26 April 2023

If the EU Picks Baby Genes

Which genes should children never inherit?

This is a fundamental question. It arises acutely in medically assisted reproduction, such as in vitro fertilisation (IVF) or artificial insemination. One must frequently choose between different donors, sperm, eggs or embryos. Genetics are one way to choose. But where should the line be drawn? Some see any genetic selection as impermissible. Others argue for selecting out the most severe, purely genetic diseases, such as Huntington’s disease or cystic fibrosis. Some go further: future children should be protected from any genetic diseases or even dispositions to diseases, such as from genes increasing the risk of heart disease. And a few favour selecting for certain non-medical characteristics, such as intelligence or height.

The EU wishes to draw this line. Its proposed Substances of Human Origin Regulation (‘SoHO’, ‘the Proposal’), currently before the Parliament and Council, would do so for all medically assisted reproduction. This post argues that SoHO fails to notice the delicate ethical and political territory into which it treads. Because of this, SoHO silently takes three controversial positions on genetic selection: it excludes most ethical considerations; it draws the line blurrily and below existing standards; and it is mandatory. These three problems lead, moreover, to two legal risks: Member State derogation and Charter violation. The legislature – so far uninterested – should remedy these flaws.

Genetic selection under SoHO

SoHO aims to ensure the ‘quality and safety’ of tissues, cells and other ‘substances of human origin’ intended for human application (Art.1). This includes, for instance, blood transfusion, tissue grafts or sperm donation. Such safety standards are necessary, as was shown by the infected blood scandals in France, Britain and elsewhere in the 1970s and 1980s. Harmonised EU standards, currently provided by separate Directives on blood and tissues and cells, ensure that safety whilst enabling the free movement of such substances between Member States.

SoHO would replace and expand upon those Directives. What interests us is one of the reform’s five major objectives: to ‘ensure safety and quality for SoHO donors and for children born from donated eggs, sperm or embryos’ (Explanatory Memorandum (‘EM’), p 6). This aim is partly achieved by protecting the child from any diseases or from environmental contamination. But it is also achieved by managing the donated cells’ genetic make-up: what the Commission calls ‘genetic risk’ (recital 14 SoHO).

The existing law includes ‘very limited and outdated’ genetic screening obligations (Impact Assessment, p 22; original emphasis). ‘[B]road spectrum genetic screening is available…to verify the risk of genetic disease transmission…but is not a mandatory requirement set under EU legislation’ (ibid, p 14; compare here, Annex III § 3.6). SoHO addresses this through an overarching duty on clinics and other ‘SoHO entities’ to ‘protect the health of…offspring from medically assisted reproduction from risks posed by SoHO preparations’ by ‘identifying, minimising or eliminating’ such risks (Art.57 SoHO). ‘They shall, in particular, achieve a high level of assurance that pathogens, toxins or genetic conditions are not transmitted to…offspring from medically assisted reproduction’ (Art.58(1) SoHO).

These duties apply in respect of all such children, including those born from a couple’s own sperm or eggs without a third-party donor. But they are further specified by minimum obligations in cases involving third-party donation, such as screening and testing (Art.58(3)(b) SoHO) and treating birth with a genetic condition as a ‘serious adverse occurrence’ requiring investigation and notification to the authorities (Arts 3(28) and 35-37 SoHO; compare Arts 47(2) and 58(14)(b)).

In short, SoHO erects a binding regime seeking to prevent children from being born with a ‘genetic condition’ following medically assisted reproduction.

Much may be said in favour of genetic selection in medically assisted reproduction and of harmonised EU rules in this respect. This post does not seek to enter that broad debate. It rather argues that SoHO takes ethically controversial positions on three questions without justifying – or even acknowledging – them.

Problem one: ethics’ exclusion

The first problem is that SoHO’s does not acknowledge the ethical stakes of genetic selection in medically assisted reproduction.

It is widely recognised that genetic selection of sperm, eggs and embryos is ethically sensitive (e.g. CDBI, pp 31-34; Wilkinson and Garrard; IBC, § 54; ESHRE). Protecting the future child’s health is one interest. But other interests are in play too: the parents’ freedom of choice; the costs of screening and testing, including the risk of false positives; and the risk of eugenics. This last concept is particularly contested and historically charged (Wilkinson and Garrard, ch1), but in this context typically expresses three fears concerning attempts to improve human genetics:

  1. state or societal control over individual reproduction,
  2. selection out of genes that merely deviate from a debatable ‘genetic norm’, rather than inevitably causing serious suffering, and
  3. selection aiming to enhance or to choose non-medical characteristics (such as intelligence or eye colour).

