uterine growth restriction (sIUGR) in monochorionic twins (MCT) are still unknown. This study explored the characteristics of hypoxia/oxidative stress in the placenta shares of MCT and their possible connections with sIUGR. METHODS: The expression levels of hypoxia inducible factor-1α (HIF-1α, hypoxia marker), malondialdehyde (MDA, lipid oxidation marker), and 8-hydroxydeoxyguanosine (8-OHdG, oxidative DNA damage marker) were evaluated in the placental shares of normal MCT (Group A) and sIUGR MCT (Group B). Expression was evaluated in the larger twins (A1/B1) and smaller twins (A2/B2). The relationship between birth weight and each marker was analysed. RESULTS: HIF-1α expression was significantly higher in the placenta shares of sIUGR MCT, indicating more severe hypoxia. HIF-1α expression was also significantly higher in the placenta share of the growth-restricted fetus (B2) than in the co-twin (B1) (P < 0.05). There were also larger inter-twin differences in HIF-1α expression in sIUGR MCT than normal MCT. The oxidative stress markers MDA and 8-OHdG were significantly higher in the placenta share of growth restricted fetus (B2) than the co-twin (B1) in sIUGR MCT (P < 0.05). The inter-twin differences in MDA and 8-OHdG were also significantly larger in sIUGR MCT than normal MCT (P < 0.05). CONCLUSIONS: The distinctive pattern of hypoxia and oxidative stress observed in sIUGR MCT pregnancies suggests that hypoxia related oxidative stress plays an important role in the pathogenesis of sIUGR.

7-1
Clinical implementation of NIPT: Understanding the role of non-specialist provider education

Subhashini Chandrasekharan, Subarna Pradhan, Anthony Hung, Mary Anne McDonald
Duke Global Health Institute, Duke University, Durham, NC, United States OBJECTIVES: As cell-free DNA (cfDNA) prenatal screening expands to low risk pregnancies and for microdeletions non-specialist providers (general Ob/Gyns, certified nurse midwives (CNM) may increasingly be required to counsel women/families before and after testing. Yet these providers have few opportunities to achieve and maintain competence in medical genetics or genetic counseling leaving. The objectives of this research are to (1) determine stakeholders' perspectives on gaps in physician education about cfDNA screening, especially for non-specialists and (2) to systematically assess suitability of currently available online courses for educating non-specialist providers about technical, clinical, and ethical aspects of cfDNA screening. METHODS: Thematic content analysis was performed on 75 semi-structured interviews conducted with representatives of eight stakeholder groups to identify barriers to appropriate clinical implementation of cfDNA in the US. Online continuing medical education (CME) courses were identified by keyword searches. We conducted content analysis of ten online CME courses on prenatal genomic screening along the following broad themes: (1) types and uses of prenatal genetic screening and testing in general; (2) information about cfDNA screening, specifically types of tests available, their uses and performance; and (3) ethical legal and social issues (ELS) surrounding cfDNA screening. RESULTS: Stakeholders concerns surrounded the (1) expanding responsibilities for general Ob/Gyns versus limited time and capacity they have for providing pre-test and post-test genetic counseling for cfDNA screening; (2) gaps in Ob/Gyn's knowledge about cfDNA screening and genetic conditions tested; and (3) potential lack of accurate and unbiased educational resources due to companies playing a strong role in educating non-specialists. CME courses were high variable in their content and topics covered. Topics with low coverage included discussion of lower specificity of cfDNA screening for conditions other than Trisomy 21, PPV of microdeletion testing, incidental findings, reasons for false positives and their implications for informed-decision making. CONCLUSIONS: Participants from all stakeholders groups perceive educating non-specialist providers as one of the most important barriers for appropriate clinical implementation of cfDNA screening. Further empirical data are needed to understand nonspecialist provider's educational experiences and specific needs surrounding cfDNAscreening education. Variable and poor coverage of technical, clinical and ELS issues in currently available online CME courses raises need for more systematic development and assessment of educational resources targeted to such nonspecialists. These approaches can better prepare non-specialist Ob/Gyn providers to facilitate patient informed decision-making and foster more ethical and effective use of cfDNA screening.

7-2
Ensuring high quality NIPD and NIPT for the patient

EMQN, Manchester, United Kingdom
OBJECTIVES: Many laboratories plan to provide non-invasive prenatal testing (NIPT) by analysing cell free DNA (cfDNA) in maternal plasma. When new testing approaches are introduced, all stakeholders must be assured that high quality testing is performed. This can be achieved through participation in external quality assessment (EQA) schemes; however, EQA for cfDNA is challenging. Initial schemes using spiked DNA plasma samples for fetal sex determination were sub-optimal. Here, we describe the development of EQA for aneuploidy NIPT and a scoping exercise for the need for EQA for non-invasive prenatal diagnosis (NIPD) for fetal sexing and single gene disorders. METHODS: Collaboration between EQA providers (CEQAS, EMQN, and UK NEQAS for Molecular Genetics) has been established to provide a pilot EQA to assess the standard of testing for NIPT for fetal aneuploidies and diagnosis of single gene disorders/fetal sexing in maternal plasma. In order to determine current and future laboratory practice for testing an online survey was distributed to over 1200 laboratories in January 2016. An exploratory pilot EQA is planned for six laboratories each receiving three maternal plasma samples for aneuploidy testing and the results expected to be reported according to the clinical case scenarios supplied with the samples. RESULTS: Surveys received from 122 laboratories indicated 51% laboratories provide NIPT for aneuploidy, with a monthly testing volume of 8 to 15 000 samples (mean 565) reporting within 3 to 15 days. All other laboratories are implementing testing. Seven laboratories perform/plan to introduce NIPD for single gene disorders (6%), and laboratories perform fetal sexing either diagnostically or as part of the analysis. Next Generation Sequencing is the most predominant method used (88%). The range of methods, fetal fraction testing, reporting other abnormalities, and sample requirements will be presented. The exploratory pilot aneuploidy NIPT EQA run results and plans for a global EQA will be discussed. CONCLUSIONS: The survey highlights the introduction of NIPT for aneuploidies into clinical service, the variable approaches employed and the need for external assessment and guidelines to promote and ensure high quality testing for the benefit of patients. Such guidelines need to include reporting standards and EQA of laboratory testing. Given that early spiking schemes were sub-optimal, this will require collection of larger volumes of plasma from affected pregnancies and potentially the use of commercially available reference materials if demonstrated to reflect the nature of 'real' plasma samples. A global pilot EQA will be offered.

MN, United States
OBJECTIVES: Results for cfDNA aneuploidy screening generally do not include a PPV. Therefore, some individuals use published sensitivity and specificity information to determine the PPV for cfDNA testing in their patients. This information has been helpful in the counseling of patients with positive cfDNA screening tests. However, as calculations of PPV are strongly dependent on the sensitivity and specificity of the test, as well as the prevalence of the disorder in the tested population, small inaccuracies, especially in the specificity, have a strong effect on PPV. Therefore, calculated PPVs do not necessarily reflect what's seen in the clinical laboratory and in real-world observations. METHODS: Four reference laboratories in the United States and Australia queried their internal laboratory information systems to identify patients who underwent cytogenetic testing (sample types: products of conception, chorionic villi, amniotic fluid, or blood) after abnormal cfDNA testing and determined if the results were concordant with the screening results. When the result was screen positive for autosomal aneuploidy, and maternal age was available, a PPV for each case was calculated using the PPV calculator developed in conjunction with NSGC (https://www. perinatalquality.org/Vendors/NSGC/NIPT/) which is based on sensitivity and specificity information derived from a metaanalysis of cfDNA studies, as opposed to individual laboratory validation data. RESULTS: A total of 538 cfDNA-positive cases were identified with 384 results confirming the cfDNA results (OAPR 71.4%). The OAPRs for each chromosomal aneuploidy was as follows: T13-48.1%. T18-73.2%. T21-92.1%. Monosomy X-18.7%. XXX-54.6%. XXY-63.3%. And XYY-88.9%. A PPV was calculated for each autosomal aneuploidy case with a maternal age. The average calculated PPV for each disorder was significantly lower than the observed TPR: T21-82.4% v. 91.9% (p-0.003); T18-46.7% v. 75.6% (p = 0.0001); T13-29.2% v. 48.9% (p = 0.0001). CONCLUSIONS: We compared our observed TPRs to those of previously published observational studies and determined that our OAPRs were not statistically different from those seen in previous studies. This indicates that the OAPRs are relatively consistent over time. Additionally, the disparity between the calculated PPV and observed TPR demonstrates that the PPV calculator significantly underestimated the performance of cfDNA screening for all three autosomal aneuploidies. This underestimation, along with the general consistency of OAPR across multiple studies, supports the use of the OAPR as a useful tool in the counseling of patients with positive cfDNA screening results.

7-4
Fetal aneuploidy screening results in maternal plasma samples redrawn due to insufficient fetal cfDNA in the initial sample Thomas Musci 1 , Maximilian Schmid 1 , Ken Song 2 , Eric Wang 1 , Kathleen Fergus 1 1 Ariosa Diagnostics, Inc., San Jose, CA, United States 2 Roche, San Jose, CA, United States OBJECTIVES: The objective of this study was to determine the redraw success rate of samples initially receiving no cfDNA result due to low fetal fraction (FF) and to characterize the results of redrawn samples. METHODS: A review of initial cfDNA results of 229 454 samples submitted to Ariosa Diagnostics for HarmonyTM Prenatal Test was conducted. cfDNA samples were reviewed and categorized as having an increased probability for trisomy 21, 18 and 13, decreased probability for trisomy and no reportable results due to low FF. Samples that received no reportable results due to low FF (<4%) were matched to repeat cfDNA samples. The results of redrawn samples were categorized and compared to the results of samples receiving a result on their initial cfDNA sample. The overall and redraw success rates were evaluated. Pregnancy outcome data was not obtained. RESULTS: Of the 229 454 samples submitted, 97.4% (223 649) received a result on their initial cfDNA sample, with 1.49% (3287) receiving results indicating an increased probability for trisomy 21, 18, or 13. Non-reportable results due to low fetal fraction were seen in 1.8% (4100) of the initial samples. Of this group, 74.2% (3041) elected to have a repeat cfDNA sample collected, with 68% (2058) receiving a reportable result on their second sample. While the mean fetal fraction was lower in the redrawn samples (6.58% compared to 11.49% in the initial samples), reportable results indicating a low probability for trisomy 21, 18. CONCLUSIONS: Of cfDNA samples redrawn after an initial sample failure due to low fetal fraction, 68% received a result with 98% of those results indicating a low probability for the common autosomal trisomies. It has been recommended that providers refer women receiving 'no call' cfDNA screen results for diagnostic testing because of an increased aneuploidy risk. Reflexing to invasive testing would result in an increased number of invasive prenatal diagnostic tests and concomitant increase in fetal loss. This study shows that submitting a second sample after an initial non-reportable cfDNA result due to low fetal fraction will yield a reportable result. Correspondence between autopsy and ultrasound was defined as agreement (same diagnosis), additional (additional findings undetected by ultrasound), unconfirmed (false positive/false negative ultrasound). PRISMA guidelines. RESULTS: A total of 4191 fetuses were pooled. Agreement was 70%, additional 21%, unconfirmed 9%. False positive and false negative rates were 3% and 4%, respectively. Malformation was described in 3153 (75%) fetuses, in which TOP occurred frequently for CNS (34%) and rarely for thorax (1%) malformations. The highest agreement between autopsy and PD was observed in GU anomalies (81%), followed by CNS (80%), genetics (79%), skeleton (77%), CHD (75%), thorax (70%); GI (63%), multiple (37%), limbs (23%). False positive rates were highest for thorax and lowest for CNS malformations. False negative rates were highest for limb and lowest for CNS malformations. CONCLUSIONS: Fetal autopsy is indicated in cases of TOP for malformations, since it provides additional information in 21% of cases. In only 3% of cases, ultrasound findings lead to unjustified TOP and this is mainly true for thorax anomalies.

8-2
Head circumference of fetuses with congenital heart disease decreases in the second half of pregnancy Ilse Haveman 1 , Roel de Heus 1 , Edu Mulder 1 , Hanneke Fleurke-Rozema 2 , Katia Bilardo 2 OBJECTIVES: Congenital heart disease (CHD) occurs in 6 to 8 per 1000 livebirths and is associated with poor perinatal outcome and risk of neurodevelopmental impairment. Brain injury, altered brain development and smaller brain volumes have been reported in children with CHD prior to cardiac surgery. Smaller fetal head circumference (HC) and microcephaly at birth, considered markers of fetal brain development, are reported in CHD cases. The purpose of this study is to evaluate the fetal HC growth pattern in isolated CHD. METHODS: This retrospective cohort study was performed at University Medical Center Utrecht and University Medical Center Groningen. Pregnancies with an EDD between 2008 and 2013 and isolated CHD were included and classified into seven groups: aortic arch obstruction, conotruncal anomaly, hypoplastic left heart, hypoplastic right heart, septal defect, complete transposition of the great arteries and others. Fetal biometry was assessed at four variable time points between 18 and 35 weeks' gestation (GA). Values were transformed into Zscores. Linear mixed modeling was performed to enable analysis of repeated measurements and construction of growth models. RESULTS: A total of 246 livebirths with isolated CHD (chromosomal abnormalities were excluded) were eligible for analysis. The linear growth model predicted a smaller HC at 36 weeks' GA in fetuses with a conotruncal anomaly compared to a reference group: mean Z-score À0.57, SD 0.78 and mean Z-score 0, SD 1, respectively (p < 0.0005) (Fig. 1). HC growth pattern showed a trend for a linear decrease starting from the 2nd trimester in all CHD groups , whereas abdominal circumference(AC) and femur length(FL) showed a constant growth pattern. Mean difference in HC Z-score between the first and last measurements was À0.64 SD (p < 0.0001). CONCLUSIONS: Fetuses with a conotruncal heart anomaly have smaller HC at 36 weeks' GA. Furthermore, all CHD groups showed a slight deflection in HC growth from the 2nd trimester onwards. Data on neurological outcome were not available and therefore a possible relation between HC and neurological impairment could not be analysed. However, taking into account the modest reduction seen in our study we hypothesize that HC alone may not be a reliable marker for neurodevelopmental delay in fetuses with CHD. OBJECTIVES: Migration takes place in the first and early-second trimesters and phenotype of migration in the cortex appears after 28 weeks of gestation. It has been believed that it is quite hard to detect migration disorder before 28 weeks. This study shows our experience of cases with neuronal migration disorder between 15 and 26 weeks. METHODS: Twenty-eight cases with diagnoses of migration disorder between 2011 and 2015 were enrolled in this study. Cases were divided into two groups by date of sonographic detection before and after 22 + 0 weeks. In terminated case, autopsy was done with parental consent. The rest of cases, the outcome and prognosis were investigated. RESULTS: All cases with migration disorders were complicated with other CNS and/or exCNS abnormalities. Among 28 cases, all 12 cases diagnosed as migration disorder between 15 and 21 weeks of gestation, were terminated before 22 weeks. In seven out of 12 cases, autopsy was done and brain histology was carefully examined and neuronal maldevelopment was confirmed. Among the rest of 16 cases diagnosed as migration disorder between 22 and 26 weeks of gestation, IUFD (1), infant death (1) and 12 fetuses were delivered and followed up, and two cases are still ongoing. There were no intact infants among 12 delivered infants. CONCLUSIONS: In most cases with neuronal migration disorder, neurological prognosis is not favorable. Early diagnosis of migration disorder has been possible and is essential for proper prenatal counseling, genetic investigation and management.

8-4
Diagnostic accuracy of congenital heart defects after referral from routine ultrasound screening Ilse Haveman 1 , Roel de Heus 2 , Katia Bilardo 2 , Hanneke Fleurke-Rozema 2 undergoing post-mortem evaluation following pregnancy termination due to abnormal ultrasound findings or unexplained fetal demise. There is a progressive increase in diagnostic performance as additional investigations are performed, some universally, and others as indicated. Exome sequencing is contributory and should be performed in cases with high clinical suspicion of a single gene disorder.

TA-1
The effect of a decision aid on informed decision making in the era of non-invasive prenatal testing: A randomized controlled trial OBJECTIVES: The intellectual disability in Down syndrome (DS) originates with delays in neurogenesis and brain growth during fetal brain development. We hypothesize that novel prenatal treatments can be identified by targeting specific signaling pathways that are consistently perturbed in fetal tissues from humans with DS and in several mouse models. METHODS: We generated new gene expression datasets from amniocytes from seven mid-trimester fetuses with trisomy 21, age and sex-matched euploid fetuses, and embryonic day 15.5 forebrain from Ts1Cje, Ts65Dn, and Dp16 mouse models of DS, along with littermate controls. Functional pathway analyses were performed using the GSEA, DAVID and IPA databases. The new datasets (n = 4) were compared to other publicly available datasets from fetal cerebrum and cerebellum, amniotic fluid, iPSCs and neurons derived from humans with DS (total = 9). We used the Connectivity Map database and developed a murine adaptation to identify FDA-approved drugs that can rescue consistently dysregulated pathway abnormalities. RESULTS: USP16 and TTC3 were dysregulated in all affected human cells and two mouse models. Functional pathway analyses revealed similarities and differences between humans with DS and mouse models. DSassociated pathway abnormalities were either the result of gene dosage specific effects (increased transcriptional activity, abnormal neurogenesis and neuronal differentiation, mitochondrial dysfunction, oxidative stress and inflammation) or the consequence of a global cell stress response with activation of compensatory mechanisms (abnormal cell cycle and kinetochore regulation, increased proteolysis and activation of anti-apoptotic gens). Using a high stringency cut-off, CMap analyses identified seventeen molecules with high predictive scores to rescue abnormal gene expression. CONCLUSIONS: Our novel integrated human/murine systems biology approach identified commonly dysregulated genes and pathways in both humans with DS and mouse models that can help to prioritize therapeutic molecules for further study. Studies testing the safety and efficacy of the top candidate molecules on human cells are ongoing prior to pre-clinical prenatal treatment in mice.

TA-3
Amniotic fluid transcriptomic changes in fetuses with myelomeningocele Archived AFS from sex and gestational age-matched euploid fetuses (n = 10) without MMC were used as controls. AFS cellfree RNA was isolated, processed and hybridized to Affymetrix GeneChip® Human Genome U133 Plus 2.0 arrays. Differentially-regulated genes were identified using paired ttests (p < 0.05). Significantly altered expression in genes, pathways and networks was identified using Ingenuity Pathway Analysis® using a right-tailed Fisher Exact Test (p < 0.05). Multiple testing was performed with corrected pvalues (<0.05) based on the Benjamini-Hochberg method. RESULTS: Control samples were from fetuses at slightly younger gestational ages (20.9+/À0.9 wks) than MMC samples (24.5+/ À1.0 wks). Fetuses with MMC had characteristic gene expression patterns. 284 differentially-regulated genes were identified (176 up-regulated and 108 down-regulated) in AFS. Known genes associated with MMC (PRICKLE2, GLI3, RAB23, HES1, FOLR1) were differentially regulated. In addition, novel dysregulated genes were identified in association with neurodevelopment and neuronal regeneration (up-regulated GAP43 and ZEB1) or axonal growth and guidance (down-regulated ACAP1). Pathway analysis demonstrated a significant contribution of inflammation (Z-score 2.45) to pathology and a broad influence of Wnt signaling pathways (Wnt1, Wnt5A, ITPR1, Z-score -1.3). CONCLUSIONS: Transcriptomic analyses of living fetuses with MMC using AFS cfRNA demonstrated differential regulation of specific genes and molecular pathways relevant to this CNS disorder, resulting in a new understanding of pathophysiological changes. The data also suggested the importance of pathways involving secondary pathology, such as inflammation, in MMC. These newly identified pathways may lead to hypotheses that can test novel therapeutic targets as adjuncts to fetal surgical repair. OBJECTIVES: Analysis of cell free DNA (cfDNA) in maternal plasma for the detection of fetal aneuploidy is available worldwide, but there is very limited availability of NIPD for women with pregnancies at increased risk of monogenic disorders. We offer a NIPD service for detection of paternal or de-novo alleles for the diagnosis of FGFR3 and FGFR2-related conditions and for cystic fibrosis in pregnancies where parents are heterozygous for different mutations. Our objective here is to expand the range of NIPD available to families at risk of a range of rare monogenic conditions as a safe, earlier alternative to invasive testing. METHOD: Primers were designed to target relevant mutations and included P5 and P7 Illumina sequencing adapters along with incorporation of a 6bp unique index. Bespoke primers were tested on paternal and/or proband DNA to verify the design before application in cfDNA. For cfDNA, in addition to the mutation specific primer, primers for HLA and ZFX/ZFY were included for each patient enabling us to confirm the presence