4 Statutory regulation of PGD in the UK: unintended consequences and future challenges
Until 2009, pre-implantation genetic diagnosis (PGD) in the UK was regulated by the Human Fertilisation and Embryology Authority (HFEA), through its Code of Practice and its licensing and inspection regime. The statute which set up the HFEA and lays out the circumstances in which fertility treatment, carried out under licence, is lawful in the UK – the Human Fertilisation and Embryology Act 1990 – did not mention PGD, although it was assumed that it was nevertheless a licensable treatment. In part as a result of its absence from the statutory scheme, the scope of the HFEA’s powers was subject to a legal challenge which culminated in the House of Lords confirming that the 1990 Act gave the Authority very broad discretion when formulating guidance on the circumstances in which PGD could be provided. The 2008 amendments to the 1990 Act, which came into force on 1 October 2009, now set out, on the face of the statute, the criteria which the HFEA must use when deciding whether to vary a clinic’s licence to allow it to provide PGD for a particular condition. Since the statutory criteria are broadly similar to the most recent pre-2009 HFEA guidance, it could be argued that this is a change of form, rather than substance. This chapter will argue, to the contrary, that far from simply being a procedural technicality, the decision to codify the regulation of PGD has had and will continue to have some potentially far-reaching consequences.
A very brief history of the regulation of PGD in the UK
Although the pioneers of in vitro fertilisation (IVF) had realised that it might be possible to biopsy cells from embryos as early as the late 1960s, the first successful cycle of PGD did not take place until 1989, at the Hammersmith Hospital in London. The Warnock Committee, which was appointed in 1982 to consider IVF and embryo research and to make recommendations, did not devote much space to PGD in its report, believing its clinical application to be some way off. The Warnock Report, published in 1984, described the use of embryo biopsy for detecting abnormalities before implantation as a potentially useful technique, particularly since it could ‘avoid … the difficult decision for the parents of whether to seek a termination where abnormality is detected’.1 However, its conclusion was that, ‘given the present relatively low success rates for pregnancy following IVF, it is unlikely that embryonic biopsy will become a feasible method of detecting abnormal embryos for some considerable time’.2
Although the 1990 Act itself does not mention PGD, for two reasons it has generally been assumed that Parliament’s intention in 1990 was to allow PGD to be carried out under the licensing regime introduced by the Human Fertilisation and Embryology Act. First, when the Human Fertilisation and Embryology Bill was introduced, in the 1989/90 session, the fact that IVF might be used not only to help infertile couples to have children, but also to prevent serious disease is believed to have been a factor which greatly assisted its passage through Parliament.3 The pictures of smiling pregnant women, whose embryos had been sexed in order to ensure that they did not give birth to boys who might have inherited the X-linked diseases which ran in their families, is widely credited with giving a significant boost to the legislation.4
Second, one of the specified research purposes under the Act for which a licence to carry out embryo research may be issued was to ‘improve methods of detecting genetic and chromosomal abnormalities in embryos’.5 If research to improve PGD techniques can be licensable, it would be extraordinary if the technique itself was a criminal offence (as anything done to an embryo without a licence is under the legislation).6 The statute’s silence on the circumstances in which PGD could be carried out, coupled with the assumption that licensing it fell within the HFEA’s statutory powers, meant that it was left to the HFEA to devise its own rules through successive editions of its Code of Practice.
In the first ten years of the HFEA’s existence, licensing PGD to enable couples to avoid passing on very serious genetic conditions to their offspring proved to be relatively uncontroversial. It was only when a new use for PGD techniques emerged at the start of the twenty-first century that the scope of the HFEA’s licensing powers was challenged. Tissue or HLA (human leukocyte antigen) typing involves taking a cell from an early embryo, in the same way as in ordinary PGD, and testing it to see if the resulting child would be a good tissue match for a sick sibling in need of a bone marrow transplant. If the selected embryo is a good tissue match, when the baby is born, blood taken from her umbilical cord can be used to treat the older sibling.
When it first considered the issue in 2001, the HFEA initially decided that tissue-typing could be legitimate in certain circumstances; in particular, the child to be born had to be at risk of suffering from the same genetic condition as its older sibling.7 In the extensive media coverage that this issue attracted, a distinction was then drawn between two couples, each of whom wished to use PGD with HLA-typing in order to treat their seriously ill son.8 The Hashmis’ son Zain was suffering from beta-thalassaemia major, and any future child they might have might also be at risk of suffering from the same condition. This meant that the Hashmis might legitimately use PGD without tissue-typing, in order to avoid having a child with the same condition as Zain, and according to the HFEA, carrying out an additional test on the biopsied cell – in order to find out whether the embryo would be a good tissue match for Zain – would be acceptable. In contrast, the Whitakers’ son Charlie suffered from Diamond–Blackfan anaemia, which is not an inherited condition. Given the HFEA’s requirement that HLA-typing could only be carried out if the couple could also legitimately access PGD to screen out the older child’s condition, the Whitakers were not able to access tissue-typing in the UK. They therefore decided to have treatment in Chicago, and subsequently gave birth to a son who proved to be a good tissue match for Charlie.
The reason for drawing this distinction between the Hashmis and the Whitakers was said to be that – in the case of Zain – embryo biopsy would be carried out to benefit the child to be born as well as to benefit his or her older brother, whereas in the case of Charlie Whitaker, the embryo biopsy would be done solely to benefit him, and not to benefit the child to be born. In practice, however, because PGD does not treat the inherited genetic condition, the only way it ‘benefits’ the future child is by enabling the latter to be selected and hence born, and of course this ‘benefit’ would exist for both Zain and Charlie’s younger sibling. Subsequently, and in my view rightly, the HFEA abandoned this rather dubious distinction, and permitted PGD with HLA-typing for both inherited and non-inherited conditions, including Diamond–Blackfan anaemia.9
Following the HFEA’s 2001 decision to allow PGD accompanied by HLA-typing, its power to do so was challenged by a ‘pro-life’ pressure group, Comment on Reproductive Ethics (CORE). Schedule 2 of the 1990 Act specified that a licence ‘cannot authorise any activity unless it appears to the Authority to be necessary or desirable for the purpose of providing treatment services’. ‘Treatment services’ were defined in section 2 as ‘medical, surgical or obstetric services’ provided ‘for the purpose of assisting women to carry children’. CORE argued that since PGD and HLA-typing were not done in order to help women to conceive, they did not fall within section 2’s definition of treatment services, and hence could not be licensed by the Authority. At first instance, Maurice Kay J. agreed with CORE: since HLA-typing was not necessary in order to help a woman to bear a child, it did not fall within the statutory definition of treatment services, and could not therefore be licensed by the HFEA. Of course, this is also true for ordinary PGD: it is not necessary in order for women to conceive, and hence if this decision had not been successfully appealed, it would have prevented the HFEA not only from licensing HLA-typing, but also from licensing any PGD treatment. On appeal, the Court of Appeal unanimously reversed his decision, taking a much more purposive interpretative approach, and its decision was upheld, again unanimously, by the House of Lords.10
The House of Lords drew attention to Schedule 2 of the Act, which listed a number of activities that fell within the definition of ‘services provided for the purpose of assisting women to carry children’; in particular paragraph 1(1)(d) referred to ‘practices designed to secure that embryos are in a suitable condition to be placed in a woman or to determine whether embryos are suitable for that purpose’. The House of Lords found that the licensing power of the Authority was ‘defined in broad terms’, and it was therefore entitled to construe the concept of suitability broadly, to include whether the embryos were suitable ‘for the purposes of the particular mother’. This could include ‘determining the genetic characteristics of the embryo by way of PGD or HLA typing’, and the Authority was ‘entitled to take the view’ that PGD or HLA-typing was ‘necessary or desirable’ for the purpose of providing such treatment services.11
There was a gradual evolution of the HFEA’s criteria for licensing PGD from the First to the Seventh Code of Practice. For example, the Code used to specifically equate PGD with ‘current practice in the use of (post-implantation) prenatal diagnosis (PND)’.12 The intention was to equate the circumstances in which it is acceptable to carry out PGD with the ‘foetal abnormality’ ground for abortion in the Abortion Act 1967, namely that abortion is lawful where there is ‘a substantial risk that, if the child were born, it would suffer from such physical or mental abnormalities as to be seriously handicapped’.13 Translating this to apply to PGD, the Sixth Code specified that PGD ‘will be available only where there is a significant risk of a serious genetic condition’.14
The abortion model was imperfect for a number of reasons, however. First, it was not entirely clear whether the Code was equating PGD with the circumstances in which prenatal diagnosis can be carried out or whether it was equating it with the criteria for the lawful termination of pregnancy. The Code appeared to elide the two, but they are not the same thing. The decision to carry out PND is a matter of clinical discretion, whereas termination of pregnancy can lawfully be carried out only for one of the four statutory grounds in the Abortion Act 1967. Ultrasound – while commonly regarded as the first chance to see one’s baby – is a prenatal diagnostic technique, and although practice varies, it is not uncommon for pregnant women having the standard 20-week anomaly scan to be told the sex of their baby. Since sex selection, other than for medical reasons, was not permitted by the HFEA, its intention in the Sixth Code of Practice was plainly not to equate current practice in the use of ultrasonography with PGD.
Second, it is possible for a woman to be denied access to abortion under the foetal abnormality ground if two doctors, acting in good faith, believe that the foetus’s abnormality is insufficiently serious. A woman can, in theory at least, be forced to carry a pregnancy to term if the foetus’s condition is not serious and she cannot avail herself of one of the other statutory grounds. It is, however, hard to see how a woman could be forced to have an embryo transferred. In short, in relation to PND, there are no formal limits on testing; rather the restrictions relate to the criteria which must be satisfied in order to lawfully terminate an established pregnancy. In contrast, in relation to PGD, the HFEA did not intend to restrict the woman’s right to refuse embryo transfer, but rather to restrict what tests could be carried out in the first place.
In addition to the target of the restrictions being different in each case, there was already some internal inconsistency in the HFEA’s rules. Tissue-typing is not done to avoid a serious genetic condition in the embryo, and hence fell outside the wording in the Sixth Code. Given these multiple pressures on the abortion analogy, it was abandoned in the Seventh Code, which instead set out specific criteria to cover both standard PGD and tissue-typing.
The non-PND linked criteria in the Seventh Code concentrated on the need for the child to be born to be at substantial risk of suffering from a serious genetic condition, or, in the case of tissue-typing, for the affected sibling’s condition to be serious. Sex selection would be permitted only where there was a risk of a serious sex-linked disease. In all three cases then, PGD could be licensed only if the relevant genetic condition was judged to be ‘serious’, making it important to have a clear definition of what seriousness means.
Many genetic conditions self-evidently meet any threshold level of seriousness. If a child is born with Tay–Sachs disease, for example, her nervous system will start to degenerate during her first year of life, and she will die within three or four years. Tay–Sachs is undoubtedly a serious condition, and it is readily understandable why parents who are at risk of having a child with Tay–Sachs disease – many of whom will have already had a child with this condition – would want to use PGD to try to ensure that their future children do not die in infancy. At other times, however, drawing a line between serious and non-serious conditions may be more difficult. In particular, is seriousness an objective clinical fact, or does it depend upon the subjective views and experience of the individual family? The HFEA, through its Eighth Code of Practice, has taken a mixed approach: some of the criteria – such as ‘the availability of effective therapy’ and ‘the speed of degeneration in progressive disorders’ – can be judged objectively, whereas others – such as ‘the views of the people seeking treatment in relation to the condition to be avoided, including their previous reproductive experience’ and their ‘family circumstances’ – are family-specific and hence variable.15 And it is clear that this mixed approach reflects the reality of PGD in clinical practice, where the previous reproductive experience of the people seeking treatment – usually of having an existing child with the condition – is commonly what prompts interest in PGD and this will, in addition to more objective consideration of the condition’s seriousness, be taken very seriously by clinic staff when deciding whether PGD is appropriate.16 This comes across very clearly in Scott et al.’s interviews with healthcare professionals working in licensed centres.
I saw a couple last week who came for Cystic Fibrosis, a fertile, intelligent couple, who have a Cystic Fibrosis child. And I said, ‘What are you doing this for? Why don’t you just have another pregnancy?’ And they couldn’t consider terminating a Cystic child because, firstly they said, ‘[W]e do not want to have another child that we have to watch die or be very ill. But on the other hand, if we kind of [terminate the] pregnancy it’s like terminating [our existing child] … And we feel we can’t do that. And we want some other way of approaching this.’17
Of course, it should be remembered that the rules contained in the HFEA’s Code of Practice are directed towards clinics, rather than the HFEA itself. It might then be argued that the criteria contained in the successive editions of the Code were simply instructing centres when they could legitimately offer PGD to a particular couple. However, given that the statute itself did not set out when the HFEA could legitimately issue a licence for PGD, the HFEA had decided to use the criteria set out in the Code itself when determining whether a licence should be varied to permit PGD. Hence, the criteria in the Code were, in practice, applied twice: first by the HFEA when deciding whether to vary a clinic’s licence to include PGD for a particular condition, and then by the clinic, when determining whether PGD was appropriate for a particular couple.
In sum, then, before the 2008 reforms to the 1990 Act, the HFEA had reserved to itself the power to decide when a clinic could obtain a licence to carry out PGD – which it did via a mixed objective–subjective ‘seriousness’ test – and its broad discretion to do so had received the unanimous support of the House of Lords.
The process of reforming the 1990 Act began with the House of Commons Science and Technology Select Committee’s decision to embark on a review of the 1990 Act, in the light of technological changes and shifts in ethical and social attitudes. Its report, Human Reproductive Technologies and the Law, was published in 2005. In relation to PGD, the Committee found that:
[t]he HFEA’s room for manoeuvre can be seen as both its major strength and its biggest weakness. We have seen that by giving the HFEA jurisdiction over what is a ‘suitable condition’ for an embryo to be transferred, it has been able to license PGD despite legal challenge. Professor Kenyon Mason told the committee that ‘the very real problem is that there is too much in this act that is not in the act but is in the Code of Practice’.18
The then Chair of the Authority, Suzi Leather, agreed that greater clarity was desirable, and told the Committee that she would like to see the permitted uses of PGD set out in legislation, in a similar way to the research purposes which set out a strictly limited set of purposes for which the HFEA is permitted to issue licences to use embryos in research.
The Science and Technology Committee concluded that it had ‘concerns about the criteria imposed by the HFEA’.19 Given the inherent limitations of PGD – in particular that ‘it cannot be used to create ‘designer babies’ – the Committee saw ‘no reason why a regulator should seek to determine which disorders can be screened out using PGD’.20