So far from recognising these stakes, SoHO denies them. SoHO purports to set purely technical ‘quality and safety’ standards. The Proposal recognises only one value in play in genetic selection: ensuring a ‘high level of human health protection’, in this case for children born by medically assisted reproduction (Art.1 SoHO). Neither SoHO nor its Impact Assessment considers the ethical or eugenic risks this entails.

Indeed, SoHO purports to exclude ethics: ‘Member States remain responsible for decisions of an ethical and organisational nature’ and so ‘rules relating to ethical aspects…are not included in the proposal’ (EM, pp3-4). This aims to avoid debate about controversial ancillary issues concerning the uses rather than standards of cells and tissues: for instance, who may access IVF. But here, the ethical issue is not ancillary: it is raised squarely by the genetic standards themselves.

The result is not, as SoHO professes, ethical neutrality. Rather, the result is a particularly controversial ethical position adopted without argument: that the future child’s health is the only value in play in genetic selection. This leads to two further problems.

Problem two: lowering and blurring the bar

The second problem is the looseness and vagueness of the line SoHO draws: ‘genetic conditions’.

We started this post with a spectrum of possible positions: where one might draw the line in selecting genes. Existing national and international standards draw the line for sperm, egg and embryo selection at serious or grave genetic diseases (DH-BIO, p 20; ESHRE, p 4; R(90)13, Principle 2; Mukherjee, pp 456-462). This flows from the balance of interests described above: it seeks to minimise unnecessary testing and the risk of eugenic selection.

SoHO draws the line not at serious genetic conditions but at genetic conditions. On its face, this lowers the bar. True, one of the specific testing duties refers to genetic conditions presenting the ‘highest risk’ given their ‘prevalence or severity’ (Art.58(3)(b)(i) SoHO). But this only underlines that less severe (but highly prevalent) genetic conditions must also be tested for. And SoHO clinics’ overarching duty remains the minimisation of all ‘genetic conditions’ (Arts 57 and 58(1)).

What, then, is a ‘genetic condition’? Astonishingly, SoHO does not define this pivotal term. The Proposal implies that the condition must be negative or harmful: recital 14 refers to ‘harmful genetic condition’, and that recital and Art.58(3)(b) refer to ‘genetic risk’. It should presumably be interpreted broadly given its objective (the protection of the future child’s health). But beyond this, SoHO provides no clues as to its meaning.

Yet this concept is crucial, as it marks the line between ‘genetic conditions’ (which clinics must prevent) and mere deviation from the genetic norm (which they must not). SoHO leaves open the key questions: where precisely is that line? How ‘harmful’ must a condition be to warrant screening out? What about the ‘major dilemma’ of genetic dispositions (IBC, §§ 89-92)? Can such dispositions – such as DNA implying an elevated risk of breast cancer or of depression – be ‘genetic conditions’? Or is the concept limited to abnormalities which always cause a disorder, such as Down’s Syndrome?

By leaving these questions open, SoHO does not so much draw the line as hand the pen to the law’s interpreters. In this case, this means administrators and judges. ‘Genetic conditions’ is an autonomous EU concept requiring uniform interpretation and application, as it includes no reference to Member State law (e.g. C-624/20 E.K. §§ 19-20). The Court of Justice would therefore have the final say on its meaning. But SoHO also empowers the Commission to adopt ‘binding rules on the implementation’ of the relevant standards, subject to Member State approval in committee (Arts 59(1) and 79(2) SoHO). (The Commission may even add ‘additional standards’ for offspring protection by delegated acts, subject to Council or Parliament veto: Arts 58(15) and 77(2) SoHO). The contours of this ethically and politically sensitive line, deserving of broad democratic debate, are thus left to non-legislative bodies.

Administrators and judges will also have to resolve the legislation’s tension regarding within-couple medically assisted reproduction (e.g. a couple using their own sperm and eggs for IVF). SoHO entities must protect all children from ‘genetic conditions’. Yet the specific screening and other duties target cells by third parties alone. What steps must clinics take, then, to discharge their duty within couples? Are the standards lower? If so, how?

SoHO provides for a uniform European line. But it draws that line lower than existing standards, and it draws it in such broad strokes that the crucial questions are left to the executive and judiciary. Both of these moves may be justifiable. But SoHO neither acknowledges the issue nor makes the argument.

