Prenatal Testing and Reproductive Genetics

Published on 16/03/2015 by admin

Filed under Basic Science

Last modified 22/04/2025

Print this page

rate 1 star rate 2 star rate 3 star rate 4 star rate 5 star
Your rating: none, Average: 0 (0 votes)

This article have been viewed 2686 times

CHAPTER 21 Prenatal Testing and Reproductive Genetics

Until recently, couples at high risk of having a child with a genetic disorder had to choose between taking the risk or considering other reproductive options, such as long-term contraception, sterilization, and termination of pregnancy. Other alternatives included adoption, long-term fostering, and donor insemination (DI).

Over the past three decades, prenatal diagnosis—the ability to detect abnormalities in an unborn child—has been widely used. Although it may be very difficult for a couple to decide to pursue prenatal diagnosis because of the possibility that this will lead to termination of pregnancy, prenatal diagnosis is an option that is chosen by many couples at high risk of having a child with a serious genetic disorder or birth defect.

The ethical issues surrounding prenatal diagnosis and selective termination of pregnancy are both complex and emotive, and are considered more fully in Chapter 24 (p. 363). In this chapter, we focus on the practical aspects of prenatal testing and diagnosis, including prenatal screening, as well as some aspects of reproductive genetics.

Techniques Used in Prenatal Diagnosis

There are several techniques that can be used for the prenatal diagnosis of hereditary disorders and structural abnormalities (Table 21.1).

Table 21.1 Standard Techniques Used in Prenatal Diagnosis

Technique Optimal Time (Weeks) Disorders Diagnosed
Non-Invasive
Maternal serum screening
α-Fetoprotein 16 Neural tube defects
Triple test 16 Down syndrome
Ultrasound 18 Structural abnormalities (e.g., central nervous system, heart, kidneys, limbs)
Invasive
Amniocentesis 16  
Fluid   Neural tube defects
Cells   Chromosome abnormalities, metabolic disorders, molecular defects
Chorionic villus sampling 10–12 Chromosome abnormalities, metabolic disorders, molecular defects
Fetoscopy
Blood (cordocentesis)   Chromosome abnormalities, hematological disorders, congenital infection
Liver   Metabolic disorders (e.g., ornithine transcarbamylase deficiency)
Skin   Hereditary skin disorders (e.g., epidermolysis bullosa)

Amniocentesis

Amniocentesis involves the aspiration of 10 to 20 ml of amniotic fluid through the abdominal wall under ultrasonographic guidance (Figure 21.1). This is usually performed around the 16th week of gestation. The sample is spun down to yield a pellet of cells and supernatant fluid. The fluid can be used in the prenatal diagnosis of neural tube defects by assay of α-fetoprotein (p. 328). The cell pellet is resuspended in culture medium with fetal calf serum, which stimulates cell growth. Most of these cells in the amniotic fluid, that have been shed from the amnion, fetal skin, and urinary tract epithelium, are non-viable, but a small proportion will grow. After approximately 14 days, there are usually sufficient cells for chromosome and DNA analysis, although a longer period may be required before enough cells are obtained for biochemical assays. Increasingly, sensitive polymerase chain reaction (PCR) techniques make direct DNA analysis possible without the need for culture.

When a couple is considering amniocentesis, they should be informed of the 0.5% to 1% risk of miscarriage associated with the procedure, and that if the result is abnormal they will be facing the possibility of having to consider a mid-trimester termination of pregnancy that involves an induction of labor.

Trials of amniocentesis earlier in pregnancy, at 12 to 14 weeks’ gestation, yielded comparable rates of success in obtaining results with a similar risk of miscarriage. However, concerns have been expressed regarding the reduction in amniotic fluid at this early stage of pregnancy, and early amniocentesis is not widely practiced. Although it has the advantage of allowing a result to be given earlier in the pregnancy, a mid-trimester termination of pregnancy is still usually required if the fetus is found to be affected.

Chorionic Villus Sampling

In contrast to amniocentesis, chorionic villus sampling (CVS), first developed in China, enables prenatal diagnosis to be undertaken during the first trimester. This procedure is usually carried out at 11 to 12 weeks’ gestation under ultrasonographic guidance by either transcervical or, more usually, transabdominal aspiration of chorionic villus (CV) tissue (Figure 21.2). This tissue is fetal in origin, being derived from the outer cell layer of the blastocyst (i.e., the trophoblast). Maternal decidua, normally present in the biopsy sample, must be removed before the sample is analyzed. Placental biopsy is the term used when the procedure is carried out at later stages of pregnancy.

Chromosome analysis can be undertaken on CV tissue either directly, looking at metaphase spreads from actively dividing cells, or after culture. Direct chromosomal analysis of CV tissue usually allows a provisional result to be given within 24 hours. Nowadays rapid, direct fluorescent in-situ hybridization (FISH) probing (p. 34), or DNA analysis by the multiplex ligation-dependent probe amplification technique (p. 66), is used to test for common chromosome aneuploidies prior to a standard karyotype following culture of CV tissue; this may also detect other chromosome abnormalities and balanced rearrangements. For single-gene disorders, sufficient CV tissue is usually obtained to allow prenatal diagnosis by immediate biochemical assay or DNA analysis using uncultured CV tissue.

The major advantage of CV sampling is that it offers first-trimester prenatal diagnosis, although it has the disadvantage that even in experienced hands the procedure conveys a 1% to 2% risk of causing miscarriage. There is also evidence that this technique can cause limb abnormalities in the embryo if carried out before 9 to 10 weeks’ gestation; for this reason is not performed before 11 weeks’ gestation.

Ultrasonography

Ultrasonography offers a valuable means of prenatal diagnosis. It can be used not only for obstetric indications, such as placental localization and the diagnosis of multiple pregnancies, but also for prenatal diagnosis of structural abnormalities not associated with known chromosomal, biochemical or molecular defects. Ultrasonography is particularly valuable because it is non-invasive and conveys no known risk to the fetus or mother. It does, however, require expensive equipment and a skilled, experienced operator. For example, a search can be made for polydactyly as a diagnostic feature of a multiple abnormality syndrome, such as one of the autosomal recessive short-limb polydactyly syndromes that are associated with severe pulmonary hypoplasia—invariably lethal (Figure 21.3). Similarly, a scan can reveal that the fetus has a small jaw, which can be associated with a posterior cleft palate and other more serious abnormalities in several single-gene syndromes (Figure 21.4).

Until a few years ago, detailed ultrasonography for structural abnormalities was offered only to couples who had a child with a genetic disorder or syndrome for which there was no chromosomal, biochemical, or molecular marker. Increasingly, however, detailed ‘fetal anomaly’ scanning is being offered routinely to all pregnant women at around 18 weeks’ gestation as screening for structural abnormalities such as neural tube defects or cardiac anomalies. This technique can also identify features that suggest the presence of an underlying chromosomal abnormality. Such a finding would lead to an offer of amniocentesis or placental biopsy for definitive chromosome analysis.

The future of fetal scanning holds the prospect of three-dimensional imaging and magnetic resonance imaging being used more widely and routinely. Although this will clearly enable the unborn baby to be visualized in far greater detail, it will also generate bigger challenges for the dysmorphologist, who might be expected to diagnose serious disorders on the basis of very subtle features.

Prenatal Screening

The history of widespread prenatal (antenatal) screening really began with the finding, in the early 1970s, of an association between raised maternal serum α-fetoprotein (AFP) and neural tube defects (NTDs). Estimation of AFP levels was gradually introduced into clinical service, and the next significant development was ultrasonography, followed, in the 1980s, by the identification of maternal serum biochemical markers for Down syndrome. These are discussed in more detail below. Where the incidence of a genetic condition was high, for instance thalassemia in Cyprus, prenatal screening came into practice, as described in Chapter 20 (p. 321). However, molecular genetic advances, rather than biochemical, mean that the range of prenatal screening is continuing to evolve.

Testing for cystic fibrosis and fragile X syndrome are available in the UK, mainly for those willing to pay privately, and in Israel, for example, a wide range of relatively rare diseases can be screened for on the basis that they are more common in specific population groups that were originally isolates with multiple inbreeding, and therefore certain mutations are prevalent. Besides Tay-Sachs disease (carrier testing in this case is biochemical; see Chapter 20), familial dysautonomia, Canavan disease, Bloom syndrome, ataxia telangiectasia (North African Jews), limb-girdle muscular dystrophy (Libyan Jews) and Costeff syndrome (Iraqi Jews) are among the conditions for which screening is available. It does not come free of charge but the level of uptake of this screening is high, revealing the lengths to which some societies will go in order to avoid having children with serious genetic conditions. As DNA testing becomes more automated, rapid, and affordable, there will be pressure from some quarters to screen for many conditions, even though they are individually very rare. This challenge is already emerging with the potential use of microarray CGH in prenatal testing. If the use of microarray becomes routine there may be great difficulty in interpreting the consequences of rare or unique copy number variants. This ethical challenge is discussed more fully in Chapter 24.

Neural Tube Defects

In 1972 it was recognized that many pregnancies in which the baby had an open NTD (p. 258) could be detected at 16 weeks’ gestation by assay of AFP in maternal serum. AFP is the fetal equivalent of albumin and is the major protein in fetal blood. If the fetus has an open NTD, the level of AFP is raised in both the amniotic fluid and maternal serum as a result of leakage from the open defect. Open NTDs fulfil the criterion of being serious disorders, as anencephaly is invariably fatal, and between 80% and 90% of the small proportion of babies who survive with an open lumbosacral lesion are severely handicapped.

Unfortunately maternal serum AFP screening for NTDs is neither 100% sensitive nor 100% specific (p. 319). The curves for the levels of maternal serum AFP in normal and affected pregnancies overlap (Figure 21.6), so that in practice an arbitrary cut-off level has to be introduced below which no further action is taken. This is usually either the 95th centile, or 2.5 multiples of the median (MoM); as a result around 75% of screened open spina bifida cases are detected. Those pregnant women with results that lie above this arbitrary cut-off level are offered detailed ultrasonography; which is usually sufficient to diagnose NTD. In fact, ultrasonography has more or less superceded maternal serum screening as a means of diagnosing NTD. Anencephaly shows a dramatic deficiency in the cranium (Figure 21.7) and an open myelomeningocele is almost invariably associated with herniation of the cerebellar tonsils through the foramen magnum. This deforms the cerebellar hemispheres, which then have a curved appearance known as the ‘banana sign’; the forehead is also distorted, giving rise to a shape referred to as the ‘lemon sign’ (Figure 21.8). A posterior encephalocele is readily visualized as a sac in the occipital region (Figure 21.9) and always prompts a search for additional anomalies that might help diagnose a recognizable condition such as Meckel-Gruber syndrome.

A raised maternal serum AFP concentration is not specific for open NTDs (Box 21.1). Other causes include threatened miscarriage, twin pregnancy and a fetal abnormality such as exomphalos, in which there is a protrusion of abdominal contents through the umbilicus.

As a result of these screening modalities there has been a striking decline in the incidence of open NTDs in liveborn and stillborn babies. Other contributory factors are a general improvement in diet and the introduction of periconceptional folic acid supplementation (p. 258). In England and Wales the combined incidence of anencephaly and spina bifida in liveborn and stillborn babies fell from 1 in 250 in 1973 to 1 in 6250 in 1993.

Down Syndrome and Other Chromosome Abnormalities

The Triple Test

Confirmation of a chromosome abnormality in an unborn baby requires cytogenetic or molecular studies using material obtained by an invasive procedure such as CVS or amniocentesis (p. 325). However, chromosome abnormalities, and in particular Down syndrome, can be screened for in pregnancy by taking into account risk factors such as maternal age and the levels of three biochemical markers in maternal serum (Table 21.2).

Table 21.2 Maternal Risk Factors for Down Syndrome

Advanced age (35 years or older)
Maternal serum
MoM*
α-Fetoprotein (0.75)
Unconjugated estriol (0.73)
Human chorionic gonadotrophin (2.05)
Inhibin-A (2.10)

* Values in parentheses refer to the mean values in affected pregnancies, expressed as multiples of the median (MoMs) in normal pregnancies.

This latter approach is based on the discovery that, at 16 weeks’ gestation, maternal serum AFP and unconjugated estriol levels tend to be lower in Down syndrome pregnancies than in normal pregnancies, whereas the level of maternal serum human chorionic gonadotropin (hCG) is usually raised. None of these parameters gives absolute discrimination, but taken together they provide a means of modifying a woman’s prior age-related risk to give an overall probability that the unborn baby is affected. When this probability exceeds 1 in 250, invasive testing in the form of amniocentesis or placental biopsy is offered.

Using age alone as a screening parameter, if all pregnant women aged 35 years and over opt for fetal chromosome analysis approximately 35% of all Down syndrome pregnancies will be detected (Table 21.3). If three biochemical markers are also included (this being the so-called triple test), 60% of all Down syndrome pregnancies will be detected when a risk of 1 in 250 or greater is the cut-off for offering amniocentesis. This approach will also result in the detection of approximately 50% of all cases of trisomy 18 (p. 275). In the latter condition all the biochemical parameters are low, including hCG.

Table 21.3 Detection Rates Using Different down Syndrome Screening Strategies

Screening Modality Percent of All Pregnancies Tested Percent of Down Syndrome Cases Detected
Age alone    
40 years and older 1.5 15
35 years and older 7 35
Age + AFP 5 34
Age + AFP, µE3 + hCG 5 61
Age + AFP, µE3, hCG + inhibin-A 5 75
NT alone 5 61
NT + age 5 69
hCG, AFP + age 5 73
NT + AFP, hCG + age 5 86

AFP, α-fetoprotein; µE3, unconjugated estriol; hCG, human chorionic gonadotrophin; NT, nuchal translucency.

It has recently been shown that another biochemical marker, inhibin-A, is also increased in maternal serum in Down syndrome pregnancies. If this fourth marker is used as part of a ‘quadruple’ serum screening test, the proportion of Down syndrome pregnancies detected rises from 60% to 75% when amniocentesis is offered to the 5% of mothers with the highest risk.

Published results from California provide a useful indication of the outcome of a triple-test prenatal screening program. In a population of 32 million, all pregnant women were offered the triple-test. This was accepted by 67% of all eligible women, of whom 2.6% went on to have amniocentesis, resulting in the detection of 41% of all cases of Down syndrome. These figures are similar to those observed in other studies and serve to illustrate the discrepancy between what is possible in theory (i.e., a detection rate of 60%) and what actually happens in practice.

Ultrasonography

Almost all pregnant women are routinely offered a ‘dating’ scan at around 12 weeks’ gestation. At around this time there is a strong association between chromosome abnormalities and the abnormal accumulation of fluid behind the baby’s neck—increased fetal nuchal translucency (NT) (see Figure 21.5). This applies to Down syndrome, the other autosomal trisomy syndromes (trisomies 13 and 18; p. 275), Turner syndrome, and triploidy, as well as a wide range of other fetal abnormalities and rare syndromes. The risk for Down syndrome correlates with absolute values of NT as well as maternal age (Figure 21.10) but, because NT also increases with gestational age, it is more usual now to relate the risk to the percentile value for any given gestational age. In one study, for example, 80% of Down syndrome fetuses had NT above the 95th percentile. By combining information on maternal age with the results of fetal NT thickness measurements, together with maternal serum markers, it is possible to detect more than 80% of fetuses with trisomy 21 if invasive testing is offered to the 5% of pregnant women with the highest risk (see Table 21.3). Some babies with Down syndrome have duodenal atresia, which shows up as a ‘double bubble sign’ on ultrasonography of the fetal abdomen (Figure 21.11).

In many centers, it is also standard practice to offer a detailed ‘fetal anomaly’ scan to all pregnant women at 18 weeks. Although chromosome abnormalities cannot be diagnosed directly, their presence can be suspected by the detection of an abnormality, such as exomphalos (Figure 21.12) or a rocker-bottom foot (Figure 21.13) (Table 21.4). A chromosome abnormality is found in 50% of fetuses with exomphalos identified at 18 weeks, and a rocker-bottom foot is a very characteristic, though not specific, finding in babies with trisomy 18 (p. 275), who are invariably growth retarded. The use of other ultrasonographic ‘soft markers’ in identifying chromosome abnormalities in pregnancy is discussed in the following section (p. 335).

image

FIGURE 21.12 Ultrasonogram at 18 weeks showing exomphalos.

(Courtesy Dr. D. Rose, City Hospital, Nottingham, UK.)

Table 21.4 Prenatal Ultrasonographic Findings Suggestive of a Chromosome Abnormality

Feature Chromosome Abnormality
Cardiac defect (especially common atrioventricular canal) Trisomy 13, 18, 21
Clenched overlapping fingers Trisomy 18
Cystic hygroma or fetal hydrops Trisomy 13, 18, 21
Duodenal atresia 45,X (Turner syndrome)
Trisomy 21
Exomphalos Trisomy 13, 18
Rocker-bottom foot Trisomy 18

Indications for Prenatal Diagnosis

There are numerous indications for offering prenatal diagnosis. Ideally, couples at high prior risk of having a baby with an abnormality should be identified and assessed before embarking on a pregnancy so that, in an unrushed manner, they can be counseled and come to a decision about which option they wish to pursue. Certain orthodox Jewish communities are extremely well organized in this respect vis-à-vis Tay-Sachs disease, as described in Chapter 20 (pp. 313–314). A less satisfactory alternative is that couples are identified early in pregnancy so that they still have an opportunity to consider prenatal diagnostic options. Unfortunately, many couples at increased risk because of their family history or previous reproductive history are still not referred until mid-pregnancy, when it may be too late to undertake the most thorough clinical and laboratory work-up in preparation for prenatal diagnosis.

Advanced Maternal Age

This has been the most common indication for offering prenatal diagnosis. There is a well-recognized association of advanced maternal age with increased risk of having a child with Down syndrome (see Table 18.4; p. 274) and the other autosomal trisomy syndromes. No standard criterion exists for determining at what age a mother should be offered the option of an invasive prenatal diagnostic procedure for fetal chromosome analysis. Most centers routinely offer amniocentesis or CVS to women age 37 years or older, and the option is often discussed with women from the age of 35 years onward. These risk figures relate to the maternal age at the expected date of delivery. The risk figures for Down syndrome at the time of CVS, amniocentesis, and delivery differ (see Figure 18.1; p. 274) because a proportion of pregnancies with trisomy 21 are lost spontaneously during the first and second trimesters. Interestingly, despite industrial-scale efforts to screen for Down syndrome, there has been a slight rise in the numbers of live births in the United Kingdom since 2000, following a steady decline from the widespread introduction in screening in 1989 (National Down Syndrome Cytogenetic Register). However, the numbers of prenatal diagnoses and terminations for Down syndrome has also increased over this period. Both observations are attributed to the slightly older age at which women are now having children, and there may also be an increasing willingness to raise a child with the condition.

Previous Child with a Chromosome Abnormality

Although there are a number of series with slightly different recurrence risk figures, for couples who have had a child with Down syndrome because of non-disjunction, or a de novo unbalanced Robertsonian translocation, the risk in a subsequent pregnancy is usually given as the mother’s age-related risk plus approximately 1%. If one of the parents has been found to carry a balanced chromosomal rearrangement, such as a chromosomal translocation (p. 44) or pericentric inversion (p. 48), that has caused a previous child to be born with serious problems due to an unbalanced chromosome abnormality, the recurrence risk is likely to be between 1% to 2% and 15% to 20%. The precise risk will depend on the nature of the parental rearrangement and the specific segments of the individual chromosomes involved (p. 48).

Family History of a Neural Tube Defect

Careful evaluation of the pedigree is necessary to determine the risk that applies to each pregnancy. Risks can be determined based on empiric data (p. 346). In high-risk situations, ultrasonographic examination of the fetus, possibly in conjunction with assay of maternal serum AFP, can be offered. However, even with good equipment and an experienced ultrasonographer, small closed NTDs can still be missed. Fortunately, the latter types of NTD are not usually associated with the serious problems seen with large open NTDs (p. 258).

Other High-Risk Factors

These factors include parental consanguinity, a poor obstetric history, and certain maternal illnesses. Parental consanguinity increases the risk that a child will have a hereditary disorder or congenital abnormality (pp. 113). Consequently, if the parents are concerned, it is appropriate to offer detailed ultrasonography to try to exclude a serious structural abnormality. It may also be appropriate to offer to test the couple for cystic fibrosis and spinal muscular atrophy carrier status, and possibly other conditions depending on ethnicity. A poor obstetric history, such as recurrent miscarriages or a previous unexplained stillbirth, could indicate an increased risk of problems in a future pregnancy and detailed ultrasonographic monitoring. A history of three or more unexplained miscarriages should be investigated by parental chromosome studies to exclude a chromosomal rearrangement such as a translocation or inversion (pp. 44, 48). Maternal illnesses, such as poorly controlled diabetes mellitus (p. 235) or epilepsy treated with anticonvulsant medications such as sodium valproate (p. 261), would also be indications for detailed ultrasonography. Both of these factors convey an increased risk of structural abnormality in a fetus.

Special Problems in Prenatal Diagnosis

The significance of the result of a prenatal diagnostic investigation is usually clear-cut, but situations can arise that pose major problems of interpretation. Problems also occur when the diagnostic investigation is unsuccessful or an unexpected result is obtained.

An Ambiguous Chromosome Result

In approximately 1% of cases, CVS shows evidence of apparent chromosome mosaicism—i.e., the presence of two or more cell lines with different chromosome constitutions (p. 50). This can occur for several reasons:

In the case of amniocentesis, in most laboratories it is routine for the sample to be split and for two or three separate cultures to be established. If a single abnormal cell is identified in only one culture, this is assumed to be a culture artifact, or what is termed level 1 mosaicism, or pseudomosaicism. If the mosaicism extends to two or more cells in two or more cultures this is taken as evidence of true mosaicism, or what is known as level 3 mosaicism. The most difficult situation to interpret is when mosaicism is present in two or more cells in only one culture, termed level 2 mosaicism. This is most likely to represent a culture artifact, but there is up to a 20% chance that the mosaicism is real and will be present in the fetus.

To resolve the uncertainty of chromosomal mosaicism in cultured CV tissue it may be necessary to proceed to amniocentesis. If the latter test yields a normal chromosomal result, then it is usually concluded that the earlier result was not a true indication of the fetal karyotype.

Counseling couples in this situation may be extremely difficult. If true mosaicism is confirmed, it is often impossible to predict the phenotypic outcome for the baby. An attempt can be made to resolve ambiguous findings by fetal blood sampling for urgent karyotype analysis, but this too is limited in terms of the information it yields about the phenotype. Whatever option the parents choose, it is important that tissue (blood, skin, or placenta) is obtained at the time of delivery, whether the couple elects to terminate or continue with the pregnancy, to resolve the significance of the prenatal findings.

An Unexpected Chromosome Result

Three different types of unexpected chromosome results may occur, each of which usually necessitates specialized detailed genetic counseling.

