The nature of diagnosis

First there is a question of the dividing line between testing and screening at the fetal stage. We shall assume that the testing of a fetus where there is some specific reason to do so is different from routine fetal screening. Either might result in identification of a disorder. Diagnosis is not a simple matter of fact, however, or even a complicated matter of fact, it has ethical aspects. This is partly connected with questions of uncertainty. Issues here surround false positives and false negatives, and also prediction. Where genetics is concerned, in general, and in relation to the fetus in particular, diagnosis may mean the detection of an existing disorder or the prediction of a disorder to come, with all the surrounding issues of probability, penetrance and severity.

Techniques

The ethics of the technique of diagnosis relate primarily to safety, which in turn is related to degrees of invasiveness. In the 1970s, the development of real-time ultrasound imagery reported in-utero diagnosis of congenital anomalies. Prenatal diagnosis techniques, such as amniocentesis (with associated risks) and chorionic villus sampling, followed by more recent non-invasive techniques, such as cell-free fetal DNA, have come to be deployed routinely as screening procedures, rather than individual testing. This is because more than 90% of structural and chromosomal abnormalities arise in pregnancies without any obvious risk factors. The evolution of genomics, however, has introduced prenatal testing and pre-implantation genetic diagnosis techniques, which have led to conditions being diagnosed earlier and more easily than before.

What and why?

The premise upon which the practice of fetal screening rests is that to identify disorders and potential disorders is a worthwhile objective and opens up possible options for action. But the question of what counts as a disorder is itself not devoid of ethical content: in addition to challenges from disability rights groups, a difference exists between a single-gene disorder, such as cystic fibrosis, and a multifactorial one; there is a difference between those that are late onset (Huntington’s) and those that are not. There is a difference between carrier status and affected individuals.

Some ethical issues are associated with the accuracy of any given test result and also what it means. To say that a genetic disorder has been detected, is in itself not particularly helpful. This is where the importance of counselling can be seen. The counsellee may need to know how severely it will affect a future child or adult, and the probability of its developing at all. Differences in perceptions of risk further complicate the ethical issues.

The ‘why’ question may be dependent upon context. In a society with a high prevalence of a particular disorder (e.g. as has been the case historically in Cyprus), fetal screening may be a significant element in the repertoire of measures to deal with this. The Wilson–Jungner principles of screening stated that, in order to justify a screening programme, the condition sought must be an important problem and there should be some treatment intervention that could be offered in the event of a positive result. Putting the issue of false positives aside, difficult questions clearly need to be addressed. What counts as an important problem? In the 1990s, it was debated whether scientists should pursue research into the genetic basis of homosexuality. The argument against this was precisely that it could lead to the termination of fetuses on the basis of the identification of susceptibility. Genomic testing has the potential to create more accurate and earlier diagnosis of conditions that may occur, in relation to health status and behavioural traits.

When we turn to the question of intervention, some critics might point to the fact that often the ‘treatment’ on offer is a termination. This is not always the case, however, and the range of options is increasing. Although the development of prenatal genetic diagnosis could result in positive tests leading to an increase in abortions, accurate diagnosis of a fetal anomaly allows appropriate counselling and transfer to a tertiary unit, planned delivery, and specialised neonatal therapy.

The criteria for screening may seem to rest on an assumption that prenatal diagnosis is something that has to be justified. An alternative ethical position is that diagnosis of genetic disorders in the fetus is a moral duty. The argument from procreative beneficence, mentioned above, suggests that we ought to have the best children we can. It is used to argue for prenatal diagnosis and also for pre-implantation genetic diagnosis.

This argument is open to criticism from those who think that trying to ensure the best possible children is wrong in principle, perhaps because it is regarded as turning children into design objects and because ‘best’, where children are concerned, is fraught with shifting boundaries of definition, just as ‘disorder’ is.

Management of the information

When we turn to management of the information, we are at the boundary line between diagnosis and management of the disorder itself. From a diagnostic perspective, the issues are what and how genetic information is communicated and to whom, who has a right to know or not to know, and how the information might be used, perceived, or misused. Issues more clearly in the category of management of the disorder include counselling, treatment options, or lifestyle adjustment.

Rights to know and not to know

The first major issue concerns what is disclosed and to whom. In the case of genetic testing and screening of adults, considerable discussion has taken place of rights to know and not to know. First, in relation to knowledge about oneself, there is a view that genetic information can be empowering, in that it can facilitate lifestyle choices that may safeguard ones health. On the other hand, it has been argued that genetic information can disrupt one’s picture of one’s life story, which can have adverse consequences: hence the argument for a right not to know.