Problem three: mandatory selection

The final problem is that SoHO’s genetic selection regime is compulsory.

International standards require that genetic screening and testing in this context be voluntary (R(90)13, principles 7 and 9; Mukherjee, pp 458-459; ESHRE, pp 3-4). This respects the future parents’ autonomy. But it also reduces the risk of eugenics: the State does not oblige parents to avoid having certain kinds of children, but rather ensures that individual parents are informed and are free to select.

In contrast, SoHO is compulsory: rather than parents being informed of genetic risks and left with a choice, clinics are obliged to prevent such risks. Nothing on the face of the legislation permits parents to opt out. Again, it may be possible to justify this stance. But SoHO does not do so. Instead of fostering the necessary political and ethical deliberation, the Proposal avoids it.

Two legal risks

These three problems – ethics’ exclusion, a low and blurry bar, and compulsion – deserve debate for their own sake. But they also deserve debate because they entail two legal risks.

First, derogation. SoHO states that ethical matters are for the Member States. Indeed, it purports to allow Member States to derogate on ethical grounds (recital 16 and Art.1(1)). Let us assume that this is lawful. Where does this leave the ‘genetic condition’ line which, as shown above, is inextricable from ethics? Imagine that Member State X decides that it is unethical to select out a genetic condition included in an implementing act, or indeed to engage in any genetic selection. Will this justify derogation? If so, would the harmonised regime risk unravelling?

Second, we can restate SoHO’s first problem using rights. SoHO’s Impact Assessment (pp 103-105) mentions various rights, but casts genetic selection itself as only affecting the right of the future child to health (Art.35 Charter of Fundamental Rights). It does not consider the Charter’s prohibitions on ‘eugenic practices, in particular those aiming at the selection of persons’ in medicine and biology (Art.3(2)(b)) and genetic discrimination (Art.21(1)). Serious questions may be raised about this regime’s compatibility with these and other rights – questions that SoHO neither considers nor answers. Those questions, which concern extremely important and sensitive Charter prohibitions designed to provide bioethical protection, will likely be raised by future litigants. The Court would then be placed in an invidious position precisely because it would not have the benefit of the legislature’s attempt to interpret and accomodate these rights to assist it.


Genetic selection in medically assisted reproduction raises contentious ethical questions. By purporting to avoid these ethical issues, SoHO ends up taking contestable positions – but doing so silently. It excludes ethical concerns other than the future child’s health; lowers the bar from ‘serious’ to any genetic condition and does not define this; and is mandatory. This opens up legal risks: Member State ethical derogation and restriction or invalidity for breaching the Charter’s provisions, including its prohibition on eugenic practices.

During the passage of the original Directive on tissues and cells, Parliament’s ENVI Committee proposed clarifying that ‘[g]enetic characteristics of germ cells are not to be considered as quality characteristics’, noting that this ‘can lead to eugenics’ (Amendment 9). Two decades later, faced with a regime that wholeheartedly adopts this model, the Council and Parliament have thus far proven uninterested (see Council and draft EP Committee report). But amongst the 757 draft amendments to the Committee report are a small number that begin to grapple with the difficulties (inter alia 118, 357, 384, 682, 710 and 711). These include proposals to subject the regime to Member State law and to grant parents the right to refuse (amendments 682 and 711). These may, at last, spark the bioethical debate that this legislation deserves.

The author would like to thank Dr Peter Oliver, Professor Shaun D. Pattinson and the editorial team of Verfassungsblog for their comments on an earlier version of this post. The usual disclaimer applies.




SUGGESTED CITATION  Coghlan, Niall: If the EU Picks Baby Genes, VerfBlog, 2023/4/26, https://verfassungsblog.de/if-the-eu-picks-baby-genes/, DOI: 10.17176/20230426-084432-0.

Leave A Comment


1. We welcome your comments but you do so as our guest. Please note that we will exercise our property rights to make sure that Verfassungsblog remains a safe and attractive place for everyone. Your comment will not appear immediately but will be moderated by us. Just as with posts, we make a choice. That means not all submitted comments will be published.

2. We expect comments to be matter-of-fact, on-topic and free of sarcasm, innuendo and ad personam arguments.

3. Racist, sexist and otherwise discriminatory comments will not be published.

4. Comments under pseudonym are allowed but a valid email address is obligatory. The use of more than one pseudonym is not allowed.

Explore posts related to this:
Bioethik, SoHO, biolaw, biopolitics, genetic

Other posts about this region: