Very sensitive real-time PCR technology can allow identification of only a few copies of target DNA in a sample. However, these extremely sensitive techniques have the disadvantage of the ease of contamination at sample collection or within the laboratory and great care must be taken to exclude this by adequate controls. Originally, this approach was investigated as a possible tumor marker in the management of malignant disease, especially looking for residual or recurrent disease2. The identification of Y signals in pregnancies with male fetuses was the first indication of the possibility of this approach in pregnancy. Real-time quantitative PCR, utilizing TaqMan® chemistry has become the method of choice for the detection of fetal alleles in maternal plasma. Since maternal plasma contains a large excess of free, maternal-derived DNA, this technology can only detect paternal inherited alleles. Unfortunately, the maternal DNA excess also limits the detection of fetal alleles caused by single nucleotide polymorphisms (SNPs), but the application of mass spectrometry3, and perhaps other PCR variants, may allow rapid analysis of fetal SNPs using maternal plasma. Recent efforts have focused on exploring the possibility of selective enrichment of fetal DNA, based on the observation that free fetal DNA is fragmented to a greater degree than free maternal DNA4, 5. If such an objective is achieved, this will offer the exploitation of a wide range of diagnostic tests, including perhaps even SNP detection on microarrays, and quantitative polymerase chain reaction (Q-PCR)6 or Multiplex Ligation-dependent Probe Amplification7 (MLPA)-based detection of aneuploidy. Such developments will expand significantly the portfolio of diseases that can be diagnosed using this technique. Another enrichment approach to reduce the maternal-derived free DNA has been by the treatment of maternal blood samples with formaldehyde8, 9. These methods increase the proportion of fetal to maternal-derived DNA in these samples, which is of obvious advantage for diagnostics. There have been many hypotheses to explain the presence of these molecules in maternal plasma, but the issue is not completely resolved as yet. There is a significant amount of information supporting the view that the placenta is a major source of free fetal DNA10. The rapid clearance of free fetal nucleic acids from the maternal circulation after delivery of the placenta11, the presence of fetal mRNA of placenta-expressed genes in the blood of pregnant women12, 13, the structural characteristics of the placenta including the physical contact with maternal tissues14 and the increase after placental laser ablation in twin–twin transfusion syndrome10 are examples of this evidence. Most likely, the death of placental cells releases DNA into the maternal circulation directly. Another possible mechanism is that the placenta is a conduit of DNA circulating in fetal blood by transfer to the mother. Indeed, placental transfer of DNA is bidirectional but cell-free fetal DNA concentration in maternal plasma is much higher than maternal DNA concentration in umbilical plasma at delivery, which indicates that cell-free DNA is unequally transferred in the direction of fetus to mother14. An alternative hypothesis for the origin of free fetal nucleic acid in maternal blood is liberation of cell-free DNA into the maternal circulation following destruction of fetal hematopoietic cells that have crossed into the maternal circulation15, 16. Indeed, there is extensive apoptosis of fetal cells in maternal plasma17 or in the umbilical cord blood18 suggesting that this is likely to account for some fetal DNA detected in the pregnant woman even if not all16. However, evidence that the source is not only from the fetal circulation or placental transfer of fetal cells is that several groups, including ours, have reliably detected free fetal DNA 35 days postconception and in some cases as early as 14 days in maternal blood19. Since this is before the fetomaternal circulation is established, these observations support a trophoblast origin16. A final hypothesis suggests the direct transfer of DNA molecules to the amniotic fluid originating from different fetal tissues (and the levels are high), which could lead to the direct transfer of DNA molecules across placenta or membranes to the maternal circulation by diffusion16. On balance we think it is likely that the main source will be apoptosis within placental cells but perhaps with a modest contribution from fetal circulation or cells too. Since the placenta seems to be the principal source of this material, it is very likely that free fetal DNA will be a useful marker for placental disease such as in fetal growth restriction and pre-eclampsia. Prolonged persistence of fetal progenitor cells in maternal blood and bone marrow can occur20. Also, a single study suggested the presence of Y chromosome-specific DNA in women with male offspring some years after delivery21, which raised the concern that cell-free fetal DNA might persist in the maternal plasma from previous pregnancies. However, several groups investigated the clearance of circulating fetal DNA after delivery11, 22-25 and they all found that the DNA signal under investigation was very rapidly eliminated after delivery. Moreover, fetal rhesus status and sex determination have been performed for clinical management using amplification by real-time PCR of fetal DNA sequences present in maternal