Analyzing products-of-conception specimens is challenging as a result of culture failure, which occurs in 10–40% of cases,4–6 and selective overgrowth of maternally derived cells, which may account for 29–58% of cases.3 Although altering techniques for handling and processing products-of-conception material can decrease culture failure rates to less than 10%,2 maternal cell contamination prevents reporting a true fetal karyotype with certainty. Using a previously reported informatics approach that analyzes SNPs, considers parental genotypic information, and does not require cell culture,8,9 we identified and excluded samples with significant maternal cell contamination, concurrently avoiding the inherent issues with culture failure rates. In this cohort, 2,392 of 2,400 (99.7%) of specimens were less than 20 weeks of gestation; results were obtained in 2,389 (99.9%) of patients. By contrast, typical culture failure rates would translate to failure for approximately 240–960 patients. More importantly, the ability to identify specimens with maternal cell contamination allowed reporting of meaningful and accurate fetal results. Elimination of these cases prevented 528 women from receiving a misleading “normal 46,XX” result and improved the male-to-female sex ratio to 0.91, far greater than the mean ratio of 0.71 reported from multiple large products-of-conception studies.2
Of the remaining 1,861 fetal samples, and at the resolution of traditional cytogenetic G-band analysis (greater than 10 Mb; Fig. 1), 755 (40.6%) showed a normal result (in 74 [10.4%] of these cases, conception was facilitated through in vitro fertilization using an egg donor [see Appendix 3, available online at http://links.lww.com/AOG/A534]) and 1,106 (59.4%) showed classical cytogenetic abnormalities. Of these 1,106 cases, aneuploidy (single and multiple aneuploidies) accounted for 945 (85.4%), triploidy for 114 (10.3%), and structural anomalies or tetraploidy for the remaining 47 (4.2%) cases.
The majority of the 860 of 1,106 (77.8%) single aneuploidies were trisomies (794/860 [92.3%]); the remainder were monosomies (66/860 [7.7%]). Trisomy of every chromosome except 1 and 19 was identified; the most common was trisomy 16 (199/794 [25.1%]) followed by trisomy 22 (158/794 [19.9%]) (Appendix 4, available online at http://links.lww.com/AOG/A534). Of the 794 trisomic cases, 767 (96.6%) were maternally derived, and only 27 (3.4%) were paternally derived. By contrast, monosomic cases (66/860 [7.7%]) were limited to the X chromosome in 53 (80.3%) cases and to chromosome 21 in 13 (19.7%) cases, whereas the majority of monosomy X cases (36/53 [67.9%]) were paternal in origin and most monosomy 21 cases (11/13 [84.6%]) were maternal in origin.
Aneuploidies involving two or more chromosomes were observed in 85 of 1,106 (7.7%) of abnormal cases; double aneuploidies accounted for 77 of 85 (90.6%) of these cases (Fig. 1). The majority of double aneuploidies (58/77 [75.3%]) were maternally derived. Mixed origins were observed in 17 (22.1%) of the 77 double aneuploidies and two (25%) of the eight triple aneuploidies.
Appendix 5, available online at http://links.lww.com/AOG/A534, shows the predicted structural abnormality for the 38 cases of partial aneuploidy (3.4% of all abnormal cases with imbalances greater than 10 Mb). Unbalanced translocations represent the single largest category. Cases with partial aneuploidies greater than 10 Mb and an imbalance of less than 10 Mb are indicated as are partial aneuploidies between 10 and 15 Mb. The three cases with imbalances between 10 and 13 Mb may have been called normal by routine cytogenetic analysis. Five cases had both aneuploidy and structural abnormalities (Appendix 6, available online at http://links.lww.com/AOG/A534).
Of the 114 triploid samples, 98 (86.0%) were complete triploidies; the remainder (16/114 [14%]) presented as hypo-, hyper-, or pseudotriploidy (Fig. 1). Approximately two thirds of triploids were digynic in origin and the remaining one third was diandric, which is important for molar pregnancy-associated risks. Only four cases of tetraploidy were observed (0.36% of abnormal cases; for further discussion, see Appendix 7, available online at http://links.lww.com/AOG/A534). Approximately 1% of cases considered normal at the G-band level had uniparental disomy (Fig. 2). Whole-genome uniparental disomy was identified in three (0.4%) of the 755 samples, two of which were androgenetic and associated with complete molar pregnancies. Uniparental disomy involving single chromosomes was identified in another four samples, all of which were maternal uniparental disomy, raising the suspicion of an initial trisomy conception with subsequent rescue. Three uniparental disomy cases were observed in cytogenetically abnormal cases, including one case with a maternally derived chromosome 12 marker chromosome and maternal uniparental disomy 12, presumably reflecting a trisomy rescue event.
Chromosomal imbalances (microdeletions and microduplications) below traditional cytogenetic G-band analysis resolution (less than 10 Mb) were also identified using SNP chromosomal microarray analysis (Fig. 2). Specifically, of the 755 samples considered normal at the G-band level, 33 (4.4%) had a copy number change ranging in size from 400 Kb to 9.5 Mb; 12 (36.4%) were classified as clinically significant (Appendix 8, available online at http://links.lww.com/AOG/A534) and the remaining were considered variants of unknown significance (Appendix 9, available online at http://links.lww.com/AOG/A534). The average size of clinically significant copy number changes was larger than that of variants of unknown significance (5.81 compared with 2.12 Mb, respectively). Twenty samples showed copy number changes of less than 10 Mb in addition to aneuploidy; four (20%) of these were classified as clinically significant (Appendix 8, http://links.lww.com/AOG/A534), whereas the remainder were considered variants of unknown significance (Appendix 9, http://links.lww.com/AOG/A534). The average size of clinically significant copy number changes in this group was also larger than that of variants of unknown significance (2.68 compared with 2.01 Mb, respectively).
This study reports a large-scale proof-of-concept for microarray analysis of products-of-conception specimens when clinically indicated. Large-scale studies of chromosomal microarray analysis in prenatal and stillbirth specimens have recently been reported.13–15 As such, the American College of Obstetricians and Gynecologists (the College) recommends the use of chromosomal microarray analysis in cases of stillbirth or intrauterine fetal demise.16 However, chromosomal microarray analysis reports of products-of-conception specimens are limited to small array comparative genomic hybridization cohort studies focused on karyotypically normal products of conception, fetuses with multiple structural anomalies, or fetal specimens that failed to grow in culture.15,17–24 The College currently does not recommend chromosomal microarray analysis for first- and second-trimester losses owing to limited data.16 The current study reports the consecutive analysis of a large number of product-of-conception cases (greater than 2,000) without preselection using a SNP-based platform. Importantly, chromosomal microarray analysis has been reported to detect 13% additional abnormalities in multiple small-array comparative genomic hybridization cohorts compared with karyotyping,25 and Appendix 10, available online at http://links.lww.com/AOG/A534, describes cases in the current study that would be identifiable both by traditional cytogenetics and SNP chromosomal microarray analysis. This indicates that at minimum, chromosomal microarray analysis will detect the same cohort of clinically relevant abnormalities and, at best, will identify additional clinically relevant abnormalities when compared with traditional cytogenetic testing.
Analyzing products-of-conception specimens has traditionally been challenging as a result of culture failure4–6 and selective overgrowth of maternally derived cells.3 Although altering handling and processing of products-of-conception material may decrease culture failure rates to less than 10%,2 maternal cell contamination prevents reporting true fetal karyotypes with certainty. Here, typical culture failure rates would translate to failure for approximately 240–960 patients. More importantly, the ability to identify maternal cell contamination allowed reporting of meaningful and accurate fetal results for 1,861 patients and prevented 528 women from receiving a misleading “normal 46,XX” result.
The use of chromosomal microarray analysis for products-of-conception specimens is appealing given that banding resolutions are often at or below 400–450 bands, where genomic imbalances of greater than 10 Mb should be readily detectable but are often missed using standard G-banding.7,26–28 In our study, 18% of partial aneuploidies may have been missed by standard G-band analysis because they had imbalances at the 10- to 15-Mb threshold. In some cases, a second cytogenetically visible aberration was evident and if the cryptic imbalance (10–15 Mb) was not concurrently detected, the reported structural abnormality would not have accurately reflected the true fetal karyotype. Unbalanced rearrangements carry reproductive implications if the balanced form is found in a carrier parent on subsequent testing. Given that 0.5–5% of couples with recurrent miscarriages carry a balanced rearrangement,26 it is imperative that such cases are discovered at the time of products-of-conception analysis so appropriate parental follow-up studies are initiated.
Aneuploidy is expected to be the principal factor leading to pregnancy loss; the concurrent finding of a copy number change (2.1% of abnormal cases) is likely coincidental. However, 4.3% (P≤.003) of samples that appeared cytogenetically normal had a copy number change with potential or known clinical significance. The significantly higher incidence of copy number changes in the cytogenetically normal group indicates a nonincidental finding that likely contributed to miscarriage causality. For pathogenic copy number changes, this is most likely related to the size of the imbalance. Pathogenic copy number change frequency (1.6% [12/755]) was similar to the recent surprising finding reporting clinically significant copy number changes in 1.7% of patients with normal karyotypes who were referred for prenatal diagnosis because of advanced maternal age or a positive screen.14 This offers insight into the frequency with which imbalances of less than 10 Mb occur in the population and underscores the need for using microarray technology for cytogenomic investigation of prenatal and products-of-conception specimens. Indeed, the College and the International Society for Prenatal Diagnosis now recommend chromosomal microarray analysis as a first-tier test for all pregnant women with ultrasound anomalies and for the analysis of stillbirths.16
Submicroscopic imbalances in products of conception raise various questions regarding miscarriage causality. Most described pathogenic copy number changes have phenotypes that are distinguishable in pediatric or adult patients and involve a range of abnormalities not yet apparent in the first trimester. Added complexity lies in the extreme variability that is observed in many of these syndromes (Appendix 11, available online at http://links.lww.com/AOG/A534). Thus, their detection in a products-of-conception specimen has major clinical implications because parents themselves may be carriers of the same copy number change, with a 50% recurrent risk in all future pregnancies.
The use of SNPs allowed for the identification of three cases with whole-genome uniparental disomy, two of which were androgenetic and were associated with complete molar pregnancies. All single-chromosome uniparental disomy cases were of maternal origin and may represent a trisomy rescue event. Although these uniparental disomy cases did not have copy number imbalances, trisomic cells in other vital tissues cannot be ruled out. This is supported by multiple uniparental disomy reports with prenatally diagnosed trisomy mosaicism.27,28 Although uniparental disomy in products-of-conception specimens is more likely indicative of trisomy mosaicism, pathogenic uniparental disomy effects may also be the result of homozygosity of a lethal autosomal-recessive mutation. Here, most uniparental disomy cases did not involve chromosomes known to be imprinted, and large cohort studies using SNP chromosomal microarray analysis are required to assess the effect of imprinted chromosomes on miscarriage causality.
Establishing miscarriage causality helps reduce self-blame during the grieving process and provides a basis for estimating reproductive recurrence risks (Lathi RB, Huynh D, Keller J, Dikan J, Rabinowitz M. Patient desire for chromosome analysis of products of conception following miscarriage: a national survey [abstract]. Fertil Steril 2011;96:S91).29 Microarray technology eliminates pitfalls associated with traditional cytogenetic products-of-conception specimen analysis. The ability to analyze nonviable tissue potentially yields chromosomal microarray analysis success rates in excess of 99%, translating to higher diagnostic returns, and interrogating SNPs extends the scope of detectable genomic abnormalities. Significantly, SNP chromosomal microarray analysis facilitates reporting “true” fetal results. Together, this supports use of SNP chromosomal microarray analysis for cytogenomic evaluation of products-of-conception specimens when clinically indicated.
1. Clendenin TM, Benirschke K. Chromosome studies on spontaneous abortions. Lab Invest 1963;12:1281–92.
2. Menasha J, Levy B, Hirschhorn K, Kardon NB. Incidence and spectrum of chromosome abnormalities in spontaneous abortions: new insights from a 12-year study. Genet Med 2005;7:251–63.
3. Bell KA, Van Deerlin PG, Haddad BR, Feinberg RF. Cytogenetic diagnosis of “normal 46,XX” karyotypes in spontaneous abortions frequently may be misleading. Fertil Steril 1999;71:334–41.
4. Dejmek J, Vojtassák J, Malová J. Cytogenetic analysis of 1508 spontaneous abortions originating from south Slovakia. Eur J Obstet Gynecol Reprod Biol 1992;46:129–36.
5. Kajii T, Ferrier A, Niikawa N, Takahara H, Ohama K, Avirachan S, et al.. Anatomic and chromosomal anomalies in 639 spontaneous abortuses. Hum Genet 1980;55:87–98.
6. Stephenson MD, Awartani KA, Robinson WP. Cytogenetic analysis of miscarriages from couples with recurrent miscarriage: a case-control study. Hum Reprod 2002;17:446–51.
7. Shaffer LG, Bejjani BA. A cytogeneticist's perspective on genomic microarrays. Hum Reprod Update 2004;10:221–6.
8. Firth HV, Richards SM, Bevan AP, Clayton S, Corpas M, Rajan D, et al.. DECIPHER: Database of Chromosomal Imbalance and Phenotype in Humans Using Ensembl Resources. Am J Hum Genet 2009;84:524–33.
9. MacDonald JR, Ziman R, Yuen RK, Feuk L, Scherer SW. The database of genomic variants: a curated collection of structural variation in the human genome. Nucleic Acids Res 2014;42:D986–92.
10. Riggs ER, Wain KE, Riethmaier D, Savage M, Smith-Packard B, Kaminsky EB, et al.. Towards a Universal Clinical Genomics Database: the 2012 International Standards for Cytogenomic Arrays Consortium Meeting. Hum Mutat 2013;34:915–9.
11. Shaikh TH, Gai X, Perin JC, Glessner JT, Xie H, Murphy K, et al.. High-resolution mapping and analysis of copy number variations in the human genome: a data resource for clinical and research applications. Genome Res 2009;19:1682–90.
12. Cooper GM, Coe BP, Girirajan S, Rosenfeld JA, Vu TH, Baker C, et al.. A copy number variation morbidity map of developmental delay. Nat Genet 2011;43:838–46.
13. Reddy UM, Page GP, Saade GR, Silver RM, Thorsten VR, Parker CB, et al.. Karyotype versus microarray testing for genetic abnormalities after stillbirth. N Engl J Med 2012;367:2185–93.
14. Wapner RJ, Martin CL, Levy B, Ballif BC, Eng CM, Zachary JM, et al.. Chromosomal microarray versus karyotyping for prenatal diagnosis. N Engl J Med 2012;367:2175–84.
15. Menten B, Swerts K, Delle Chiaie B, Janssens S, Buysse K, Philippé J, et al.. Array comparative genomic hybridization and flow cytometry analysis of spontaneous abortions and mors in utero samples. BMC Med Genet 2009;10:89.
16. The use of chromosomal microarray analysis in prenatal diagnosis. Committee Opinion No. 581. American College of Obstetricians and Gynecologists. Obstet Gynecol 2013;122:1374–7.
17. Schaeffer AJ, Chung J, Heretis K, Wong A, Ledbetter DH, Lese Martin C, et al.. Comparative genomic hybridization-array analysis enhances the detection of aneuploidies and submicroscopic imbalances in spontaneous miscarriages. Am J Hum Genet 2004;74:1168–74.
18. Benkhalifa M, Kasakyan S, Clement P, Baldi M, Tachdjian G, Demirol A, et al.. Array comparative genomic hybridization profiling of first-trimester spontaneous abortions that fail to grow in vitro. Prenatal Diagn 2005;25:894–900.
19. Shimokawa O, Harada N, Miyake N, Satoh K, Mizuguchi T, Niikawa N, et al.. Array comparative genomic hybridization analysis in first-trimester spontaneous abortions with “normal” karyotypes. Am J Med Genet A 2006;140:1931–5.
20. Robberecht C, Schuddinck V, Fryns JP, Vermeesch JR. Diagnosis of miscarriages by molecular karyotyping: benefits and pitfalls. Genet Med 2009;11:646–54.
21. Zhang YX, Zhang YP, Gu Y, Guan FJ, Li SL, Xie JS, et al.. Genetic analysis of first-trimester miscarriages with a combination of cytogenetic karyotyping, microsatellite genotyping and array CGH. Clin Genet 2009;75:133–40.
22. Rajcan-Separovic E, Diego-Alvarez D, Robinson WP, Tyson C, Qiao Y, Harvard C, et al.. Identification of copy number variants in miscarriages from couples with idiopathic recurrent pregnancy loss. Hum Reprod 2010;25:2913–22.
23. Rajcan-Separovic E, Qiao Y, Tyson C, Harvard C, Fawcett C, Kalousek D, et al.. Genomic changes detected by array CGH in human embryos with developmental defects. Mol Hum Reprod 2010;16:125–34.
24. Warren JE, Turok DK, Maxwell TM, Brothman AR, Silver RM. Array comparative genomic hybridization for genetic evaluation of fetal loss between 10 and 20 weeks of gestation. Obstet Gynecol 2009;114:1093–102.
25. Dhillon RK, Hillman SC, Morris RK, McMullan D, Williams D, Coomarasamy A, et al.. Additional information from chromosomal microarray analysis (CMA) over conventional karyotyping when diagnosing chromosomal abnormalities in miscarriage: a systematic review and meta-analysis. BJOG 2014;121:11–21.
26. Gardner RJM, Sutherland GR, Shaffer LG. Chromosome abnormalities and genetic counseling. New York (NY): Oxford University Press; 2012.
27. Kotzot D. Prenatal testing for uniparental disomy: indications and clinical relevance. Ultrasound Obstet Gynecol 2008;31:100–5.
28. Kotzot D, Utermann G. Uniparental disomy (UPD) other than 15: phenotypes and bibliography updated. Am J Med Genet A 2005;136:287–305.
29. Nikcevic AV, Tunkel SA, Kuczmierczyk AR, Nicolaides KH. Investigation of the cause of miscarriage and its influence on women's psychological distress. Br J Obstet Gynaecol 1999;106:808–13.
Supplemental Digital Content
© 2014 by The American College of Obstetricians and Gynecologists.