Clinical Manifestations in Carriers

Occasionally, carriers for certain disorders can have mild clinical manifestations of the disease (Table 20.1), particularly with some of the X-linked disorders. These manifestations are usually so slight that they are apparent only on careful clinical examination. Such manifestations, even though minimal, are unmistakably pathological; for instance, the mosaic pattern of retinal pigmentation seen in manifesting female carriers of X-linked ocular albinism. Unfortunately, in most autosomal and X-linked recessive disorders there are either no manifestations at all in carriers, or they overlap with the variation seen in the general population. An example would be female carriers of hemophilia, who have a tendency to bruise easily—this is not a reliable sign of carrier status, as this is seen in a significant proportion of the general population, for different reasons. In X-linked adrenoleukodystrophy, a proportion of carrier females manifest neurological features, sometimes relatively late in life when the signs might easily be confused with the problems of aging. Thus clinical manifestations are helpful in detecting carriers only when they are unmistakably pathological, which is the exception rather than the rule with most single-gene disorders.

Table 20.1 Clinical and Biochemical Abnormalities Used in Carrier Detection of X-Linked Disorders

G6PD, Glucose 6-phosphate dehydrogenase.

Biochemical Abnormalities in Carriers

By far the most important approach to determining the carrier status for autosomal recessive and X-linked disorders has been the demonstration of detectable biochemical abnormalities in carriers of certain diseases. In some disorders, the biochemical abnormality seen is a direct product of the gene and the carrier status can be tested for with confidence. For example, in carriers of Tay-Sachs disease the range of enzyme activity (hexosaminidase) is intermediate between levels found in normal and affected people. In many inborn errors of metabolism, however, the enzyme activity levels in carriers overlap with those in the normal range, so that it is not possible to distinguish reliably between heterozygote carriers and those who are homozygous normal.

Carrier testing for Tay-Sachs disease in many orthodox Jewish communities, which are at significantly increased risk of the disorder, is highly developed. Because of faith-based objections to termination of pregnancy, carrier testing may be crucial in the selection of life partners. A couple considering betrothal will first see their rabbi. In addition to receiving spiritual advice, they will undergo carrier testing for Tay-Sachs disease. If both prove to be carriers, the proposed engagement will be called off, leaving them free to look for a new partner. If only one proves to be a carrier the engagement can proceed, although the rabbi does not disclose which one is the carrier. Although such a strategy to prevent genetic disease may be possible in many communities where inbreeding is the norm, and their ‘private’ diseases have been well characterized either biochemically or by molecular genetics, in practice this is very rare.

In many single-gene disorders, the biochemical abnormality used in the diagnosis of the disorder in the affected individual is not a direct result of action of the gene product but the consequence of a secondary or downstream process. Such abnormalities are further from the primary action of the gene and, consequently, are usually even less likely to be useful in identifying carriers. For example, in Duchenne muscular dystrophy (DMD) there is an increased permeability of the muscle membrane, resulting in the escape of muscle enzymes into the blood. A grossly raised serum creatine kinase (CK) level often confirms the diagnosis of DMD in a boy presenting with features of the disorder (p. 307). Obligate female carriers of DMD have, on average, serum CK levels that are increased compared with those of the general female population (Figure 20.1). There is, however, a substantial overlap of CK values between normal and obligate carrier females. Nevertheless, this information can be used in conjunction with pedigree risk information (p. 344) and the results of linked DNA markers (p. 345) to help calculate the likelihood of a woman being a carrier for this disorder.

FIGURE 20.1 Creatine kinase (CK) levels in obligate carrier females of Duchenne muscular dystrophy and women from the general population.

(Adapted from Tippett PA, Dennis NR, Machin D, et al 1982 Creatine kinase activity in the detection of carriers of Duchenne muscular dystrophy: comparison of two methods. Clin Chim Acta 121:345–359.)

There is another reason for difficulty with carrier testing in the case of X-linked recessive disorders. Random inactivation of the X chromosome in females (p. 103) means that many, often the majority, of female carriers of X-linked disorders cannot be detected reliably by biochemical methods. An exception to this involves analysis of individual clones to look for evidence of two populations of cells, as with peripheral blood lymphocytes in female carriers of some of the X-linked immunodeficiency syndromes (p. 204). In a clinical setting, this is referred to as ‘X-inactivation studies’.

Linkage between a Disease Locus and a Polymorphic Marker

DNA Polymorphic Markers

The advent of recombinant DNA technology revolutionized the approach to carrier detection. Relatively seldom, nowadays, do conventional biochemical and blood group polymorphisms have a role because there are few that are sufficiently informative to be of practical clinical value. The large number of different types of DNA sequence variants (p. 17) in the human genome means that, if sufficient numbers of families are available, linkage of any disease with a polymorphic DNA marker is possible, providing the disease is not genetically heterogeneous (see locus heterogeneity below). After this is achieved, the information can be applied to smaller families. The demonstration of linkage between a DNA sequence variant and a disease locus overcomes the need to identify a biochemical defect or protein marker and the necessity for it to be expressed in accessible tissues. In addition, use of markers at the DNA level also overcomes the difficulties that occur in carrier detection due to X-inactivation for women at risk for X-linked disorders (p. 103).

Linked polymorphic DNA markers were frequently used in determining the carrier status of females in families where DMD has occurred; nowadays, direct gene sequencing is more likely to be used. An example is shown in Figure 20.2; individual III₃ wants to know whether she is a carrier and therefore at risk for having sons affected with DMD. Analysis of the pedigree reveals that her mother, II₄, along with her sister, II₁, and their mother, I₂, are all obligate carriers of DMD. The family is informative for a polymorphic CA dinucleotide repeat (p. 69) in the closely flanking region 5′ to the dystrophin gene known as Dys 5′ II, which can be demonstrated by polymerase chain reaction (PCR) (p. 56). The mutation in the dystrophin gene in the family is segregating with allele 1 and, because individual III₃ has inherited this allele from her mother, she is likely to be a carrier. Linked polymorphic DNA markers can be used for prenatal diagnosis to predict whether a male fetus is likely to be affected with DMD, even without knowing the specific dystrophin gene mutation in the affected male(s) (p. 308).

FIGURE 20.2 Family with Duchenne muscular dystrophy showing segregation of the CA repeat 5′ to the dystrophin gene known as Dys 5′ II.

(Courtesy J. Rowland, Yorkshire Regional DNA Laboratory, St. James’s Hospital, Leeds, UK.)

Clinical Examination

In some dominantly inherited disorders, simple clinical means can be used for presymptomatic diagnosis, taking into account possible pleiotropic effects of a gene (p. 109). For example, individuals with neurofibromatosis type I (NF1) can have a variety of clinical features (p. 298). It is not unusual to examine an apparently unaffected relative of someone with NF1, who has had no medical problems, only to discover that they have sufficient numbers of café-au-lait spots or cutaneous neurofibromas to confirm that they are affected. However, NF1 is a relatively rare example of a dominantly inherited disorder that is virtually 100% penetrant by the age of 5 or 6 years, with visible external features. With many other disorders, clinical examination is less helpful.

In tuberous sclerosis (TSC) a number of body systems may be involved and the external manifestations, such as the facial rash of angiokeratoma (Chapter 7; see Figure 7.5, A, p. 111) may not be present. Similarly, seizures and learning difficulties are not inevitable. In autosomal dominant polycystic kidney disease, which is extremely variable and may have a delayed age of onset, there may be no suspicion of the condition from routine examination, and hypertension may be borderline without raising suspicions of an underlying problem. Reaching a diagnosis in Marfan syndrome (p. 300) can be notoriously difficult because of the variable features and overlap with other joint hypermobility disorders, even though very detailed diagnostic criteria have been established.

Specialist Investigation

In conditions where clinical assessment leaves diagnostic doubt or ambiguity, special investigations of relevant body systems can serve to clarify status and presymptomatic diagnosis. In TSC imaging, studies of the brain by computed tomography to look for intracranial calcification (Figure 20.3) is more or less routine, as well renal ultrasonography to identify the cysts known as angiomyolipoma(ta) (Figure 20.4). Use of these relatively non-invasive tests in relatives of persons with TSC can often detect evidence of the condition in asymptomatic persons.

FIGURE 20.3 Intracranial calcification (arrows) in an asymptomatic person with tuberous sclerosis.

FIGURE 20.4 Renal ultrasonogram of an asymptomatic person with tuberous sclerosis showing abnormal echogenicity due to presumed angiomyolipomata (arrows).

Similarly, assessment for Marfan syndrome involves ophthalmic examination for evidence of ectopia lentis, echocardiography for measurement of the aortic root diameter, and sometimes magnetic resonance imaging of the lumbar spine to look for evidence of dural ectasia—all of these features count as major criteria in the disorder.

It is important to point out, however, that the absence of these findings on clinical or specialist investigation does not always exclude the diagnosis, although it does reduce the likelihood of the person having inherited the gene. If the relative frequencies of positive findings in persons with the disorder and in persons from the general population are known, it is possible to calculate a residual relative likelihood of having inherited the gene. This information can be used in conjunction with other information, such as the pedigree risk, in genetic counseling.

Biochemical Tests

For a number of autosomal dominant disorders, biochemical tests can determine whether or not a person at risk has inherited a gene. Examples include the use of serum cholesterol levels in those at risk for familial hypercholesterolemia (p. 175), currently more straightforward than genetic testing, and assay of the appropriate urinary porphyrins or the specific enzyme deficiency in the various dominant porphyrias (p. 179).

Linked DNA Markers

Linked DNA polymorphic markers can be used in presymptomatic diagnosis of dominantly inherited disorders but the principles and pitfalls discussed earlier apply. A common difficulty in dominantly inherited conditions is that of informativity, because often a key individual is deceased or unavailable for some other reason. Despite this, the availability of linked DNA markers has found widespread use in presymptomatic or predictive testing for a number of single-gene disorders inherited in an autosomal dominant manner. Increasingly, however, direct mutation analysis is possible and has replaced the use of linked markers—for example, in Huntington disease (HD). In NF1, TSC, and Marfan syndrome, mutation analysis is expensive and used more discerningly because it is not guaranteed to identify the causative mutation even where there is confidence about the diagnosis. Judicious use of linked DNA markers can be very helpful. Box 20.1 lists some of the more common conditions in which DNA analysis is regularly used to offer presymptomatic diagnosis, but there are of course many more.

The Disease

To justify the applied effort and resources allocated to screening, the disease should be sufficiently common and have potentially serious effects that are amenable to prevention or amelioration. This may involve early treatment, as in phenylketonuria diagnosed in the neonatal period (p. 167), or the offer of termination of pregnancy for disorders that cannot be treated effectively and are associated with serious morbidity and/or mortality.

The Test

The test should be accurate and reliable with high sensitivity and specificity. Sensitivity refers to the proportion of cases that are detected. A measure of sensitivity can be made by determining the proportion of false-negative results (i.e., how many cases are missed). Thus, if a test detects only 70 of 100 cases, it shows a sensitivity of 70%. Specificity refers to the extent to which the test detects only affected individuals. If unaffected people test positive, these are referred to as false positives. Thus, if 10 of 100 unaffected individuals have a false-positive test result, the test shows a specificity of 90%. Table 20.2 explains this further. Of great interest too is the positive predictive value of a screening test, which is the proportion of positive tests that are true positives; this is illustrated in Table 20.3.

Table 20.2 Sensitivity and Specificity

The Program

The program should be offered in a fair and equitable manner, and should be widely available. It must also be morally acceptable to a substantial proportion of the population to which it is offered. Participation must be entirely voluntary in the case of prenatal programs, but the ethical principles are more complex in neonatal screening for conditions where early treatment is essential and effective in preventing morbidity. In these situations, the principles of beneficence (doing good) and non-maleficence (not doing harm) are relevant. Easily understood information and well-informed counseling should both be readily available.

It is often stated that the cost of a screening program should be reasonable and affordable. This does not mean that the potential savings gained through a reduction in the number of affected cases requiring treatment should exceed or even balance the cost of screening, although this argument is popular with health economists who have to fund the program. It is reasonable to point out that cost–benefit analyses also have to take into account non-tangible factors such as the emotional costs of human suffering borne by both the affected individuals and those who care for them.

Phenylketonuria

Routine biochemical screening of newborn infants for phenylketonuria was recommended by the Ministry of Health in the United Kingdom in 1969, after it had been shown that a low-phenylalanine diet could prevent the severe learning disabilities that previously had been a hallmark of this condition (p. 167). The screening test, which is sometimes known as the Guthrie test, is carried out on a small sample of blood obtained by heel-prick at age 7 days. An abnormal test result is further investigated by repeat analysis of phenylalanine levels in a venous blood sample. A low-phenylalanine diet is extremely effective in preventing learning disabilities, and, although it is not particularly palatable, most affected children can be persuaded to adhere to it until early adult life when it can be relaxed. However, because high phenylalanine levels are toxic to the developing brain, a woman with phenylketonuria who is contemplating pregnancy should adhere to a strict low-phenylalanine diet both before and during pregnancy (p. 261).

Galactosemia

Classic galactosemia affects approximately 1 in 50,000 newborn infants and usually presents with vomiting, lethargy, and severe metabolic collapse within the first 2 or 3 weeks of life. Newborn screening is based on a modification of the Guthrie test with subsequent confirmation by specific enzyme assay. The early introduction of appropriate dietary restriction can prevent the development of serious complications such as cataracts, liver failure, and learning disability.

Congenital Hypothyroidism

Screening for congenital hypothyroidism was first introduced in the United States in 1974 and is now undertaken in most parts of the developed world. The test is based on assay of either thyroxine or thyroid-stimulating hormone. This disorder is particularly suitable for screening as it is relatively common, with an incidence of approximately 1 in 4000, and treatment with lifelong thyroxine replacement is extremely effective in preventing the severe developmental problems associated with the classic picture of ‘cretinism’. The most common cause of congenital hypothyroidism is absence of the thyroid gland rather than an inborn error of metabolism (p. 167). Congenital absence of the thyroid gland is usually not caused by genetic factors but on rare occasion is part of a wider syndrome.

Cystic Fibrosis

Newborn screening for cystic fibrosis has been introduced in several countries with a significant population of northern European origin. It is based on the detection of a raised blood level of immunoreactive trypsin, which is a consequence of blockage of pancreatic ducts in utero, supplemented by DNA analysis. The rationale for screening is that early treatment with physiotherapy and antibiotics improve the long-term prognosis. Initial results are encouraging but absolute confirmation of long-term benefit is awaited. In 2006 it became policy in England to introduce screening for all newborns.

Sickle Cell Disease and Thalassemia

Newborn screening based on hemoglobin electrophoresis is undertaken in many countries with a significant Afro-Caribbean community. As with cystic fibrosis, it is hoped that early prophylaxis will reduce morbidity and mortality, thereby improving the long-term outlook. In the case of sickle cell disease, treatment involves the use of oral penicillin to reduce the risk of pneumococcal infection resulting from immune deficiency secondary to splenic infarction (p. 152). Even in Western countries with good medical facilities, a significant proportion of sickle cell homozygotes, possibly as many as 15%, die as a result of infection in early childhood. In the case of thalassemia, early diagnosis makes it possible to optimize transfusion regimens and iron-chelation therapy from an early stage. Neonatal screening programs for both of these hemoglobinopathies were implemented in the United Kingdom in 2005, and antenatal screening (the mother, followed by the father if necessary) is under way in some areas of high risk. In some low-risk areas, there is a preference for antenatal screening to be targeted to high-risk couples after completion of an ethnicity questionnaire (p. 164).

Thalassemia

α- and β-thalassemia are caused by abnormal globin chain synthesis because of mutations involving the α- and β-globin genes or their promoter regions (p. 160). Both disorders show autosomal recessive inheritance and are extremely common in certain parts of the world, notably South-East Asia (α-thalassemia), Cyprus and the Mediterranean region, Italy, and the Indian subcontinent (β-thalassemia).

In Cyprus in 1974 the birth incidence of β-thalassemia was 1 in 250 (carrier frequency 1 in 8). After the introduction of a comprehensive screening program to determine the carrier status of young adults, which had the support of the Greek Orthodox Church, the incidence of affected babies declined by more than 95% within 10 years. Similar programs in Greece and Italy have seen a drop in the incidence of affected homozygotes of more than 50%.

If it is acceptable to judge the outcome of these screening programs on the basis of a reduction in the births of affected babies, then they have been very successful, due largely to the efforts of highly motivated staff interacting with a well-informed target population that has usually opted not to have affected children.

Sickle Cell Disease

In contrast to the Cypriot response to β-thalassemia screening, early attempts to introduce sickle cell carrier detection in the black population of North America were disastrous. Information pamphlets tended to confuse the sickle cell carrier state, or trait, which is usually harmless, with the homozygous disease, which conveys significant morbidity (p. 159). Several US states passed legislation making sickle cell screening in black people mandatory, and carriers suffered discrimination by employers and insurance companies. It is not surprising that public criticism was aroused, leading to abandonment of the screening programs and amendment of the ill-conceived legislation.

This experience emphasizes the importance of ensuring voluntary participation and providing adequate and appropriate information and counseling. Later pilot studies in the United States and in Cuba have shown that individuals of Afro-Caribbean origin are perfectly receptive to well planned, non-directive sickle cell screening programs.

Cystic Fibrosis

The discovery in 1989 that the Phe508del deletion/mutation (formerly deltaF508, or ΔF508) accounts for a high proportion of all cystic fibrosis heterozygotes soon led to the suggestion that screening programs could be implemented for carrier detection on a population basis. In the white population of the United Kingdom, the cystic fibrosis carrier frequency is approximately 1 in 25 and the Phe508del mutation accounts for 75% to 80% of all heterozygotes. A further 10% to 15% of carriers can be detected relatively easily and cheaply using a multiplex PCR analytical procedure.

Initial studies of attitudes to cystic fibrosis carrier detection yielded quite divergent results. A casual, written invitation generates a poor take-up response of around 10%, whereas personal contact during early pregnancy, whether mediated through general practice or the antenatal clinic, results in uptake rates of more than 80%. Studies have been undertaken to explore attitudes to cystic fibrosis screening among specific groups, such as school leavers and women in early pregnancy.

Two approaches for screening pregnant women have been considered. The first is referred to as two-step and involves testing pregnant mothers at the antenatal clinic. Those who test positive for a common mutation (approximately 85% of all cystic fibrosis carriers) are informed of the result and invited to bring their partners for testing—hence ‘two-step’ testing. If both partners are found to be carriers, an offer of prenatal diagnosis is made. This approach has the advantage that all carriers detected are informed of their result and further family studies—cascade screening—can be initiated. The disadvantage is that women whose partners are subsequently found to have a normal result experience considerable unnecessary anxiety, which creates a need for counseling and support.

The second approach is referred to as couple screening. This involves testing both partners simultaneously and disclosing positive results only if both partners are found to be carriers. In this way much less anxiety is generated, but the opportunity for offering tests to the extended family when only one partner is a carrier is lost. The results of pilot studies indicated that these ‘two-step’ and ‘couple screening’ approaches are equally acceptable to pregnant women, with take-up rates of approximately 70%. However, there is no publicly available cystic fibrosis screening for adults in the United Kingdom.

Positive and Negative Aspects of Population Screening

Well-planned population screening enhances informed choice and offers the prospect of a significant reduction in the incidence of disabling genetic disorders. These potential advantages have to be weighed against the potential disadvantages that can arise from the overenthusiastic pursuit of a poorly planned or ill-judged screening program (Box 20.4). Experience to date indicates that in relatively small, well-informed groups, such as the Greek Cypriots and American Ashkenazi Jews, community screening is welcomed. When screening is offered to larger populations the outcome is less certain.

A 3-year follow-up of almost 750 individuals screened for cystic fibrosis carrier status in the United Kingdom revealed that a positive test result did not cause undue anxiety, although some carriers had a relatively poor perception of their own general health. A more worrying outcome was that almost 50% of the individuals tested could not accurately recall or interpret their results. This emphasizes the importance of pretest counseling and the provision of accurate information that is easily processed and understood.