circulation with no reported false-positive results related to persistent fetal DNA from a previous pregnancy24, 26, 27. Therefore, persistence of free DNA from a previous pregnancy does not seem to be a problem and non-invasive prenatal diagnosis using free fetal DNA is reliable. The data also show that since the half-life of free DNA is very short, there must be a continuous source of the material in pregnancy. Some reports have documented detection of Y-signal in the urine of pregnant women with a male fetus28-30 leading to the possibility of using maternal urine as a source of fetal DNA for diagnostic purposes. However, there have been several reports where male-specific sequences were not detected despite the use of very sensitive nested PCR assay nor a highly reproducible real-time PCR31, 32. Moreover, even under pathological conditions known to increase kidney permeability, such as pre-eclampsia and the HELLP syndrome, the detection of free fetal DNA was negative32. We conducted a study with the aim of clarifying this issue and we were unable to find Y chromosome DNA in any of 20 samples of maternal urine, including 12/20 samples in which Y chromosome DNA was detected in maternal plasma that were confirmed as male karyotype33. In summary, the data suggest that cell-free fetal DNA is not easily detectable in maternal urine, at least with the laboratory techniques that we have today. Another surprising finding was the detection of cell-free fetal RNA in maternal plasma. It surprised many because, if a laboratory sample is ‘spiked’ with RNA material, the signal disappears very quickly indeed. However, samples taken from pregnant women have RNA signals that are stable. This is probably because of membrane binding or even vesiculation of RNA, perhaps during the apoptosis process34. Poon et al., using a two-step reverse transcription (RT)-PCR assay, demonstrated for the first time the presence of fetal-derived, male-specific mRNA in the plasma of pregnant women with male fetuses12. As with DNA, the placenta seems to be the most important source of fetal mRNA because a signal is found for placenta-expressed genes such as corticotropin-releasing hormone (CRH), and there is a correlation with mRNA and protein levels in maternal blood13. These results suggest that circulating mRNA could be very useful for research into obstetric complications and pregnancy physiology and not only limited to detection of fetal genetic polymorphisms. Circulating mRNA seems to provide information regarding gene expression patterns in the placenta, which are altered in some pathologies. Ng et al. have demonstrated that the concentration of maternal plasma CRH mRNA is increased in pregnancies complicated with pre-eclampsia when compared with normal pregnancies35 and may also be related to the severity of the disease36. These results have encouraged the development of microarray technology, which allows the systematic identification of a large panel of fetal specific mRNA markers in maternal plasma37 that directly reflect the gene expression patterns in the placenta. This has particular implications for the generation of new markers for the studying and monitoring of pregnancy disorders related to placental pathology34, 37. Hemolytic disease of the fetus and newborn (HDFN) due to maternal alloantibodies directed against paternally inherited antigens on fetal red cells, was a significant cause of fetal and neonatal morbidity and mortality until the introduction of anti-D immunoglobulin during pregnancy and shortly after delivery38, 39. However, in England and Wales, about 500 fetuses develop HDFN each year, of which about 25–30 die40. The fetuses at risk are those with a D + phenotype, and so prenatal determination of fetal D blood group for management of pregnancies at risk for HDFN is important41. The identification of blood group genes and subsequent detection of the molecular bases of blood group polymorphisms has made it possible to predict blood group phenotypes from amniotic fluid42. The source of DNA used to predict fetal blood group was initially amniotic fluid or chorionic villi26. However, the related risk of the obstetric procedures is 0.5–1% for fetal loss43; amniocentesis carries a 17% risk of fetomaternal hemorrhage44, which is associated with increased incidence of gestational maternal immunization45. The demonstration that cell-free fetal DNA is present in the plasma of pregnant women in sufficient quantities for determination of fetal RhD genotype46, leads to the possibility of fetal D typing using a non-invasive approach. Combination of the sensitivity of real-time PCR with improved rhesus D (RhD) typing assay enables a highly accurate prediction of fetal D status from maternal plasma26, and this is now available as a service world-wide26, 41, 47. This new approach has significantly reduced the number of invasive procedures carried out in our unit for fetal D grouping47. The future possibilities for the use of circulating nucleic acids for fetal blood grouping include undertaking this test in all rhesus negative pregnant women in order to administer anti-D only to those women with a D-positive fetus. Also, the possibilities for testing for other blood group antigens, such as Kell, Rhc, RhE, or other alloimmune diseases such as neonatal alloimmune thrombocytopenia, are being explored48. Since the first identification of Y-chromosome sequences as part of the fetal cell-free DNA pool in maternal plasma1, the technique has proved reproducible in many laboratories2. When maternal blood is used to determine fetal sex, the rate of successful recognition of fetal gender is nearly 100%, even in the first trimester2, 27, 49. In our study, we made a correct diagnosis of gender in 25 of 26 cases, with one false-positive male result19. These results are having a huge impact in clinical practice for the management of X-linked genetic disorders and the technique has even been used to guide fetal therapy in prenatal cases at risk of 21-hydroxylase deficiency49-51. It is already possible to consider only undertaking an invasive test such as chorionic villus sampling (CVS) in order to show if a male is affected or not by the disease when the fetus is shown to be male by free fetal DNA. This means CVS can be avoided when the fetus is female. Another advantage is the ability to determine fetal sex early in pregnancy. CVS is usually not undertaken before 11 weeks' gestation, whereas fetal gender determination from maternal plasma is reliable from 7 weeks' gestation but male signals can be detected as early as 2 weeks postconception19. The diagnosis of fetal gender early in the first trimester may reduce anxiety for parents who are carriers of an X-linked condition or avoid unnecessary exposure to steroids in pregnancies at risk of 21-hydroxylase deficiency50. There is increasing evidence that maternal plasma circulating nucleic acids can also be useful for the non-invasive prenatal diagnosis of certain genetic disorders. The diagnoses of autosomal dominant disorders such as myotonic dystrophy52 or Huntington disease53, or single gene mutation as in achondroplasia54, are examples of the wide range of problems where the use of free fetal DNA could or has been employed. In cases of autosomal recessive disorders such as cystic fibrosis (CF), a serious and common genetic disorder in the Caucasian population, prenatal diagnosis currently requires CVS. When the parents are discordant for a mutation the diagnosis of an unaffected fetus may be made by exclusion of a paternally inherited cystic fibrosis transmembrane conductance regulator (CFTR) gene. In that way, invasive prenatal diagnosis could be limited to those pregnancies which have inherited the paternal mutation and so have a potentially affected fetus55. Similarly, another common autosomal recessive single-gene disorder is β-thalassemia. Chiu et al. designed allele-specific primers and a fluorescent probe for the detection of one of the mutations in the β-globin gene from maternal plasma by real-time PCR56 and they were able to exclude fetal inheritance of this specific paternally transmitted mutation. The application of Q-PCR to the detection of placentally derived mRNAs has been suggested for the diagnosis of aneuploidy. Oudejans et al.57 described detecting various mRNAs in maternal plasma, all encoded by chromosome 21. Thus, using suitable housekeeping controls, quantification of these mRNAs compared to other non-chromosome 21-encoded, placentally derived mRNAs may provide an opportunity to diagnose the presence of trisomy 21 pregnancies. These studies clearly require expanding, especially considering that other non-chromosome 21-encoded mRNAs may be altered in trisomy 21 placental tissue. An increase in fetal DNA levels in maternal plasma during the last 8 weeks of pregnancy has been demonstrated58. Indeed, any pathological process that disturbs the placenta, such as pre-eclampsia, fetal growth restriction or fetomaternal hemorrhage seems to be associated with increased apoptosis of trophoblast cells59, 60. This increased level of apoptosis could explain the raised cell-free fetal DNA levels in maternal blood observed in pre-eclamptic patients61, 62 or in cases of fetal growth restriction63, and these measurements may serve as a marker for placental abnormalities. Quantification of fetal DNA may also be helpful in predicting pre-eclampsia64 because its concentration in maternal plasma is increased prior to the onset of the disease and it could be associated with an eightfold increased risk65. A two-phase elevation has been observed in women who will develop pre-eclampsia: the first occurs at 13 to 17 weeks' gestation and a second about 3 weeks before the onset of symptoms. This may be due to increased liberation from placenta or impairment in kidney function later in the disease66. The same concept may apply to the prediction of cases of fetal growth restriction with reported correlations between increasing fetal DNA values in fetoplacental hypoxia as a result of placental dysfunction or high altitude63, 67, 68. In summary, non-invasive prenatal diagnosis has already become part of patient care for fetal blood grouping and in pregnancies at risk of a sex-linked disease. There are extensive other possible applications to ultrasound in obstetrics and gynecology such as for single gene disease screening in common problems such as echogenic bowel (cystic fibrosis) or perhaps investigation of a genetic cause of short limbs on ultrasound. Major studies on both sides of the Atlantic (such as the FP6 EU Network ‘Special Advances in Fetal Evaluation’ (SAFE)69) are underway. We think this will become a major source of material for fetal genetic testing following ultrasound diagnosis of fetal abnormalities or pregnancy complications in the future.
No takes yet. Share an insight, caveat, or question.
Illanes et al. (2005) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: