1 Pre-implantation genetic diagnosis: safely born but not designed
The possible uses of assisted conception technology – in vitro fertilisation (IVF) – have changed in breadth dramatically since its first intended application and clinical success in 1978: namely, to overcome infertility caused by tubal disease.1 There has been a rapid expansion in its use for the treatment of infertility, including alleviation of male infertility using intracytoplasmic sperm injection (ICSI).2 There are now over 50,000 cycles conducted each year in the UK,3 and in the USA around 150,000.4 It is estimated that over 3.5 million babies have been born worldwide following IVF.
With the increasing confidence as to its application and safety, the possibility already mooted in press coverage in the early days of IVF – that access to the embryo in vitro might provide an alternative means of preventing transmission of genetic disease – has become a reality. Proof of concept that embryos carrying X-linked diseases could be sexed, and only female embryos transferred, thus avoiding the birth of affected males, was demonstrated in 1968.5 However, it was with the advent of novel molecular technologies such as fluorescence in situ hybridisation (FISH) and amplification of deoxyribonucleic acid (DNA) using the polymerase chain reaction (PCR),6 that the breadth and potential scale of the tests that could be offered became apparent as a clinical reality.7 FISH on the interphase nucleus of a single biopsied cell allowed sexing to be undertaken by amplifying specific sequences on the Y chromosome.8 Indeed, the first two successful pregnancies achieved after using this method to test embryos coincidentally occurred during the passage of the Human Fertilisation and Embryology Act 19909 in the UK, and probably significantly influenced the discussions in the UK Parliament. For the first time parliamentarians and the public could see important medical uses for the technology, rather than relying on speculation and theory. Its successful application in these cases probably facilitated the relatively permissive legislation that exists today in the UK.10
The very power of this technology to amplify even the most minute amount of DNA, in itself produced a raft of difficulties of precision – specific and stringent precautions had to be put in place to avoid amplifying extraneous DNA, such as from the operator, in order to prevent false diagnoses. Equally, there was the inherent danger that absence of a signal due to amplification failure could be misinterpreted in diagnosis; for example, in tests for sex-linked disorders, the absence of the appearance of a Y signal could lead to the erroneous assumption that the embryo was female and suitable for transfer, which proved to be the case in the early days of sexing by PCR.11 By the mid-1990s, the alternative technique of FISH, whereby coloured fluorescent complementary DNA probes could be used to identify highly specific areas within interphase nuclear chromosomes,12 simplified the means of making a sexing diagnosis,13 and its application soon expanded to use in translocations14 and aneuploidy screening using a variety of different probes.15
The use of these powerful genetic technologies has expanded horizons, not only in being able to help further in enabling procreation, but also in areas outside of conventional reproductive biology. These advances also present new ethical and legal dilemmas.
Pre-implantation genetic diagnosis (PGD)
PGD is an early alternative to prenatal diagnosis (PND) and is currently deemed suitable for a small group of patients who are at substantial risk of conceiving a pregnancy affected by a known genetic disorder.16 To label PGD as it is presently performed as the pursuit of ‘designer babies’ is to misunderstand its methodology, and the serious and often lethal genetic diseases that can be prevented by its use.
Genetic disease may result from a new mutation in the genes that manifests as disease in an offspring for the first time in the family. More commonly it is passed on by the parents; either via one parent in a dominantly inherited condition, in which case that parent will have, or will develop the disorder, or in a recessively inherited condition where both parents will be carriers of the condition although not manifesting the disease itself. For a child to inherit the condition will require it to inherit a mutated gene from each parent, which would happen approximately one-quarter of the time. Some diseases are sex linked. Here, the mutated gene is carried on the X chromosome, which in females can be balanced by their other (normal) X and hence generally they do not manifest the disease, although they are carriers for it and can pass it on to their children. Should that mutation be inherited by a male child, where the Y chromosome does not carry information to balance it, the boy will be affected by the disease.
A number of factors will influence when and to what extent a mutation might produce disease. In some cases the effect of the mutation may not manifest until late in adult life (such as Huntington’s disease (HD)), whereas in others it may appear early in childhood (for example, spinal muscular atrophy), or even in utero. It is also possible that the effects of the mutation may be so mild as not to be recognised or, where it is picked up, to be of no concern. In others, its appearance as a disease may be so variable that the detection of a mutation can only convey a susceptibility, rather than a definitive diagnosis that it will occur (e.g. some breast cancer genes).
There is a vast list of genetic conditions caused by gene mutations, for which there is now the technology to enable diagnosis on a single cell, thus making the use of PGD a possibility. However, the application of PGD to each of these conditions may present different and often unique ethical dilemmas; for example, these will be affected by whether the disease will manifest in early death or physical and mental compromise, in which sex and at what age, whether carriers manifest symptoms, and whether there is any treatment for the disease. Besides inheritance of a single gene mutation, it may be the case that a whole chromosome is inherited incorrectly resulting in aneuploidy (number imbalance), or only part of a particular chromosome may be abnormal resulting in a chromosomal rearrangement (translocation, deletion, inversion). The effects of aneuploidy or rearrangement can be lethal, or result in identifiable syndromes causing mental and physical disability, or early or late miscarriage. Chromosome aberrations might be sporadic (de novo), or result from environmental effects or advancing age, or may be inherited; it is sufficient for one of the couple to pass on the imbalance for it to manifest. In rearrangements, the likelihood that they will be passed on and manifest as disease is highly variable, being dependent on the random nature of meiotic segregation, and the gamete involved in the fertilisation process – egg or sperm. It is generally more difficult for couples to comprehend this inheritance and the unpredictability of whether or not their offspring will be affected,17 when compared with those disorders where there is a more defined inheritance – one in four, or one in two.
For couples affected by genetic disease the choices are stark. They can play genetic roulette and take the one in four, one in two, or random chance of having a child who manifests the condition. Alternatively, they could take the chance but decide to terminate an affected pregnancy. It may also be possible to substitute one partner’s set of gametes – that is, to opt for egg or sperm donation. In some cases, the choice may be to remain childless. PGD provides a further option, meaning that intending parents can choose to have an embryo created in vitro, tested before implantation and only those embryos found to be unaffected by the disease made available for replacement into the uterus, with the intention of establishing a pregnancy.
A sample for genetic analysis may be taken from an egg, a cleavage stage embryo, or from the blastocyst – the immediate pre-implantation stage of development (day 5), where the tissues have differentiated into the inner cell mass (ICM) from which the true embryo/foetus will develop, and the surrounding trophectoderm, from which placenta and other extraembryonic tissues will form.
A single cell of the eight or so to which the embryo will have cleaved by day 3 in vitro, is easily accessible and removable after a hole is made in the shell (zona pellucida) which surrounds each embryo. This can be done by using an acidified medium or a laser. Early evidence suggests that the loss of a blastomere does not harm the embryo, as its further development is still plastic and uncommitted. However, the removal of more than one blastomere may be detrimental to the establishment of a pregnancy.18 Removal of a portion of the trophectoderm as a biopsy on day 5 has the practical advantage that fewer embryos will be available, as not all fertilised eggs progress to this stage, thus reducing the load on the laboratory. Equally, each of these embryos has better developmental potential to have reached this stage in vitro. The disadvantage is that the diagnosis will either have to be made within 24 hours in order to enable use to be made of the limited implantation window (by day 6), or the embryo will need to be frozen for subsequent replacement if that time interval cannot be met.19 Removal of the first polar body (and preferably also the second polar body) is easily effected at the time of fertilisation, and will reveal information from the maternally transmitted genome.20 It can also reveal important transmissible meiotic chromosome errors that account for the majority of sporadic aneuploidies.21
The clinical results reported for PGD vary, and depend on the type of disease being tested for, the experience of the clinic and, like IVF in general, substantially on the age of the female patient, from whom the eggs are obtained and into whom the embryo(s) are replaced.22 As might be expected, with recessive disease around 75 per cent of the embryos may be found to have a genetically transferable result (being normal or carrier). Around the same number is likely for sex-linked disorders where molecular technology is used, so allowing unaffected males also to be detected. The pregnancy rate is reduced in sex selection and in dominant disease, where around one in two embryos might be expected not to be transferable, and is even lower for reciprocal translocations where the risk is variable, but where the odds are stacked against normality. At our Centre for Preimplantation Genetic Diagnosis at Guy’s Hospital London, only 65 per cent of the embryos from reciprocal translocations are transferable. The odds of a successful pregnancy following transfer are of the same order as those when unbiopsied embryos are transferred for infertile couples using IVF, and depend largely on the quality of the embryo, itself largely dictated by how many eggs were retrieved and fertilised, thus allowing for more choice in selection of the one (or two) healthy embryos for transfer. The outcomes for children born following embryo biopsy for PGD are reassuring, as PGD does not appear to affect development.23
However, this technology has raised other more controversial uses,24 some of which are briefly summarised here. While the technology may be well established, and apparently safe, the consequences of its application remain controversial, at least for some.
HLA (human leukocyte antigen) typing of embryos is undertaken in order to find one for transfer which has a tissue match to a sibling with a life-threatening genetic or other disorder that might be ameliorated or cured by transplantation of stem cells obtained from the umbilical cord blood of the born child.25 Considerable hype has surrounded these techniques and they have been the subject of public, regulatory, parliamentary, ethical and legal debate.26 27 However their use is limited, partly because of the infrequency of the need – since in many cases matching stem cells may be available from international bone marrow banks – and the poor odds of finding an embryo match (three in eight) when PGD with tissue-typing is required. This is aggravated further by the fact that, when required, it may be by older women as they already have one or more children, and their response to ovarian stimulation is waning. In addition, if the child suffers from an acute illness such as leukaemia, there may not be sufficient time to perform the IVF, type the embryo by PGD and achieve a successful match and a live birth.28 Notwithstanding this, there have been many clinical successes which have transformed lives,29 30 but the long-term personal outcome for the family dynamic is still unexplored, as sensationalised in the novel My Sister’s Keeper.31
Social sex selection
The availability of PGD makes it possible to select embryos based on their sex. While this may be used to avoid the transmission of genetic disease, it could also be used to select embryos purely because of sex, for example for ‘family balancing’. In some families there may be a preponderance of children of one sex, where the parents might wish for a child of the other, and the possibility of sex selection through assisted conception technology is a topic of considerable debate.32 33