Normal prenatal ultrasound findings reflect outcome in case of trisomy 14 confined placental mosaicism developing after preimplantation genetic diagnosis

Preimplantation genetic diagnosis (PGD) is a well-established technology that is widely applied in assisted reproductive medicine. However, misdiagnosis by PGD can still occur and thus invasive prenatal diagnosis may be necessary1. Until recently, non-invasive cell-free DNA (cfDNA) testing was advocated for confirmation of a PGD result. However, in certain situations, for example, when there is confined placental mosaicism (CPM), results of cfDNA testing might not correlate with the clinical phenotype2. We describe a case of a normal liveborn baby with a prenatal karyotype of mosaic trisomy 14, despite being conceived after PGD. Detailed prenatal genetic assessments, together with frequent ultrasound examinations, indicated a possible case of CPM, which was confirmed by postnatal diagnosis of the neonate and placenta. The timeline of our investigations is illustrated in Figure 1. A 30-year-old nulliparous woman opted for PGD because she was a carrier of Robertsonian translocation, rob(14;21). During the in-vitro fertilization cycle, a total of five day-5 embryos underwent trophectoderm biopsy (6–9 cells were obtained from each embryo), of which four had nuclei available for PGD. Fluorescence in-situ hybridization (FISH) for chromosomes 14 and 21 using the Qbiogene TEL 14q DNA probe (Qbiogene Inc., Carlsbad, CA, USA) and Vysis LSI 21 SpectrumOrange probes (Abbott Molecular, Des Plaines, IL, USA) revealed two embryos with a normal/balanced karyotype, one with trisomy 21 and one with monosomy 21 (Figure 2a–d). Singleton pregnancy was achieved after the two normal/balanced embryos were transferred. At 12 weeks of gestation, cfDNA testing with genome-wide normalized score algorithm3 was performed at the woman's request and the result indicated a boundary risk for trisomy 14 (P = 0.05). To exclude the possibility of PGD misdiagnosis, chorionic villus sampling (CVS) was performed at 14 weeks. Interphase FISH using IGH/CCND1 dual color translocation probes (Cytocell, Cambridge, UK) for villi revealed 30% of cells with trisomy 14 (nuc ish(IGH × 3, CCND1 × 2)[30/100]). Analysis of informative short tandem repeat (STR) markers for chromosome 14 revealed evidence of uniparental disomy (UPD) in partial cells. The woman agreed to continue pregnancy after non-directive genetic counseling as there was a possibility of CPM. Interphase FISH for cultured amniocytes at 16 weeks showed that the percentage of mosaic trisomy 14 had declined to 15% (nuc ish(IGH × 3, CCND1 × 2)(15/100)) and karyotyping showed 18% of trisomy 14 (46,XX,+14,der(14;21)(q10;q10)[4]/45,XX,der(14;21)(q10;q10)[18]) (Figure 2e,f). STR analysis revealed no UPD of chromosome 14. Between 16 and 37 weeks, monthly antenatal care was provided. The pregnancy was uneventful and all ultrasound examinations were normal. At 37 + 1 weeks, a normal female neonate was born via vaginal delivery, weighing 2700 g, with Apgar scores of 8 and 9 at 1 min and 5 min, respectively. For postnatal diagnosis, interphase FISH using Qbiogene TEL 14q DNA probe (Qbiogene) and TelVysion 21q SpectrumOrange probe (Abbott Molecular) was performed on neonatal cord blood, umbilical cord and placenta. All tissues presented disomies 14 and 21 (nuc ish(D14S1419,VIJyRM2029) × 2[100]) except the placenta, which showed 2% 14qter deletion/21qter duplication (nuc ish(D14S1419 × 1,VIJyRM2029 × 3)[2/100]) and 2% 14qter duplication/21qter deletion (nuc ish(D14S1419 × 3,VIJyRM2029 × 1)[2/100]) (Figure 2g–j). The discrepancy between PGD results and prenatal diagnosis may originate from the chromosomal mosaicism in the biopsied embryos or subsequent mitotic errors after PGD. However, trophectoderm biopsy has the advantage of collecting multiple cells and chromosomal mosaicism was not found by PGD using FISH. We therefore considered that the most likely explanation in this case was a chromosome segregation error during postzygotic mitosis. Initial