Q J Med 1999; 92: 223-232
© 1999 Association of Physicians
Commentary |
Genetic markers to predict polygenic disease: a new problem for social genetics
From the Department of Human Metabolism and Genetics, St. Bartholomew's Hospital, London, and 1 Centre for Clinical Science & Measurement, School of Biological Sciences, University of Surrey, Guildford, Surrey, UK
Professor D.J. Galton, Department of Human Metabolism and Genetics, St Bartholomew's Hospital, London EC1A 7BE
 |
Summary |
|---|
 Top Summary IntroductionProtein polymorphisms and...DNA polymorphism and disease...The use of DNA...The abusesThe futureReferences |
|---|
Many genetic markers that relate to common multifactorial disease in adults have been identified during the past 15 years. Their use as adjuncts for the diagnosis, prognosis, prediction of disease or targeting therapy for these disorders has begun, good examples being the Factor V Leiden mutation for venous-thromboembolism, lipoprotein lipase mutations for hypertriglyceridaemia and the apolipoprotein E4 variant for Alzheimer's dementia. However, extensive gene–gene and gene–environment interactions make their use more complex than markers for the simpler monogenic disorders (such as cystic fibrosis, or Duchennne's muscular dystrophy). Possible misapplication of the genetic markers for multifactorial disease in the fields of risk prediction, direct sales to the public, life assurance, employment rights, and legislation for regulation of their use are discussed.
|
Introduction |
|---|
|
TopSummaryIntroductionProtein polymorphisms and...DNA polymorphism and disease...The use of DNA...The abusesThe futureReferences |
|---|
During the past 15 years there have been unprecedented advances in the genetics of polygenic (or multifactorial) disorders, such as cardiovascular and malignant disease. For example, since 1982 more than 20 genes coding for proteins involved in plasma lipid transport have been identified and chromosomally localized, and their DNA sequences characterized.1,2 Frequently occurring mutations in some of these genes have been directly related to defects of plasma lipid transport and predisposition to premature coronary artery disease. For example, a common mutation in exon 4 of apolipoprotein CIII (C3175-G) has been consistently shown to relate to defects of plasma triglyceride and HDL transport and in some cases to associate with a predisposition to premature coronary artery disease3,4 and myocardial infarction.50,51 Genetic variants of another lipid transport protein, apolipoprotein E, have been shown to relate to both disorders of plasma cholesterol transport and the development of premature coronary artery disease.5,45 Since 1982 the genetics of more than 60 proto-oncogenes have been characterized. These genes have the potential to induce neoplastic transformation, and include genes coding for growth factors, growth factor receptors, protein kinases, signal transducers and nuclear transcription factors. Mutations in some of these oncogenes have been shown to relate to the development of cancers. For example mutations of a tumour suppressor gene, the APC (adenomatous polyposis coli) gene, located on the long arm of chromosome 5q21 relate to the development of colorectal cancer,6 and mutations of the tumour suppressor gene p53 relate to the development of ovarian and breast cancers.7 The genetics of other multifactorial disorders, such as diabetes mellitus, Alzheimer's disease, and venous thromboembolism have also been partially clarified in the present decade (see Table 1
for summary), and have provided a wealth of new genetic markers. A genetic marker can be defined as an allelic variant at a particular locus that shows a strong disease association. For polygenic disease, by definition, no single marker will ever allow a definitive diagnosis or certain risk prediction, but it is hoped that certain combinations of such markers will be found to do so in the future.
|
Many scientific advances have often resulted in misguided applications, even without the backup of a new technology. Darwin's Theory of Evolution published in 1859 was applied uncritically to the sociobiology of human populations. This led Francis Galton (born 1822, died 1911) to develop the subject of eugenics (eu: good; gen-: create) which he defined as `the science of improving inherited stock not only by judicious matings, but by all other influences'. On the basis of the key concepts of Darwin's revolutionary theories of `natural selection' favouring `the survival of the fittest' in the `struggle for existence', Galton made several proposals to improve the inherited constitution of human populations by a type of `artificial selection' of the ablest members of society. These briefly included: (i) providing social and financial incentives for the more able members of society to procreate; (ii) taking into account the achievements of family relatives when making appointments of individuals to key positions in society; and (iii) taking action against low achievers in society by possibly segregating those with mental deficiencies or other serious defects (insanity, habitual criminality) to restrict them from having offspring. By the early 20th century, item (iii) had been transformed by many authoritarian politicians in America, Norway, Sweden and Germany to provide the compulsory sterilization of individuals with mental deficiency, criminal tendencies, homosexuality or other antisocial tendencies. If such disastrous applications can occur without the benefit of a new technology, how much more potentially dangerous are these sociobiological concepts with the benefit of a set of brand new methods and techniques to characterize, manipulate and alter genetic material in humans. With the development of the field of genetic markers for polygenic disease and other complex biological traits, there are already indications for potential or actual abuse. For example, the complex behavioural trait of male homosexuality has been reported to be linked to a DNA polymorphism on chromosome Xq28 with a multipoint LOD score of 4 (p<0.0005) from a study of 33 pedigrees and linkage studies of 113 individuals.32 Their conclusions were that the development of male homosexuality appears to involve genetic factors including a locus at Xq28. Even before corroboration from other laboratories, this has been taken up by the informed theological press to consider testing foetuses for the `gay gene' and then to speculate whether we `should see abortion as means of cleansing our society of a gay population'.33 Although such extreme public hostility to homosexuality has in most European-based societies diminished, and such a suggestion would meet with deserved derision, the same cannot be claimed for paedophilia. A claim to identify a genetic determinant for the latter condition might well lead to aggressive forms of social legislation. One problem is that complex behavioural traits such as overeating and obesity can, in very rare instances, be related to defects of a single gene. In the majority of cases, overeating leading to morbid obesity is due to a complex interplay of nutritional, social, emotional, metabolic, and genetic factors. But in very rare instances a single homozygous frameshift mutation (involving the deletion of a single guanine nucleotide in codon 133) of the leptin gene associated with very low serum leptin levels may directly relate to early-onset obesity in humans.8 Other examples would include the development of ischaemic heart disease and mutations of the LDL receptor gene.9 This type of result has led to an uncritical transfer of the ideas underlying the genetic determinism of rare monogenic disorders such as phenylketonuria, cystic fibrosis, and Ducheme muscular dystrophy, to an oversimplified genetic determinism for complex multifactorial disease such as cancer, ischaemic heart disease, Alzheimer's dementia, etc. This conceptual transfer from monogenic disorders to polygenic disease is quite inappropriate, because polygenic disease involves the co-inheritance of several genetic determinants that usually have to interact with environmental factors before the disease becomes manifest. The genetic determinants for a phenotype can be variable and they may interact in different ways; some of the genetic factors can even be protective for the occurrence of the disease. This paper therefore presents a review and critique of some medical and social issues arising from the use of genetic markers for polygenic disease, and highlights some of the basic limitations of the current approach.
|
Protein polymorphisms and disease associations |
|---|
|
TopSummaryIntroductionProtein polymorphisms and...DNA polymorphism and disease...The use of DNA...The abusesThe futureReferences |
|---|
Multifactorial disease can be defined as a disorder with a clear-cut hereditary component (i.e. it aggregates in families but without the usual Mendelian inheritance ratios for monogenic disease) and where environmental factors are required for the occurrence of the disease (Figure 1
).
|
The first major study of disease association was the relationship between stomach cancer and blood groups (A,B,O polymorphisms). Blood group O and stomach cancer are both more common in the North of England. Aird and his colleagues discovered that the strongest association was in fact with blood group A.46 Later studies did reveal a weak association between blood group O, duodenal ulcers and non-secretor status, with a contribution of 2.5% to the total variance. Informative markers were next found at the polymorphic HLA locus on chromosome 6 which showed strong disease associations;47 for example, HLA B27 is found in more than 90% of subjects with ankylosing spondylitis, but it is of no use in predicting the disease, because approximately 10% of the healthy European population also carry HLA B27; whereas only approximately 1% of them will develop ankylosing spondylitis. Another striking example is the almost 100% association of narcolepsy with the HLA DR2 antigen, yet ~15% of the European population possess this particular polymorphism without developing narcolepsy. Clearly such polymorphic variants would be of no use for risk prediction in the population, but they may help in research to elucidate the mechanisms of pathogenesis of the disease, and they may be used as an adjunct to diagnosis, i.e. if a patient with sacroileitis also possesses the HLA B27 variant, then the diagnosis of ankylosing spondylitis is much more likely before other manifestations of the disease become apparent.
Other examples of protein polymorphisms which have proved useful adjuncts in diagnosis and which have also elucidated possible pathogenic mechanisms include the apolipoprotein E polymorphisms. More than 90% of subjects with Type III hyperlipoproteinaemia have the apolipoprotein E phenotype of E2/E2 which occurs between 0.2 and 1.6% of an unselected healthy population studied in various countries, but only 2% of subjects with E2/E2 will actually develop the dyslipidaemia.10 So like the HLA markers, it cannot be used for risk prediction, but can be used as an adjunct to diagnosis. With regard to pathogenic mechanisms, it has been shown that the affinity of the apo E2 peptide for hepatic remnant receptors is less than that of apo E3 which in turn is less than that of apo E4; and the suggestion is that remnant lipoprotein particles carrying E2/E2 cannot be cleared effectively from the bloodstream as those particles possessing the other apolipoprotein E variants of E3 or E4.11 Although more than 95% of E2/E2 homozygotes do not develop Type III hyperlipidaemia, an increase in their VLDL-cholesterol content is detectable, probably because of the slower clearance of their apo-E2-containing lipoproteins. It is therefore apparent that other factors, either genetic or environmental, must occur for Type III hyperlipidaemia to develop. So even in this well established and characterized example, the use of genetic polymorphisms cannot be used for risk prediction, although they are of use as an adjunct to diagnosis.
|
DNA polymorphism and disease associations |
|---|
|
TopSummaryIntroductionProtein polymorphisms and...DNA polymorphism and disease...The use of DNA...The abusesThe futureReferences |
|---|
With the use of recombinant DNA techniques a wealth of DNA markers have been found throughout the human genome, which has greatly enlarged the scope of studying the genetics of disease associations or linkage other than at the HLA locus. These markers were initially identified as restriction fragment length polymorphisms and then with refinement of techniques to exonic (some of which cause a change in protein structure or function) or intronic mutations; and currently microsatellite markers are being used to scan the human genome for disease-related alleles either by linkage or association studies. The first DNA polymorphisms that were demonstrated to show disease relationships and have withstood the test of time are the variable number of tandem repeats (vntr) 5' to the insulin gene on chromosome 11 that relates to insulin-dependent diabetes mellitus;12,13 and a biallelic polymorphism in exon 4 of the apolipoprotein CIII gene on chromosome 11 where C3175 is transverted to G.3 This mutation is transcribed but not translated, and the rarer allele shows disease relationships with dyslipidaemia (raised plasma triglyceride and low HDL-cholesterol).14–17 Common genetic variants of lipoprotein lipase have also been identified that show relationships with dyslipidaemia.43–45 Subsequently many other DNA polymorphic variants have been demonstrated to relate to a variety of other disorders including cancer, the dementias, venous thrombosis, and other metabolic diseases. Some of these, with their DNA markers, are listed in Table 1
. |
The use of DNA polymorphisms as disease markers |
|---|
|
TopSummaryIntroductionProtein polymorphisms and...DNA polymorphism and disease...The use of DNA...The abusesThe futureReferences |
|---|
Screening adults
The discovery of disease-related alleles as listed in Table 1
has improved our understanding of the pathogenesis of such disorders and the pathways involved in their development (for example, the apolipoprotein E gene variants relating to the dyslipidaemias as described previously). Some DNA polymorphisms can be used as adjuncts to diagnosis in the same way that some HLA markers are used in the diagnosis of, for example, ankylosing spondylitis. Thus apolipoprotein E4 has been proposed as an adjunct to diagnosis in subjects presenting with an early dementia with a provisional diagnosis of Alzheimer's disease. DNA variants can also be used in the prognosis of disease. For example, there are more than 230 mutations described for the low-density lipoprotein receptor,18 and the site of these mutations can have very different effects on the phenotype, familial hypercholesterolaemia. Thus point mutations in exon 16 or 17 coding for the transmembrane portion of the molecule can have serious effects on LDL clearance from plasma because the receptor comes adrift from the cell membrane leading to a severe monogenic form of familial hypercholesterolaemia. Other mutations in, for example, exons 7–14, that only affect the region of epidermal growth factor precursor-homology of the molecule, may minimally affect ligand binding of LDL or attachment of the receptor to the cell surface, leading to a much less severe disease phenotype19 or even require other genetic variants (e.g. at the apolipoprotein E locus) for overt hypercholesterolaemia to develop as a polygenic disorder.
Another use for the identification of the mutations as listed in Table 1 is to reveal new therapeutic targets. For example, the discovery that familial hypercholesterolaemia is due to mutations of the LDL receptor has led to the development of a new class of drugs, the statins, to increase the number of defective cell-surface LDL receptors in patients with heterozygous familial hypercholesterolaemia and thereby improving LDL clearance from the bloodstream. This class of drugs has been used successfully to treat familial hypercholestolaemia and can reduce the incidence of long-term complications of coronary atherosclerosis.20,21 It may also be possible in the future to stratify patients with a given disorder by their particular genotype and so predict whether they will respond to a pharmacological agent. This will obviate the need for placing patients on long-term drug therapy when it can be predicted from their genotype analysis that they will not respond.
Finally, some of the genetic variants in Table 1 have been proposed for use in risk prediction for cancer.22,23 The APC gene variants to predict risks of developing colorectal cancer, or the Her2/neu gene to predict recurrence of breast cancer have both been approved by the Food and Drug Administration in the United States. The Her2/neu gene to predict the recurrence of breast cancer detects the levels of amplification of this gene on chromosome 17 in a subset of breast cancers. In clinical trials, 31% of patients with localized breast cancer at the time of diagnosis who had a positive Her2/neu gene test died within 5 years of surgery, whereas 97% of those with a negative test were still alive 5 years after surgical resection of their tumour. This test may indicate which group of patients with breast cancer will need more aggressive therapy and more frequent follow-up. However, no such genetic markers have been reliably validated to be of use in the risk prediction of complex behavioural traits in the healthy population such as intelligence, cognition, sociopathy, violent behaviour, homosexuality, paedophilia, etc. There have been several reports, the strongest perhaps being a genetic marker on the X chromosome to predict gender preference in men. A Dutch group recently reported a genetic marker on the X chromosome that may identify a predisposition to violent behaviour.35 In a large Dutch kindred, all affected males showed aggressive and sometimes violent behaviour such as arson, attempted rape and exhibitionism. The locus for this disorder was assigned to Xp 11-21 with a maximal multipoint LOD score of 3.69. A postulated mechanism for disturbance of monoamine metabolism was considered since the locus was very close to the monoamine oxidase type A gene. As in the example of overeating and morbid obesity, there could possibly be a locus that predisposes to violent behaviour in very rare instances in a small subset of people, but it is clearly not possible to extrapolate to all violent members of the general population until the evidence from many research groups has confirmed the relationship.
Screening the blastocyst
The mutations listed in Table 1 could also be used to screen pre-implantation embryos and if any are found the embryo could be discarded and another zygote used in its place. This is not done at present. The feasibility of preimplantation genetic diagnosis was first established in the 1980s after pressure from married couples who were at risk of transmitting an X-linked disorder and who had moral objections to abortion. Polymerase chain reaction (PCR) amplification of DNA isolated from one or two cells removed from an 8–10-cell embryo after in vitro fertilization, now permits pre-implantation diagnosis of cystic fibrosis, Tay-Sachs disease, Duchenne muscular dystrophy, and mutations of the APC gene for familial adenomatous polyposis coli.24,25 Worldwide, more than 100 babies have been born after preimplantation genetic diagnosis with no reported increase in congenital anomalies. The main advantage of preimplantation diagnosis is that it obviates the need for selective abortion; the main disadvantage is that it requires in vitro fertilization which currently has only a success rate of about 1 : 5. The UK Human Fertilization and Embryology Authority (HFEA) is currently licensing several hospitals to perform such diagnosis for monogenic disease of preimplantation embryos, and of course the ethics of extending such techniques to screen for polygenic disorders to create `designer babies' for healthy married couples has now to be critically assessed. It is natural that parents would want, and have the right, to chose the best possible health for their children, and pre-implantation screening may be one way of ensuring `optimal' gene transfer to their offspring. One recommendation by a group of advisors on the Ethical Implications of Biotechnology of the European Commission was made for pre-natal diagnosis (of which pre-implantation diagnosis is a special case) and stated that `it should always be considered a medical act and be offered on the basis of specific medical indications. The choice of sex or other characteristics for non-medical reasons is an ethically unacceptable indication for pre-natal diagnosis and should be prohibited'. All the conditions listed in Table 1 would fall into the category of medical conditions and therefore be potential candidates for detection. The technique would fall within the original definition of eugenics by Galton, and is potentially open to all the abuses of discrimination that followed in the train of this field in the early 20th century. It is covertly making a negative value statement about the other members of society who possess the trait that is being screened and rejected for implantation. That is not to say that the use of the technique should be banned, only that extreme care should be taken in its implementation.
|
The abuses |
|---|
|
TopSummaryIntroductionProtein polymorphisms and...DNA polymorphism and disease...The use of DNA...The abusesThe futureReferences |
|---|
The basic problem here would be to distinguish between legitimate and illegitimate uses of genetic markers to discriminate individuals in the areas of: disease control (from the time of conception to adult life), employment, career promotion, life and health assurance.
Risk prediction
The unresolved status of the `gay' or the `violence' genes emphasizes the importance of strict criteria to validate the use of genetic markers for prediction before ever being considered in the public domain of journalism or textbooks. One suggested set of criteria for validation of a genetic marker is set out in Table 2. The cases so far discussed of the `gay' and `violence' genes have only reached stage 2 of the suggested criteria for validation of a genetic marker and in the case of the `violence' gene the phenotype has only been studied in a single extended pedigree. Yet many informed monographs written for the general public by ethicists, philosophers and theologians are discussing the implications of these findings as though they have already direct relevance to social problems of crime and sexual abuse.33,36 Even the most well established and characterized genetic variants that predispose to a well defined phenotype such as colorectal cancer and the APC gene mutations or breast cancer are only at stage 3 of the criteria for validation in Table 2. The use of DNA markers for predicting risks of polygenic disease are always going to suffer from some limitations until, probably, the whole genome has been mapped and its function ascertained. However, some combinations of genetic markers may well be found beforehand to be of use for diagnosis or risk prediction of multifactorial disease. But uncertainty will arise from the nature of polygenic disease that (i) involves environmental interactions for expression, (ii) involves gene–gene interactions and can require the assessment of protective genotypes acting against the appearance of the disease phenotype.
|
Environmental interactions
The genetic variants underlying multifactorial conditions are not the sole determinants of the disease. For the dyslipidaemias, type II diabetes or premature atherosclerosis, environmental factors such as food intake, food composition, physical exercise, and obesity may all be involved before full manifestation of the disease. For example, the genotype underlying the inherited basis of non-insulin-dependent diabetes mellitus (type II diabetes mellitus) often requires caloric excess and obesity for the disease to manifest, and when the patient returns to normal body mass index, the disorder of abnormal blood sugars can resolve. Some genetic variants of apolipoprotein CIII only manifest as a dyslipidaemia when the patient consumes excess amounts of alcohol, and when alcohol is restricted, the dyslipidaemia resolves.26 The effects of a high cholesterol diet on serum cholesterol levels are influenced by genetic factors, such as apolipoprotein B and E genotypes.48 Thus risk prediction on the basis of genetics alone is incomplete until the extent and impact of the exposure to environmental factors have been assessed.
Gene–gene interactions
The fact that two or more genetic variants are required for expression of the disease phenotype implies that there could be gene–gene interactions. With current methodology these are difficult to demonstrate, but good examples of genetic interactions include observations where some genetic variants appear to protect against the development of the disease. A common mutation occurring at a frequency of about 10–15% in the healthy population in the gene coding for lipoprotein lipase involves a C to G transversion at codon serine 447, converting it to a stop codon, and thereby truncating the protein by two amino acids at the carboxy-terminal end.27 This might be expected to impair the function of the enzyme for clearance of plasma triglycerides, but several research groups have shown the opposite. The mutation is found more frequently in healthy controls than in subjects with dyslipidaemia28 or premature coronary atherosclerosis.29 Although the kinetic constants (Km and Vmax) of the mutant enzyme appear no different from those of the wild type, there is greater release of the mutant enzyme from the vascular endothelium compared to the wild-type enzyme. This may account for its enhanced activity in vivo. Alternatively, the mutation may affect dimerization of the enzyme for full catalytic activity. So this variant appears to be protective for the development of dyslipidaemia and coronary artery disease. Other examples of protective genotypes are: (i) some HLA variants such as HLA DR 2/DQB1*0602 appear to have a stronger protective role against the development of insulin dependent diabetes (type I diabetes mellitus) than the susceptibility genotypes of HLA DR 3/4;37 and (ii) a common genetic variant of the apolipoprotein A II gene, detectable as a restriction fragment length polymorphism using the enzyme MspI appears to be associated with higher plasma triglyceride levels when co-inherited with an uncommon variant of the apolipoprotein AI-CIII-AIV gene cluster.49
So before accurate risk prediction can be achieved in any individual, all the susceptibility and protective genotypes will have to be mapped, and this may require an extensive genome search. With the development of the new oligonucleotide array (DNA chip) technology this may become feasible within the next decade, but it is certainly not possible to perform now.
Direct marketing of genetic tests
The discovery of genetic variants that predispose individuals to inherited forms of cancer have provoked widespread public interest and created demands for such tests which in part have been satisfied by commercial laboratories. Assays for mutations of the APC gene that predisposes individuals to colorectal cancer have been marketed and performed by a private company La Corp (Baltimore USA) following referral by a physician or gastroenterologist. In one study,38 out of the 177 patients tested in 1995, only 18.6% received genetic counselling before the test and only 16.9% provided written informed consent. In 31.6% of the cases, the referring physicians misinterpreted the test results. Some patients at risk for familial adenomatous polyposis coli would have been given a false-negative result, because the test cannot detect APC gene mutations in about 20% of patients with familial adenomatous polyposis. Hence genetic testing rules out this disorder in a person at risk only when no mutation is found in that person, and a mutation has been identified in an affected family member. The APC gene is just one of a large array of DNA-based tests for assessment of risk of developing early dementias or cancer. The Her2/neu gene amplification test for recurrence of breast cancer is marketed by Oncor and has received FDA approval for risk prediction as described previously. The scope for misapplication and misinformation of patients, and a creation of public anxiety in view of variable environmental factors and the presence of protective genotypes are enormous, and might have many undesirable consequences on lifestyle choices by the patients, as well as the ability for purchasing some forms of life or health assurance.
Life and health assurance and genetic discrimination
The insurance industry has already laid out guidelines for the declaration of genetic tests before the purchase of life insurance, disability income insurance, and critical illness insurance. In the UK, for purchase of life insurance up to a total of £100 000 which is directly linked to a new mortgage for a private dwelling, the results of any genetic tests must be reported but may not be taken into account by the insurance company.39 It is implied that for sums greater than £100 000, such genetic tests will be taken into account and the premiums adjusted accordingly. Most insurance companies work on the basis of equity, insisting that the premium paid must truly reflect the risks of disability of the individual. Otherwise people at high risk will purchase a higher-claim policy, and will make financial gain out of the premiums paid by lower-risk individuals. Other possible modes of operating are on the basis of either equality, where the inherited risks of disease are not taken into account since the individual should not be held responsible or penalized for the genetic endowment he receives arbitrarily from his parents; or even on the basis of solidarity, where the premiums are reduced for individuals carrying genetic risks for the development of chronic disease and his financial claims would be paid out of the premiums of the more fortunate healthy members of society. Both models of equality or solidarity, although appealing to notions of social justice, are unacceptable to the insurance industry which is a commercial organization in the business of predicting risks to health or life.40 However, risk discrimination by the insurance industry may turn into genetic discrimination with the availability of all the new genetic tests for chronic illness appearing on the market; and this may profoundly affect the purchase of long-term care insurance or critical illness cover for individuals found to possess susceptibility genes for chronic disease. A somewhat analogous situation may arise to that for type 1 diabetics, who may find it difficult to obtain travel insurance. Currently a moratorium has been declared for 2 years by the UK Human Genetics Advisory Commission to prevent the disclosure of existing genetic tests to insurance companies until more information is available on the actuarial relevance of such tests. In the European Council at Strasbourg Article 11 of the document on Human Rights and Biomedicine (1997) states that `Any form of discrimination against a person on the grounds of his or her genetic heritage is prohibited'.41 Clearly, the claims for actuarial fairness by the insurance industry versus the claims for social justice by the European Council will require new forms of legislation to clarify this problem.
Employment rights and genetic discrimination
At least six American corporations screen employees for sensitivity to toxic substances that they may encounter during their work and, reasonably, deny employment to such allergic individuals. This has been extended to genetic screening, and denial of employment has already occurred in the USA for individuals who carry a single mutant allele for sickle-cell anaemia, even though clinically they are unaffected.42 The problem arose out of the demand for sickle-cell screening for African Americans of marriageable age. If both partners were discovered to carry the sickle-cell trait, they were counselled with regard to their future family. Sickle-cell screening laws were enacted in seventeen states, often under the sponsorship of African American legislators. In 1972, Congress passed the National Sickle Cell Anaemia Control Act which provided for research, screening, counselling and education. But in the preamble to this Act, it stated erroneously that 2 million Americans suffered from sickle-cell disease. In fact 2 million were carriers of the harmless sickle-cell trait and fewer than 100 000 had the disease. However, the American Air Force Academy, acting on this erroneous statement, restricted the entry of heterozygous subjects to their Academy, Commercial Airlines restricted sickle cell carriers to ground employment only, and African Americans found their career promotion blocked to high-quality jobs. The justification for this was the fear that sickling of red cells in heterozygotes would occur at high altitudes with minor degrees of oxygen deficiency. There was no clinical evidence for this. Spokespersons for the African American community in the USA indicted the compulsory sickle-cell screening programme as a form of racial discrimination and eventually the law was repealed.30 The sickle-cell screening programme began with the good intentions of reducing the birth rate of babies with sickle-cell disease, but due to misapplication became a means of genetic discrimination against one subset of the American population for high-quality employment and insurance. However, employers should have the rights to information about genetic defects of prospective employees if this would put at hazard the lives of themselves or others during the course of their work. Mutations of the rhodopsin gene leads to some forms of night blindness, and clearly such individuals should not be employed as airline pilots, sailors, or involved in driving vehicles at night. Similar considerations could apply to the genetic defects underlying colour blindness or narcolepsy, or to the genetic defects of the LDL receptor that can predispose to early and sudden myocardial infarction. With the array of new genetic tests that are becoming available, new legislation will be required to balance employment rights for at risk individuals with the possible hazards that they may create for themselves or others in the work-place.
|
The future |
|---|
|
TopSummaryIntroductionProtein polymorphisms and...DNA polymorphism and disease...The use of DNA...The abusesThe futureReferences |
|---|
With the development of the new genetic technology, eugenics can be more widely defined as `the use of science applied to the qualitative and quantitative improvement of the human genome'. It is a salutary lesson that neither Francis Galton (nor Charles Darwin who largely supported the concept) could foresee the gross misapplication of these eugenic ideas by authoritarian politicians, less than 30 years from the death of Francis Galton.31 Such practices have been reported to have continued up to 1994, where in Norway, sterilization experiments using radiation were performed on mentally retarded subjects in hospitals to investigate the effects of nuclear fallout on fertility.52 What use might be expected for the new genetic markers and voluntary eugenics in the future, and what safeguards should be put in place to prevent future gross misapplications? One can expect the availability of a large array of susceptibility and protective genes to be identified within the next two decades which will be of use to partly predict the risks of developing common disorders such as cancer, premature atherosclerosis, hypertension, diabetes mellitus, and Alzheimer's dementia, etc. A good start has already been made in this direction, as described in Table 1
. Such markers can be used either for direct or indirect eugenic purposes. Direct eugenic methods could in the future include the screening of genetic markers for polygenic disorders in preimplantation embryos and discarding those who possessed a large number of such disease-related alleles. This procedure is already in practice for diagnosis of monogenic disorders in several infertility clinics throughout the UK, and could become much more widespread for healthy married couples to choose `designer' embryos, especially if any genetic disease was present in their families. It is possible but doubtful whether selective abortion of implanted embryos following antenatal screening would ever be an acceptable practice (as is done currently for some homozygous monogenic disorders) since the methods for risk prediction of these polygenic disorders are too uncertain at present. Indirect eugenic effects may arise from the use of genetic markers for disease prediction by creating a genetic `underclass' and making it more difficult for affected individuals to obtain mortgages, life assurance products, employment, or job promotion. All such social factors arising from genetic discrimination may influence an individual's reproductive choice and size of family. There is clearly a need for a regulatory framework derived from reasoned ethical values in which to use the new genetics and which should at the least take into account the following factors: respect for individual autonomy and privacy; personal reproductive choices uninfluenced by exterior pressures related to genetic discrimination; rights to as normal a family life as possible; and protection of the rights of other family members and members of the rest of society and their safeguard from harm.
Currently there are a multitude of regulatory bodies (but without statutory powers) ranging from local ethical committees of hospitals, National Funding Bodies such as the MRC or Wellcome Trust, various advisory bodies (including the Human Genetics Advisory Commission, the Advisory Committee on Genetic Modification, the UK Xenotransplantation Interim Regulatory Authority, the UK Human Fertilization and Embryology Authority, the European Medicines Evaluation Agency) to the Convention of Human Rights and Biomedicine of the Council of Europe. The difficulty of course is finding the correct balance between individual personal freedoms and legitimate interference by the various government agencies. Even the most private and intimate of personal actions can be legitimately the subject of state interference if other members of society could be put at risk (e.g. the intention to make it a criminal offence if a man has sexual intercourse with a woman knowing that he is suffering from AIDS). The new laws relating to the new genetics should be primarily made for the benefit of the community, and it is the duty of the legislators that no injustice be done even to an individual. The legal framework has to be equitable and not transgress the common rights of citizens; it should be flexible as new developments occur and be able to accommodate every particular foreseeable case as new discoveries are made. But legal responses are not easy to formulate in such a rapidly changing field such as the new (predictive) genetics. It may be better to react to new situations as they arise as precedents, rather than to attempt pro-active legislation.
|
Acknowledgments |
|---|
This work was supported by Grant PL 931211 of the Commission of the European Communities (to DJG), by the British Heart Foundation (GAF), and the Joint Research Board of St Bartholomew's Hospital.
|
References |
|---|
|
TopSummaryIntroductionProtein polymorphisms and...DNA polymorphism and disease...The use of DNA...The abusesThe futureReferences |
|---|
1. Dammerman M, Breslow Jl. Genetic basis of lipoprotein disorders. Circulation 1995; 91:505–12.
2. Breslow Jl. Apolipoprotein genes and atherosclerosis. J Clin Invest 1992; 70:377–84.
3. Rees A, Shoulders CC, Stocks J, Galton DJ, Baralle FE. DNA polymorphism adjacent to human apoprotein A-1 gene: relation to hypertriglyceridaemia. Lancet 1983; 1:444–6.[Web of Science][Medline]
4. Rees A, Stocks J, Williams LG, et al. DNA polymorphisms in the apolipoprotein C-III and insulin genes and atherosclerosis. Atherosclerosis 1985; 58:269–75.[Web of Science][Medline]
5.
Mahley RW. Apolipoprotein E: cholesterol transport protein with expanding role in cell biology. Science 1988; 240:622–30.
6. Ravelingien N, Pector JC, Velu T. Contribution of molecular oncology in the detection of colorectal carcinomas. Acta Gastroenterol Belg 1995; 58:270–3.[Medline]
7.
Cho Y, Gorina S, Jeffrey PD, Pavletich NP. Crystal Structure of a p53 Tumor Suppressor-DNA Complex: Understanding Tumorigenic Mutations. Science 1994; 265:346–55.
8. Montague CT, Farooqi IS, Whitehead JP, et al. Congenital leptin deficiency is associated with severe early-onset obesity in humans. Nature 1997; 387:903–8.[Medline]
9. Hobbs HH, Brown MS, Goldstein JL. Molecular genetics of the LDL receptor gene in familial hypercholesterolemia. Hum Mutat 1992; 1:445–66.[Medline]
10. Mahley RW, Innerarity TL, Rall SC, Jr, Weisgraber KH. Lipoproteins of special significance in atherosclerosis. Insights provided by studies of type III hyperlipoproteinemia. Ann N Y Acad Sci 1985; 454:209–21.[Web of Science][Medline]
11. Schneider WJ, Kovanen PT, Brown MS, et al. Familial dysbetalipoproteinemia. Abnormal binding of mutant apoprotein E to low density lipoprotein receptors of human fibroblasts and membranes from liver and adrenal of rats, rabbits, and cows. J Clin Invest 1981 Oct; 68:1075–85.
12. Hitman GA, Tarn AC, Winter RM, et al. Type 1 (insulin-dependent) diabetes and a highly variable locus close to the insulin gene on chromosome 11. Diabetologia 1985; 28:218–22.[Web of Science][Medline]
13. Raffel LJ, Hitman GA, Toyoda H, Karam JH, Bell GI, Rotter JI. The aggregation of the 5' insulin gene polymorphism in insulin dependent (type I) diabetes mellitus families. J Med Genet 1992 Jul; 29:447–50.[Web of Science][Medline]
14. Humphries SE, Talmud PJ. Apolipoprotein C111 gene variation and dyslipidaemia. Current Opin Lipidol 1997; 8:154–8.[Web of Science][Medline]
15. Hayden MR, Kirk H, Clark C, et al. DNA polymorphisms in and around the Apo-A1-CIII genes and genetic hyperlipidemias. Am J Hum Genet 1987; 40:421–30.[Web of Science][Medline]
16.
Dammerman M, Sandkuijl LA, Halaas JI, Chung W, Breslow JI. An apolipoprotein CIII haplotype protective against hypertriglyceridemia is specified by promoter and 3' untranslated region polymorphisms. Proc Natl Acad Sci USA 1993; 90:4562–6.
17. Zeng Q, Dammerman M, Takada Y, Matsunaga A, Breslow JL, Sasaki J. An apolipoprotein CIII marker associated with hypertriglyceridemia in Caucasians also confers increased risk in a west Japanese population. Hum Genet 1995; 95(4):371–5.[Web of Science][Medline]
18. Hobbs HH, Russell DW, Brown MS, Goldstein JL. The LDL receptor locus in familial hypercholesterolemia: mutational analysis of a membrane protein. Ann Rev Genet 1990; 24:133–70.[Web of Science][Medline]
19. Russell DW, Lehrman MA, Sudhof TC, et al. The LDL receptor in familial hypercholesterolemia: use of human mutations to dissect a membrane protein. Cold Spring Harb Symp Quant Biol 1986; 51(Pt 2):811–9.
20. Havel RJ. Analysis of angiographic trial data in women. Drugs 1994; 47(Suppl 2):11–15.
21. Betteridge DJ, Durrington PN, Fairhurst GJ, et al. Comparison of lipid-lowering effects of low-dose fluvastatin and conventional-dose gemfibrozil in patients with primary hypercholesterolemia. Am J Med 1994; 96:45S–54S.[Medline]
22. Brandt B, Vogt U, Schlotter CM, et al. Prognostic relevance of aberrations in the erbB oncogenes from breast, ovarian, oral and lung cancers: double-differential polymerase chain reaction (ddPCR) for clinical diagnosis. Gene 1995; 159(1):35–42.[Web of Science][Medline]
23. Ormiston W. Hereditary breast cancer. Eur J Cancer Care (Engl) 1996; 5:13–20.[Medline]
24. WDelhanty JDA, Handyside AH, Winston RML. Preimplantation diagnosis. Lancet 1994; 343:1569–70.
25. Schulman JD, Edwards RG. Preimplantation diagnosis in disease control, not eugenics. Hum Reprod 1996; 11:463–4.
26. Stocks J, Holdsworth G, Galton D. Hypertriglyceridaemia associated with an abnormal triglyceride-rich lipoprotein carrying excess apolipoprotein C-III-2. Lancet 1979; 2:667–71.[Web of Science][Medline]
27. Stocks J, Thorn JA, Galton DJ. Lipoprotein lipase genotypes for a common premature termination codon mutation detected by PCR-mediated site-directed mutagenesis and restriction digestion. J Lipid Res 1992; 33:853–7.[Abstract]
28.
Mattu RK, Needham EW, Morgan R, et al. DNA variants at the LPL gene locus associate with angiographically defined severity of atherosclerosis and serum lipoprotein levels in a Welsh population. Arterioscler Thromb 1994; 14:1090–7.
29. Zhang Q, Cavanna J, Winkelman BR, et al. Common genetic variants of lipoprotein lipase that relate to lipid transport in patients with premature coronary artery disease. Clin Genet 1995; 48:293–8.[Web of Science][Medline]
30. Duster T. Backdoor to Eugenics. Keegan Paul, USA, 1991.
31.
Galton DJ, Galton CJ. Francis Galton: and eugenics today. J Med Ethics 1998; 24:99–105.
32. Hu S, Pattatucci AM, Patterson C, et al. Linkage between sexual orientation and chromosome Xq28 in males but not females. Nature Gen 1995; 11:248–56.[Web of Science][Medline]
33. Peters T. Playing God: genetic determinism and human freedom. New York, Routledge, 1997.
34. Josefson D. FDA approves genetic test for women with breast cancer. Br Med J 1998; 316:168.
35. Brunner HG, Nelen MR, van Zandroort P, et al. X-linked borderline mental retardation with prominent behavioural disturbance: phenotype, genetic localisation and evidence for disturbed monoamine metabolism. Am J Hum Genet 1993; 52:1032–9.[Web of Science][Medline]
36. Kitcher PK. The lives to come. London, Penguin Books, 1996.
37.
Todd JA. Genetic analysis of type 1 diabetes using whole genome approaches. Proc Nat Acad Sci USA 1995; 92:8560–5.
38.
Giardiello FM, Brensinger JD, Peterson GM, et al. The use and interpretation of commercial APC gene testing for familial adenomatous polyposis. N Engl J Med 1997; 336:823–7.
39. Genetic Tests. Association of British Insurers document L.627. Dec 1997.
40. Pokorski RJ. Insurance underwriting in the genetic era. Am J Hum Genet 1997; 60:205–16.[Web of Science][Medline]
41. Convention on human rights and biomedicine. Council of Europe. European Treaty Series 164, Oviedo 1997.
42. Kevles DJ. In the name of eugenics. Harvard, Harvard University Press, 199?.
43. Mailly F, Fisher RM, Nicaud V, et al. Association between the LPL-D9N mutation in the lipoprotein lipase gene and plasma lipid traits in myocardial infarction survivors from the ECTIM study. Atherosclerosis 1996; 122:21–8.[Web of Science][Medline]
44. Fisher RM, Humphries SE, Talmud PJ. Common variation in the lipoprotein lipase gene: effects on plasma lipids and risk of atherosclerosis. Atherosclerosis 1997; 135:145–59.[Web of Science][Medline]
45. Boerwinkle E, Sing CF. The use of measured genotype information in the analysis of quantitative phenotypes in man. 111. Simultaneous estimation of the frequencies and effects of the apolipoprotein E polymorphism and residual polygenetic effects on cholesterol and triglyceride levels. Ann Hum Genet 1987; 51:211–26.[Web of Science][Medline]
46. Aird JM, Bentall HH, Roberts JAF. A relationship between cancer of the stomach and the ABO blood groups. Br Med J 1953; I:799–801.
47. McDevitt HO, Bodmer WF. HLA immune response genes and disease. Lancet 1974; I:1269–75.
48.
Gylling H, Kontula K, Koivisto UM, Miettinen HE, Miettinen TA. Influence of apoprotein E and B gene polymorphisms on LDL levels during increased cholesterol intake. Arterioscl Thromb Vasc Biol 1997; 17:38–44.
49. Ferns GAA, Shelley CS, Stocks J, Rees A, Baralle F, Galton DJ. A DNA polymorphism of the apoprotein AII gene in hypertriglyceridaemia. Hum Genet 1986; 74:302–6.[Web of Science][Medline]
50. Ferns GAA, Stocks J, Ritchie C, Galton DJ. DNA polymorphisms of the genes for apoprotein CIII and insulin in survivors of myocardial infarction. Lancet 1985; ii:300–1.
51. Ferns GAA, Galton DJ. Haplotypes of the human apoprotein A1/C111/AIV gene cluster in coronary atherosclerosis. Hum Genet 1986; 73:245–9.[Web of Science][Medline]
52. Glasse A. US and Norway `used insane Nazi-style tests'. The Times, 29.4.98.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. R. Golomb, B. P. Garg, L. E. Walsh, and L. S. Williams Perinatal stroke in baby, prothrombotic gene in mom: Does this affect maternal health insurance? Neurology, July 12, 2005; 65(1): 13 - 16. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Cushman Inherited Risk Factors for Venous Thrombosis Hematology, January 1, 2005; 2005(1): 452 - 457. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||