The nature of diagnosis

First there is a question of the dividing line between testing and screening at the fetal stage. We shall assume that the testing of a fetus where there is some specific reason to do so is different from routine fetal screening. Either might result in identification of a disorder. Diagnosis is not a simple matter of fact, however, or even a complicated matter of fact, it has ethical aspects. This is partly connected with questions of uncertainty. Issues here surround false positives and false negatives, and also prediction. Where genetics is concerned, in general, and in relation to the fetus in particular, diagnosis may mean the detection of an existing disorder or the prediction of a disorder to come, with all the surrounding issues of probability, penetrance and severity.

Techniques

The ethics of the technique of diagnosis relate primarily to safety, which in turn is related to degrees of invasiveness. In the 1970s, the development of real-time ultrasound imagery reported in-utero diagnosis of congenital anomalies. Prenatal diagnosis techniques, such as amniocentesis (with associated risks) and chorionic villus sampling, followed by more recent non-invasive techniques, such as cell-free fetal DNA, have come to be deployed routinely as screening procedures, rather than individual testing. This is because more than 90% of structural and chromosomal abnormalities arise in pregnancies without any obvious risk factors. The evolution of genomics, however, has introduced prenatal testing and pre-implantation genetic diagnosis techniques, which have led to conditions being diagnosed earlier and more easily than before.

What and why?

The premise upon which the practice of fetal screening rests is that to identify disorders and potential disorders is a worthwhile objective and opens up possible options for action. But the question of what counts as a disorder is itself not devoid of ethical content: in addition to challenges from disability rights groups, a difference exists between a single-gene disorder, such as cystic fibrosis, and a multifactorial one; there is a difference between those that are late onset (Huntington’s) and those that are not. There is a difference between carrier status and affected individuals.

Some ethical issues are associated with the accuracy of any given test result and also what it means. To say that a genetic disorder has been detected, is in itself not particularly helpful. This is where the importance of counselling can be seen. The counsellee may need to know how severely it will affect a future child or adult, and the probability of its developing at all. Differences in perceptions of risk further complicate the ethical issues.

The ‘why’ question may be dependent upon context. In a society with a high prevalence of a particular disorder (e.g. as has been the case historically in Cyprus), fetal screening may be a significant element in the repertoire of measures to deal with this. The Wilson–Jungner principles of screening stated that, in order to justify a screening programme, the condition sought must be an important problem and there should be some treatment intervention that could be offered in the event of a positive result. Putting the issue of false positives aside, difficult questions clearly need to be addressed. What counts as an important problem? In the 1990s, it was debated whether scientists should pursue research into the genetic basis of homosexuality. The argument against this was precisely that it could lead to the termination of fetuses on the basis of the identification of susceptibility. Genomic testing has the potential to create more accurate and earlier diagnosis of conditions that may occur, in relation to health status and behavioural traits.

When we turn to the question of intervention, some critics might point to the fact that often the ‘treatment’ on offer is a termination. This is not always the case, however, and the range of options is increasing. Although the development of prenatal genetic diagnosis could result in positive tests leading to an increase in abortions, accurate diagnosis of a fetal anomaly allows appropriate counselling and transfer to a tertiary unit, planned delivery, and specialised neonatal therapy.

The criteria for screening may seem to rest on an assumption that prenatal diagnosis is something that has to be justified. An alternative ethical position is that diagnosis of genetic disorders in the fetus is a moral duty. The argument from procreative beneficence, mentioned above, suggests that we ought to have the best children we can. It is used to argue for prenatal diagnosis and also for pre-implantation genetic diagnosis.

This argument is open to criticism from those who think that trying to ensure the best possible children is wrong in principle, perhaps because it is regarded as turning children into design objects and because ‘best’, where children are concerned, is fraught with shifting boundaries of definition, just as ‘disorder’ is.

Management of the information

When we turn to management of the information, we are at the boundary line between diagnosis and management of the disorder itself. From a diagnostic perspective, the issues are what and how genetic information is communicated and to whom, who has a right to know or not to know, and how the information might be used, perceived, or misused. Issues more clearly in the category of management of the disorder include counselling, treatment options, or lifestyle adjustment.

Rights to know and not to know

The first major issue concerns what is disclosed and to whom. In the case of genetic testing and screening of adults, considerable discussion has taken place of rights to know and not to know. First, in relation to knowledge about oneself, there is a view that genetic information can be empowering, in that it can facilitate lifestyle choices that may safeguard ones health. On the other hand, it has been argued that genetic information can disrupt one’s picture of one’s life story, which can have adverse consequences: hence the argument for a right not to know.

Management of information: disclosure and counselling

The ethos of genetic counselling has long been associated with non-directiveness. This followed disquiet about purported abuses of genetics in the first half of the twentieth century and also a greater emphasis, in medicine generally, on the principle of autonomy. Discussion about the extent to which non-directiveness is possible have of course been plentiful, in light of the potentially vulnerable position of a pregnant woman in the counselling encounter with healthcare professionals; a situation where termination is on the table as a possible outcome.

A woman may, of course, prefer not to know whether the fetus is suffering from a genetic disorder, especially in light of potential uncertainties about penetrance and severity. Second, let us suppose that she does want to know, but that it is not so severe that she, with or without a partner’s input, wants to opt for a termination. How might the knowledge affect how the future child is regarded? This is likely to be dependent upon what the condition in question is. If it is one that can be ameliorated by some lifestyle intervention or treatment, then it might be beneficial. If it is not, there may be dangers of stigmatisation at one end of the spectrum to over-protection at the other. The issue is also whether, and if so when, a child has the right to know of a genetic diagnosis.

A diagnosis of carrier status has its own issues. The extent to which carrier status could be regarded as sufficient reason for a termination would depend on certain background conditions obtaining, for example in societies with very high prevalence of a recessive condition of a life-threatening sort. It is difficult to think of a case where this would be the optimal solution but scope exists, however, for misinterpretation of carrier status.

Another ethical issue also concerns incidental findings: for example, if during the diagnostic process, information emerges that is relevant to paternity, which the woman may not wish to be disclosed.

With reference to all these issues, the central question is whose interests have to be taken in to account, and how is it possible to prioritise? This will be discussed below.

Management of the disorder: therapy

Once issues concerning disclosure of the information have been addressed, the next question relates to the options for action.

In medicine in general, apart from some minority positions, it is generally presumed that to treat illness where one can is a good thing. So a fortiori , to treat genetic disorder where one can is a good thing. Two complicating factors can be added to this general statement.

The first is that it is, of course, problematic when the only ‘treatment’ possible is a termination. Technological advance makes possible other options, but also introduces new problems.

Termination

The first point to make here is that it may be preferable to speak of options for action rather than options for treatment. Even if there is no treatment option available, termination may not be the only option for action. Information may enable a woman to prepare for the birth of a child with a genetic disorder, and this may be the optimal outcome in some cases.

Nevertheless, it is fair to say that the option of termination brings the question of the status of the fetus centre stage. For some, the ethics of termination depends on that question. For others, it does not: one view is that, even if the fetus is a person (and not just a human being), termination is permissible because either the freedom of choice of the mother is, or should be, overriding, or because, all things considered, it is in the best interests of all (e.g. in the case of ‘serious’ disorder), however that is defined. Once we go down the road of trying to assess overall consequences, and the more we tolerate the deliberate ending of life, even fetal life, in society, we may have to take into account not only individual case consequences, but also the repercussions on respect for life in general. The setting of time limits in legislation is a middle way, from an ethical point of view, recognising that, as fetuses develop, they come to have interests increasingly demanding of consideration and that it is a more serious matter to end their lives. Current legislation in England and Wales, however, allows termination without limit of time in cases where the disorder is serious.

Fetal gene therapy

What are the other alternatives to termination in cases of genetic disorder, apart from proceeding to the birth of a child with disorder? The answer is some form of treatment. Classically, fetal therapies fall into four categories: transplacental drug treatment, gene therapy, invasive procedures, and fetal surgery. Although any fetal intervention has to deal, ethically, with the degree to which the fetus can be considered a patient, the prospect of fetal gene therapy raises additional issues.

Gene therapy is a medical technique that seeks to deliver genetic material to a cell, in order to generate a therapeutic effect by correcting an existing abnormality within the host cell. Gene therapy delivers the wanted genes through viral vectors such as retroviruses and adenoviruses. More recently, human immunodeficiency virus has been used as a viral vector, but a shift has taken place towards using non-viral vectors. The first use of fetal gene therapy was recorded in 1999 with the first successful therapeutic application of gene transfer in utero occurring in 2003. Fetal gene therapy is used to benefit a fetus: the fetus does not have to be the direct recipient of the gene therapy because, in certain circumstances, the recipient of the gene therapy will be the pregnant woman for the benefit of her fetus.

For an ethical assessment, the following questions will need to be answered: is the disorder in question amenable to treatment? Would the treatment be preferable to termination or non-intervention? Does the clinician feel that, after reviewing considerations such as risk–benefit calculations, the chance of ethical approval, whether the pregnant woman is likely to consent, that suggesting a fetal gene therapy approach is a viable option? Would the intervention occur at the correct time? These questions are not an exhaustive list but merely indicate the complex questions that must be asked.

Safety considerations with gene therapy in adults have produced some disappointments and setbacks scientifically and ethically, and the early promise associated with it has been replaced by a far more cautious approach. Fetal to gene therapy has emerged as a field in its own right, partly because of the perception that the ‘road blocks’ to gene therapy in general could perhaps be avoided through a fetal approach. Early signs indicated that fetal gene therapy could be successful where adult gene therapy could not. In 1990, Hatzoglou et al. used an in-utero approach in applying a replication-incompetent retrovirus vector to fetal rats in-utero. Their work showed that it mediated expression of human growth factor significantly more effectively than the application of the same vector within an adult animal model. The success of this work led to a large number of successful in-utero gene experiments, which demonstrated gene delivery to virtually all fetal tissues.

The delivery of foreign protein or therapy to the fetus can take advantage of immune tolerance, whereas, in relation to postnatal therapies, the development of gene therapy hits road blocks, because of issues of immune response and engraftment of genes into the hosts. There is a ‘window of opportunity’, which makes genetic inventions through fetal gene therapy a distinct possibility: the immune system of the fetus has not fully developed, thereby in theory eradicating the problems that babies after birth suffer because they have developed immune systems. Long-term tolerance of the engrafted transgene is induced prenatally.

Significant progress has been made in the past decade, showing promising signs in haemophilia disorders, cystic fibrosis, muscular dystrophy and central nervous system disorders. Coutelle et al. have a high incidence rate of liver tumours as a result of in-utero gene therapy when using an SIN configuration, but not when using a HIV backbone in mice, despite in-utero gene therapy being, in principle, a possible successful treatment for several conditions, including haemophilia B. Other concerns include mutagenesis, germ-line transmissions and disruption of normal development. The prospects of germ-line therapy has always been regarded as particularly problematic. Even if the gene intervention is not specifically targeted at the germ line, which develops and is compartmentalised by week 7 of gestation, the germ line could potentially be affected as a side-effect.

Within the UK, The Gene Therapy Advisory Committee conducted its own review of the potential of prenatal (fetal) gene therapy. Under its New and Emerging Technologies sub-group, it took the core principles used by the Clothier Committee, which sought to look into the issues and ethics in gene therapy in general. These core principles were as follows: (1) gene therapy is research and not innovative treatment; (2) only somatic therapy should be considered; (3) in view of safety and ethical difficulties, germ-line interventions are off limits at present; (4) gene therapy should be restricted to life-threatening disorders where no current alternative effective treatments are available; (5) individuals should participate in gene-therapy research trials only after a full explanation of the procedures, risks and benefits, and after they have given their informed consent, if they are capable of doing so; and (6) recognising that some people, including young children, may not be able to give such consent, so therapeutic research involving such patients must not put them at disproportionate risk.

The Gene Therapy Advisory Committee reviewed these core principles and the Clothier Committee report and concluded that no new ethical issues were raised by in-utero interventions or the use of gene therapy in other situations.

The implication of principle (point 4), however, is that a target disorder or disease would need to be life threatening, or associated with severe disability, with no suitable treatment available after birth, in order to justify intervention in utero. Therefore, recent advances in adult gene therapy for haemophilia B would mean that an in-utero approach would not be permitted because of the relative success and progress of a postnatal approach.

Fetus as patient

Because the beneficiary of the treatment would be a fetus, any treatment in pregnancy is clouded by the historical problems associated with experimental medicine in pregnancy such as the Thalidomide and Epilim incidents. ^c

c See (BBC 2002) for details on the thalidomide incident that resulted in birth defects between 1957 and 1961; and (The Guardian Editorial) for details regarding the soon to be dropped Epilim case where the users of the epilepsy drug ‘Epilim’ during pregnancy alleged that the drug caused the birth defects in their children.