The Presence of a Marker Chromosome

A third difficult situation is the finding of a small additional chromosome known as a marker chromosome, that is, a small chromosomal fragment the specific identity of which cannot be determined by conventional cytogenetic techniques (p. 33). If this is found to be present in one of the parents, then it is unlikely to be of any significance to the fetus. If, on the other hand, it is a de novo finding, there is up to a 15% chance that the fetus will be phenotypically abnormal. The risk is lower when the marker chromosome contains satellite material (p. 17), or is made up largely of heterochromatin (p. 32), than when it does not have satellites and is mostly made up of euchromatin (p. 32). The availability of FISH (p. 34) means that the origin of the marker chromosome can often be determined more specifically, so that it is possible to give more precise prognostic information. The most common single abnormality of this kind is a marker chromosome 15.

Ultrasonographic ‘Soft’ Markers

Sophisticated ultrasonography has resulted in the identification of subtle anomalies in the fetus, the significance of which is not always clear. For example, choroid plexus cysts are sometimes seen in the developing cerebral ventricles in mid-trimester (Figure 21.14). Initially, it was thought that these were invariably associated with the fetus having trisomy 18 but in fact they occur frequently in normal fetuses, although if they are very large and do not disappear spontaneously they can be indicative of a chromosome abnormality.

Increased echogenicity of the fetal bowel (Figure 21.15) has been reported in association with cystic fibrosis—the prenatal equivalent of meconium ileus (p. 301). Initial reports suggested this finding could convey a risk as high as 10% for the fetus having cystic fibrosis, but it is now clear that this risk is probably no greater than 1% to 2%. Novel ultrasonographic findings of this kind are often called soft markers, and their interpretation must be approached cautiously in the effort to distinguish normal from abnormal variation.

Preimplantation Genetic Diagnosis

For many couples prenatal diagnosis on an established pregnancy, with a view to possible termination, is too difficult to contemplate. For some of these couples preimplantation genetic diagnosis (PGD) provides an acceptable alternative. The second largest group of PGD users are those with subfertility or infertility who wish to combine assisted reproduction with genetic testing of the early embryo. In the procedure, the female partner is given hormones to induce hyperovulation, and oocytes are then harvested transcervically, under sedation and ultrasonographic guidance. Motile sperm from a semen sample are added to the oocytes in culture (in vitro fertilization [IVF]—the same technique as developed for infertility) and incubated to allow fertilization to occur. If genetic analysis is to be undertaken on DNA from a single cell (blastomere) from the early embryo (blastocyst) at the eight-cell stage on the third day, fertilization is achieved using intracytoplasmic sperm injection (ICSI) of a single sperm to avoid the presence of extraneous sperm.

At the eight-cell stage, the early embryo is biopsied and one, or sometimes two, cells are removed for analysis. Whatever genetic analysis is undertaken, it is essential that this is a practical possibility on genomic material from a single cell. From the embryos tested, two that are both healthy and unaffected by the disorder from which they are at risk are reintroduced into the mother’s uterus. Implantation must then occur for a successful pregnancy and this is a major hurdle—the success rate for the procedure is only about 25% to 30% per cycle of treatment, even in the best centers. A variation of the technique is removal of the first, and often second, polar bodies from the unfertilized oocyte, which lie under the zona pellucida. Because the first polar body degenerates quite rapidly, analysis is necessary within 6 hours of retrieval. Analysis of polar bodies is an indirect method of genotyping because the oocyte and first polar body divide from each other during meiosis I and therefore contain different members of each pair of homologous chromosomes.

In the United Kingdom, centers must be licensed to practice PGD and are regulated by the Human Fertilization and Embryology Authority (HFEA), though this body is due to be abolished. In numerical terms, the impact of PGD has been small to date, but a wide and increasing range of genetic conditions has now been tested (Table 21.5). The most common referral reasons for single-gene disorders are cystic fibrosis, myotonic dystrophy, Huntington disease, β-thalassemia, spinal muscular atrophy, and fragile X syndrome. The technique for identifying normal and abnormal alleles in these conditions, and DNA linkage analysis where appropriate, is PCR (p. 56). Sex selection in the case of serious X-linked conditions is available where single-gene analysis is not possible. The biggest group of referrals for PGD, however, is chromosome abnormalities—reciprocal and robertsonian translocations in particular (pp. 45, 47). Genetic analysis in these cases uses FISH technology (p. 34) and substantial work has to be undertaken for the couple prior to treatment because of the unique nature of many translocations.

Table 21.5 Some of the Conditions for Which Preimplantation Genetic Diagnosis has been Used and is Available

Mode of Inheritance Disease
Autosomal dominant

Autosomal recessive Spinal muscular atrophy Tay-Sachs disease X-linked X-linked—sexing only DMD
Ornithine transcarbamylase deficiency
Incontinentia pigmenti
Other serious disorders Mitochondrial MELAS Chromosomal

MELAS, mitochondrial myopathy encephalopathy, lactic acidosis, stroke.

In recent years, PGD has occasionally been used not only to select embryos unaffected for the genetic disorder for which the pregnancy is at risk, but also to provide a human leukocyte antigen tissue-type match so that the new child can act as a bone marrow donor for an older sibling affected by, for example, Fanconi anemia. The ethical debate surrounding these so-called savior sibling cases is discussed further in Chapter 24.

A further development using micromanipulation methods has attracted a lot of attention. To circumvent the problem of genetic disease resulting from mutation in the mitochondrial genome, the nucleus of the oocyte from the genetic mother (who carried the mitochondrial mutation) was removed and inserted into a donor oocyte from which the nucleus had been removed. This is cell nuclear replacement technology, similar to that used in reproductive cloning experiments in animals (’Dolly’ the sheep; see p. 369). The resulting fertilization led to the headline that the fetus had three genetic parents. The technique has also been used in other situations where the oocytes are generally of poor quality, but its use is extremely limited.

Assisted Conception and Implications for Genetic Disease

In Vitro Fertilization

Many thousands of babies worldwide have been born by IVF over the past 30 years, when the technique was first successful. The indication for the treatment in most cases is subfertility, which now affects one in seven couples. In some Western countries, 1% to 3% of all births are the result of assisted reproductive technologies (ARTs). The cohort of offspring conceived in this way is therefore very large, and evidence is gathering that the risk of birth defects is increased by 30% to 40% compared with the general population conceived in the normal way and about 50% more children are likely to be small for gestational age (SGA). Specifically, a small increase in certain epigenetic conditions due to defective genomic imprinting (p. 121) has been observed—Beckwith-Wiedemann (p. 124) and Angelman (p. 123) syndromes, and ‘hypomethylation’ syndrome, though the possible mechanisms are unclear. In cases studied, loss of imprinting (LOI) was observed at the KCNQ1OT1 locus (see Figure 7.27; p. 125) in the case of Beckwith-Wiedemann syndrome, and at the SNRPN locus (Figure 7.23; p. 123) in the case of Angelman syndrome. No apparent imprinting differences explain the increase in SGA babies conceived by ICSI.

Epigenetic events around the time of fertilization and implantation are crucial for normal development (p. 103). If there is a definite increased risk of conditions from abnormal imprinting after ARTs, this may relate, in part, to the extended culture time of embryos, which has become a trend in infertility clinics. Instead of transferring cleavage-stage embryos, it is now more routine to transfer blastocysts, which allows the healthier looking embryos to be selected. However, in animal models it has been shown that in vitro culture affects the extent of imprinting, gene expression, and therefore the potential for normal development.

Donor Insemination

As a means of assisted conception to treat male infertility, or circumvent the risk of a genetic disease, donor insemination (DI) has been used since the 1950s. Only relatively recently, however, has awareness of medical genetic issues been incorporated into practice. Following the cases of children conceived by DI who were subsequently discovered to have balanced or unbalanced chromosome disorders, or in some cases cystic fibrosis (indicating that the sperm donor was a carrier for cystic fibrosis), screening of sperm donors for cystic fibrosis mutations and chromosome rearrangements has become routine practice in many countries. This was recommended only as recently as 2000 by the British Andrology Society. In the Netherlands, a donor whose sperm was used to father 18 offspring developed an autosomal dominant late-onset neurodegenerative disorder (one of the spinocerebellar ataxias), thus indicating that all 18 offspring were conceived at 50% risk. This led to a ruling that the sperm from one donor should be used no more than 10 times, as against 25 before this experience. In the United Kingdom, men older than age 40 years cannot be donors because of the small but increasing risk of new germline mutations arising in sperm with advancing paternal age.

Of course, it is not possible to screen the donor for all eventualities, but these cases have served to highlight the potential conflict between treating infertility (or genetic disease) by DI and maintaining a high level of concern for the welfare of the child conceived. More high profile in this respect is the ongoing debate about how much information DI children should be allowed about their genetic fathers, and the law varies across the world.

Naturally, all of these issues apply in an equivalent way to women who wish to be egg donors.

Assisted Conception and the Law

In the United States, no federal law exists to regulate the practice of assisted conception other than the requirement that outcomes of IVF and ICSI must be reported. In the United Kingdom, strict regulation operates through the HFEA based on the Human Fertilization and Embryology Act of 1990. The HFEA reports to the Secretary of State for Health, issues licences, and arranges inspections of registered centers. The different licences granted are for treatment (Box 21.2), storage (gametes and embryos), and research (on human embryos in vitro). A register of all treatment cycles, the children born by IVF, and the use of donated gametes, must be kept. The research permitted under licence covers treatment of infertility, increase in knowledge regarding birth defects, miscarriage, genetic testing in embryos, the development of the early embryo, and potential treatment of serious disease. At the time of writing it is not clear what arrangements will be put in place when the HFEA is abolished.

Non-Invasive Prenatal Diagnosis

At the turn of the 19th century, it was discovered that fetal cells reach the maternal circulation, but confirmation that cell-free DNA of fetal origin (placentally derived) is present in the plasma of pregnant women was not made until 1997. This fact has now been exploited in clinical practice as early as 6 to 7 weeks of pregnancy to determine fetal sex by detection of Y-chromosome DNA, as well as fetal Rhesus D gene. Early determination of fetal sex is clinically useful in a pregnancy at risk of an X-linked recessive disorder, and also in congenital adrenal hyperplasia (see the following section). The problem with analyzing cell-free fetal DNA is one of isolation because maternal cell-free DNA constitutes about 95% of all the cell-free DNA in the maternal circulation. The absence of Y-chromosome DNA might indicate the fetus is female, or that the quantity of fetal DNA is very low. This is resolved by using real-time PCR to quantify the amount of fetal or total DNA present in plasma.

Much effort is now focused on enriching fetal cells from maternal blood, which would make it possible to analyse the pure fetal genome because fetal red cells are nucleated; however, only about one fetal cell is present in 1 ml of maternal blood, and consequently current techniques are limited because of the scarcity of cells. If whole cells could be efficiently enriched, in theory fetal aneuploids could be detected using FISH, or other techniques based on allele ratios. Advances are taking place rapidly and non-invasive techniques are likely to become a reality in due course, and therefore change perceptions of prenatal diagnosis dramatically.

Prenatal Treatment

So far this chapter has focused on prenatal diagnosis and screening for abnormalities with the subsequent option of termination of pregnancy, as well as other techniques designed to prevent genetic disease. Although these are the only options in most situations, there is cautious optimism that prenatal diagnosis will, in time, lead to the possibility of effective treatment in utero, at least for some conditions.

A possible model for successful prenatal treatment is provided by the autosomal recessive disorder congenital adrenal hyperplasia (CAH) (p. 174). Affected female infants are born with virilization of the external genitalia. There is evidence that in a proportion of cases the virilization can be prevented if the mother takes a powerful steroid known as dexamethasone in a very small dose from 4 to 5 weeks’ gestation onward. Specific prenatal diagnosis of CAH can be achieved by DNA analysis of CV tissue. If this procedure confirms that the fetus is both female and affected, the mother continues to take low doses of dexamethasone throughout pregnancy, which suppresses the fetal pituitary–adrenal axis and can prevent virilization of the female fetus. If the fetus is male and either affected or unaffected, the mother ceases to take dexamethasone and the pregnancy can proceed uneventfully.

Treatment of a fetus affected with severe combined immunodeficiency (p. 203) has also been reported. The immunological tolerance of the fetus to foreign antigens introduced in utero means that the transfused stem cells are recognized as ‘self’, with the prospect of good long-term results.

When gene therapy (p. 350) has been proved to be both safe and effective, the immunological tolerance of the fetus should make it easier to commence such therapy before birth rather than afterward. This will have the added advantage of reducing the period in which irreversible damage can occur in organs such as the central nervous system, which can be affected by progressive neurodegenerative disorders.

Further Reading

Abramsky L, Chapple J, editors. Prenatal diagnosis: the human side. Cheltenham. UK: Nelson Thornes, 2003.

Dealing with the legal, emotional, and ethical issues, this nevertheless contains a lot of medical information in a very readable format with interesting case studies.

Brock DJH, Rodeck CH, Ferguson Smith MA, editors. Prenatal diagnosis and screening. Edinburgh, UK: Churchill Livingstone, 1992.

A comprehensive multiauthor textbook covering all aspects of prenatal diagnosis.

Drife JO, Donnai D, editors. Antenatal diagnosis of fetal abnormalities. London, UK: Springer, 1991.

The proceedings of a workshop on the practical aspects of prenatal diagnosis.

European Society for Human Reproduction and Embryology PGD Steering Committee. ESHRE Preimplantation Genetic Diagnosis Consortium: data collection III (May 2001). Hum Reprod. 2002;17:233-246.

An up-to-date appraisal of the use of PGD.

Lilford RJ, editor. Prenatal diagnosis and prognosis. Oxford, UK: Butterworth-Heinemann, 1990.

Provides useful information on recurrence risks for Down syndrome, the prognosis for abnormalities detected by ultrasonography, and decision analysis.

Stranc LC, Evans JA, Hamerton JL. Chorionic villus sampling and amniocentesis for prenatal diagnosis. Lancet. 1997;349:711-714.

A good review of the practical and ethical aspects of the two main prenatal invasive diagnostic techniques.

Whittle MJ, Connor JM, editors. Prenatal diagnosis in obstetric practice. Oxford, UK: Blackwell, 1989.

Describes prenatal diagnostic techniques and the types of abnormalities identified.