embryonic cell divisions are known to be highly error-prone. In addition, the genetic instability linked to inter-chromosome effect has been proposed in early embryos from female carriers of Robertsonian translocation4. It is noteworthy that postnatal diagnosis confirmed CPM in this case, while different cell lines found in the analyses of both CVS and amniocytes suggested fetoplacental mosaicism during the first to early second trimester. A mechanism explaining the CPM and normal fetal outcome from early fetoplacental mosaicism may involve aneuploidy rescue5. The finding of spontaneous chromosomal correction by a decline in the percentage of cells with trisomy 14 is compatible with our previous report, in which a fetus with 100% trisomy 13 eventually regressed to a completely normal baby (100% disomy 13 for cultured cells from cord blood at birth) by a continuous decline in the percentage of cells with trisomy 13 as the gestation progressed5. As in the case presented here, most CPM pregnancies are associated with good postnatal outcome. However, some abnormal chromosomal complements may lead to pregnancy complications due to either placental insufficiency or fetal UPD caused by trisomy rescue6, 7. Therefore, when encountering CPM, detailed ultrasound and STR marker analysis should be performed to distinguish it from UPD (e.g. UPD in imprinted chromosomes 6, 7, 11, 14 and 15 have phenotypic effects associated with a known syndrome6) or a CPM condition associated with adverse fetal outcomes (e.g. trisomy 16, the most common aneuploidy observed in CPM, is associated with intrauterine growth restriction7). In our case, normal anatomy and growth pattern on serial fetal ultrasound scans without UPD provided reassurance to the patient that CPM was highly likely and thus a favorable prognosis could be anticipated. In addition to the unusual fetal condition, it is also important to note that results of cfDNA testing performed in early pregnancy were not conclusive for trisomy 14 mosaicism given that the result showed a boundary risk of trisomy 14. At 12 weeks, the concentration of fetal DNA was estimated as 9.74%, therefore the theoretical fetal DNA fraction of the trisomy-14 cells in maternal plasma was 2.92% (i.e. 9.74% × 30%), which is lower than the proposed lowest limit of 4% for fetal DNA fraction that is feasible for cfDNA testing3. This report demonstrates that normal prenatal ultrasound findings are an important prognostic indicator in the rare situation of discrepancy between PGD and subsequent genetic investigations during pregnancy. Awareness of the possible underlying mechanisms for the discrepancy, such as postzygotic mitotic errors and aneuploidy rescue seen in our case, should aid management and counseling of affected pregnancies. This study was kindly supported by research grants from the Ministry of Science and Technology (MOST 104-2314-B-371-009-MY3 to M.C.) and Changhua Christian Hospital (101-CCH-IRP-40 to G.-C.M. and 102-CCH-IRP-034 to M.C.), Taiwan. W.-J. Wu1,2,3#, G.-C.Ma1,4,5#, M.-H. Lee1, Y.-C. Chen2 and M. Chen1,2,6,7,8* 1Department of Genomic Medicine and Center for Medical Genetics, Changhua Christian Hospital, Changhua, Taiwan; 2Department of Genomic Science and Technology, Changhua Christian Hospital Healthcare System, Changhua, Taiwan; 3Department of Obstetrics and Gynecology, Changhua Christian Hospital, Changhua, Taiwan; 4Institute of Biochemistry, Microbiology and Immunology, Chung Shan Medical University, Taichung, Taiwan; 5Department of Medical Laboratory Science and Biotechnology, Central Taiwan University of Science and Technology, Taichung, Taiwan; 6Department of Obstetrics and Gynecology, College of Medicine and Hospital, National Taiwan University, Taipei, Taiwan; 7Department of Medical Genetics, National Taiwan University Hospital, Taipei, Taiwan; 8Department of Life Science, Tunghai University, Taichung, Taiwan *Correspondence. (e-mail: [email protected]; [email protected]) #W.-J.W. and G.-C.M. contributed equally to management and documentation of this case.