Sensitivity and specificity of the sequencing test to detect T21 in the analysis population (n=493) were 100% (95% CI 95.9–100.0) and 100% (95% CI 99.1–100.0), respectively (Table 4 and Fig. 3A). This included correct classification for one complex T21 karyotype, 47,XX, inv(7)(p22q32),+21, and two translocation T21 cases arising from Robertsonian translocations, one of which was also mosaic for monosomy X (45,X,+21,der(14;21)q10;q10)/46,XY,+21,der(14;21)q10;q10) and 46,XY,+21,der(21;21)q10;q10). Sensitivity and specificity to detect T18 in the analysis population (n=496) were 97.2% (95% CI 85.5–99.9) and 100% (95% CI 99.2–100.0) (Table 4 and Fig. 3B). Although censored (as per protocol) from the primary analysis, four samples with mosaic karyotypes for T21 and T18 were all correctly classified by massively parallel sequencing as “affected” for aneuploidy (Table 2). Because they were correctly detected they are indicated on the left side of Figure 3A and B. All remaining censored samples were correctly classified as unaffected for trisomies 21, 18, and 13 (Table 2). Sensitivity and specificity to detect T13 in the analysis population were 78.6% (95% CI 49.2–99.9) and 100% (95% CI 99.2–100.0) (Fig. 3C). One T13 case detected arose from a Robertsonian translocation (46,XY,+13,der(13;13)q10;q10).
There were seven unclassified samples in the chromosome 21 analysis (1.4%), five in the chromosome 18 analysis (1.0%), and two in the chromosome 13 analysis (0.4%) (Fig. 3A–C). In all categories there was an overlap of three samples that had both a censored karyotype (69,XXX) and no fetal DNA detected. One unclassified sample in the chromosome 21 analysis was correctly identified as T13 in the chromosome 13 analysis and one unclassified sample in the chromosome 18 analysis was correctly identified as T21 in the chromosome 21 analysis.
The sex chromosome analysis population for determining performance of massively parallel sequencing (female, male, or monosomy X) was 433. Our refined algorithm for classifying the sex status, which allowed for accurate determination of sex chromosome aneuploidies, resulted in a higher number of massively parallel sequencing unclassified results. Sensitivity and specificity for detecting diploid female state (XX) were 99.6% (95% CI 97.6 to more than 99.9) and 99.5% (95% CI 97.2 to more than 99.9), respectively; sensitivity and specificity to detect male (XY) were both 100% (95% CI 98.0–100.0); and sensitivity and specificity for detecting monosomy X (45,X) were 93.8% (95% CI 69.8–99.8) and 99.8% (95% CI 98.7 to more than 99.9) (Fig. 3D–F). Although censored from the analysis (as per protocol), the massively parallel sequencing classifications of mosaic monosomy X karyotypes were as follows (Table 2): two of seven classified as monosomy X, three of seven with a Y chromosome component classified as XY, and two of seven with an XX chromosome component classified as female. Two samples that were classified by massively parallel sequencing as monosomy X had karyotypes of 47,XXX and 46,XX. Eight of ten sex chromosome aneuploidies for karyotypes 47,XXX, 47,XXY and 47,XYY were correctly classified (Table 2). If we had limited the sex chromosome classifications to monosomy X, XY and XX, most of the unclassified samples would have been correctly classified as male, but we would not have identified the XXY and XYY sex aneuploidies.
In addition to accurately classifying trisomies 21, 18, 13 and sex, the sequencing results also prospectively correctly classified aneuploidy for chromosomes 16 and 20 in two samples (47,XX,+16 and 47,XX,+20) (Table 3). Chromosome 22 aneuploidy was not detected by massively parallel sequencing in two other patients with trisomy 22 (one of these had no fetal DNA detected). Interestingly, one sample with a complex karyotype involving the long arm of chromosome 6 (6q) and two duplications, one of which was 37.5Mb in size, showed an increased normalized chromosome value from sequencing tags in chromosome 6 (normalized chromosome value=3.6). Two prior articles have shown that detection of partial chromosome deletion is feasible.11,18 In another sample, aneuploidy of chromosome 2 was detected by massively parallel sequencing but not observed in the fetal karyotype at amniocentesis (46,XY). Other complex karyotype variants shown in Tables 2 and 3 include samples from fetuses with chromosome inversions, deletions, translocations, triploidy, and other abnormalities that were not detected here but potentially could be classified by massively parallel sequencing at higher sequencing density or with further algorithm optimization or both. In these cases, massively parallel sequencing correctly classified the samples as unaffected for trisomy 21, 18, or 13 and as male or female.
We also examined specific clinical variables that might affect massively parallel sequencing performance. In this study, 38 of 532 analyzed samples were from women who underwent assisted reproduction. Of these 38, 17 had chromosomal abnormalities; no false-positive or false-negative results were detected in this subpopulation.
This prospective study to determine the capability of massively parallel sequencing to detect whole chromosome fetal aneuploidy from maternal plasma was designed to emulate the real world scenario of sample collection, processing and analysis. Whole blood samples were obtained at the enrollment sites, did not require immediate processing, and were shipped overnight to the sequencing laboratory. In contrast to a prior prospective study that only involved chromosome 21,15 in this study, all eligible samples with any abnormal karyotype were sequenced. The sequencing laboratory did not have prior knowledge of which fetal chromosomes might be affected nor the ratio of aneuploid to euploid samples. The study design recruited a high-risk study population of pregnant women to assure a statistically significant prevalence of aneuploidy, and Tables 2 and 3 indicate the complexity of the karyotypes that were analyzed. The results demonstrate that: 1) fetal aneuploidies (including those resulting from translocation trisomy, mosaicism, and complex variations) can be detected with high sensitivity and specificity and 2) aneuploidy in one chromosome does not affect the ability of the massively parallel sequencing method to correctly identify the euploid status of other chromosomes. The algorithms used in the previous studies appear to be unable to effectively determine other aneuploidies that inevitably would be present in a general clinical population.10,14
With regard to mosaicism, the massively parallel sequencing analysis in this study was able to classify samples as aneuploidy that had mosaic karyotypes for chromosomes 21 and 18 in four of four affected samples. These results demonstrate the sensitivity of the massively parallel sequencing analysis for detecting specific characteristics of cell free DNA in a complex mixture. In one case, the massively parallel sequencing data for chromosome 2 indicated a whole or partial chromosome aneuploidy whereas the amniocentesis karyotype result for chromosome 2 was diploid. In two other examples, one sample with 47,XXX karyotype and another with a 46,XX karyotype, massively parallel sequencing classified these samples as monosomy X. It is possible these are mosaic cases, or that the pregnant woman herself is mosaic. (It is important to remember that the sequencing is performed on total DNA, which is a combination of maternal and fetal DNA.) Although cytogenetic analysis of amniocytes or villi from invasive procedures is currently the reference standard for aneuploidy classification, a karyotype performed on a limited number of cells cannot rule out low-level mosaicism. The current clinical study design did not include long-term infant follow-up or access to placental tissue at delivery, so we are unable to determine whether these were true-positive or false-positive results. We speculate that the specificity of the sequencing process, coupled with optimized algorithms to detect genome wide variation, may ultimately provide more sensitive identification of fetal DNA abnormalities, particularly in cases of mosaicism, than standard karyotyping.
The International Society for Prenatal Diagnosis has issued a Rapid Response Statement commenting on the commercial availability of massively parallel sequencing for prenatal detection of Down syndrome.19 They state that before routine massively parallel sequencing-based population screening for fetal Down syndrome is introduced, evidence is needed that the test performs in some subpopulations, such as in women who conceive by in vitro fertilization. The results reported here suggest that massively parallel sequencing is accurate in this group of pregnant women, many of whom are at high risk for aneuploidy.
Although these results demonstrate the excellent performance of massively parallel sequencing with optimized algorithms for aneuploidy detection across the genome in singleton pregnancies from women at increased risk for aneuploidy, more experience, particularly in low-risk populations, is needed to build confidence in the diagnostic performance of the method when the prevalence is low and in multiple gestation. In the early stages of clinical implementation, massively parallel sequencing for chromosomes 21, 18, and 13 should be used after a positive first-trimester or second-trimester screening result. This will reduce unnecessary invasive procedures caused by the false-positive screening results, with a concomitant reduction in procedure related adverse events. Invasive procedures could be limited to confirmation of a positive result from sequencing. We acknowledge, however, that there are certain clinical scenarios (eg, advanced maternal age and infertility) in which pregnant women will want to avoid an invasive procedure; they may request this test as an alternative to the primary screen or invasive procedure or both. All patients should receive thorough pretest counseling to ensure that they understand the limitations of the test and the implications of the results. As experience accumulates with more samples, it is possible that this test will replace current screening protocols and become a primary screening and ultimately a noninvasive diagnostic test for fetal aneuploidy.
1. Milunsky A, Milunsky JM. Genetic disorders and the fetus: diagnosis, prevention, and treatment. 6th ed. Hoboken (NJ): Wiley-Blackwell; 2010.
2. Malone FD, Canick JA, Ball RH, Nyberg DA, Comstock CH, Bukowski R, et al.. First-trimester or second-trimester screening, or both, for Down's syndrome. N Engl J Med 2005;353:2001–11.
3. Fan HC, Blumenfeld YJ, Chitkara U, Hudgins L, Quake SR. Noninvasive diagnosis of fetal aneuploidy by shotgun sequencing DNA from maternal blood. Proc Natl Acad Sci U S A 2008;105:16266–71.
4. Fan HC, Quake SR. Sensitivity of noninvasive prenatal detection of fetal aneuploidy from maternal plasma using shotgun sequencing is limited only by counting statistics. PLoS One 2010;5:e10439.
5. Fan HC, Blumenfeld YJ, Chitkara U, Hudgins L, Quake SR. Analysis of the size distributions of fetal and maternal cell-free DNA by paired-end sequencing. Clin Chem 2010;56:1279–86.
6. Chiu RW, Chan KC, Gao Y, Lau VY, Zheng W, Leung TY, et al.. Noninvasive prenatal diagnosis of fetal chromosomal aneuploidy by massively parallel genomic sequencing of DNA in maternal plasma. Proc Natl Acad Sci U S A 2008;105:20458–63.
7. Chiu RW, Cantor CR, Lo YM. Non-invasive prenatal diagnosis by single molecule counting technologies. Trends Genet 2009;25:324–31.
8. Lo YM, Chan KC, Sun H, Chen EZ, Jiang P, Lun FM, et al.. Maternal plasma DNA sequencing reveals the genome-wide genetic and mutational profile of the fetus. Sci Transl Med 2010;2:61ra91.
9. Chiu RW, Lo YM. Non-invasive prenatal diagnosis by fetal nucleic acid analysis in maternal plasma: the coming of age. Semin Fetal Neonatal Med 2011;16:88–93.
10. Ehrich M, Deciu C, Zwiefelhofer T, Tynan JA, Cagasan L, Tim R, et al.. Noninvasive detection of fetal trisomy 21 by sequencing of DNA in maternal blood: a study in a clinical setting. Am J Obstet Gynecol 2011;204:205.e1–11.
11. Sehnert AJ, Rhees B, Comstock D, de Feo E, Heilek G, Burke J, et al.. Optimal detection of fetal chromosomal abnormalities by massively parallel DNA sequencing of cell-free fetal DNA from maternal blood. Clin Chem 2011;57:1042–9.
12. Chen EZ, Chiu RW, Sun H, Akolekar R, Chan KC, Leung TY, et al.. Noninvasive prenatal diagnosis of fetal trisomy 18 and trisomy 13 by maternal plasma DNA sequencing. PLoS One 2011;6:e21791.
13. Lo D, Chiu RW. Plasma nucleic acid analysis by massively parallel sequencing: pathological insights and diagnostic implications. J Pathol 2011;225:318–23.
14. Chiu RW, Akolekar R, Zheng YW, Leung TY, Sun H, Chan KC, et al.. Non-invasive prenatal assessment of trisomy 21 by multiplexed maternal plasma DNA sequencing: large scale validity study. BMJ 2011;342:c7401.
15. Palomaki GE, Kloza EM, Lambert-Messerlian GM, Haddow JE, Neveux LM, Ehrich M, et al.. DNA sequencing of maternal plasma to detect Down syndrome: An international clinical validation study. Genet Med 2011:913–20.
16. Kidd KK, Pakstis AJ, Speed WC, Grigorenko EL, Kajuna SL, Karoma NJ, et al.. Developing a SNP panel for forensic identification of individuals. Forensic Sci Int 2006;164:20–32.
17. Clopper C, Pearson ES. The use of confidence or fiducial limits illustrated in the case of the binomial. Biometrika 1934;26:404–13.
18. Peters D, Chu T, Yatsenko SA, Hendrix N, Hogge WA, Surti U, et al.. Noninvasive prenatal diagnosis of a fetal microdeletion syndrome. N Engl J Med 2011;365:1847–8.
19. Benn P, Borrell A, Cuckle H, Dugoff L, Gross S, Johnson JA, et al.. Prenatal detection of Down syndrome using Massively Parallel Sequencing (MPS): a rapid response statement from a committee on behalf of the Board of the International Society for Prenatal Diagnosis, 24 October 2011. Prenat Diagn 2012 [epub ahead of print].
Supplemental Digital Content
© 2012 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.