Page URL:

Response to the Human Fertilisation and Embryology Authority's Survey Embryo Testing: Testing for More than One Condition/Abnormality at a Time

18 January 2016
This policy document is a response submitted by the Progress Educational Trust (PET) to the Human Fertilisation and Embryology Authority (HFEA)'s Survey Embryo Testing: Testing for More than One Condition/Abnormality at a Time.
The survey was circulated via the HFEA's PGD Update email alert service.

It is very difficult to answer the questions put forward for consideration by the HFEA. To ask and answer such questions assumes a clear understanding of the relevant technologies, of the reasons for their use, and of their effectiveness in answering specific clinical questions. As we discuss below, the content of the HFEA's email suggests a limited understanding of these issues.
We are also uncertain of the status of and context for this email, as we have spoken to several people who work in this area who are unaware of the email. There has been little time to formulate a response, due to the Christmas break and due to the Association of Clinical Embryologists and British Fertility Society conferences taking place on 5-8 January 2016. Are changes in policy to be based on responses to this email and the legal advice received by the HFEA, or will there be wider consultation?
There are a few general points we think the HFEA should keep in mind when considering the scenarios discussed in its email.
The fertility sector is highly commercialised, and has a track record of using technologies in a clinical setting before they have a robust evidence base.
Establishing the validity and efficacy of technologies is especially challenging where preimplantation genetic screening (PGS) is concerned, as opinion is divided in the sector and many of the relevant experts have conflicts of interest.
Genetics is a particularly complex and fast-moving area of medicine, and clinics offering preimplantation genetic diagnosis (PGD) and PGS need adequately trained staff to consider the implications of the tests they are carrying out. Any patient counselling must be carried out by a qualified genetics counsellor.
The scenarios described in the HFEA's email do not appear to proceed from a clear understanding of the relevant technologies. For example, the email mentions Huntington's disease (HD).
There are already many policies that govern predictive testing for HD, including not testing children. Furthermore, HD is caused by a triplet repeat expansion that will not be detected by array technology or (at present) by sequencing. HD is therefore of little relevance to the questions the HFEA asks.
This is an example of lack of understanding of the technical limitations of this technology. Simple consent processes are meaningless unless the individuals taking consent understand the limitations as well as the possibilities of the relevant technologies. This is one of the reasons why any patient counselling about the use and implications of these technologies must be carried out by a qualified genetics counsellor.
A number of useful guidelines, policies and position statements already exist on incidental findings, opportunistic genomic screening and gene sequencing, which provide a good background to the complexity of this area. These include:
The Position Statement on Opportunistic Genomic Screening issued by the Association of Genetic Nurses and Counsellors
Numerous reports and briefing notes in the Clinical Genome Analysis section of the Foundation for Genomics and Population Health website
The 'Policy for Pertinent, Secondary and Incidental Findings, and Feeding Back to Clinicians and Patients' in the 100,000 Genomes Project Protocol (.pdf 1MB), p42-50
The priority of the HFEA should be to address the effectiveness of these technologies, and consider where its legal obligations lie.
Speaking broadly, embryo testing can serve two main purposes. These are often conflated, but are in fact very different.

1. To improve the chance of implantation
The main question for the HFEA to consider is the effectiveness of a technology in achieving this. There have been many years of heated debate about PGS and its effectiveness, and it is not clear to us that there is any evidence for increasing the number of genetic variations that are included in PGS.
Furthermore, there is an unhelpful tendency – which the HFEA should resist, for the benefit of patients and professionals alike – to talk of PGS as a technology for improving implantation per se rather than a technology for improving the chance of implantation. This is very misleading. PGS involves selecting embryos in an attempt to improve the implantation rate per embryo transferred.
If the reason for testing is not to improve the chance of implantation but rather to avoid implanting embryos with certain disorders, then this needs to be stated clearly. There are established criteria for evaluating such screening programmes.
Setting aside public health arguments and concerns about which conditions to screen for, there are clear criteria on the need to understand the test and the pathways – false negatives, false positives and potential benefits and harms. It would be useful to refer to ongoing discussion about how to implement prenatal arrays, particularly in relation to the complexity of interpreting and managing results.

2. To offer a reproductive choice to patients whose offspring are at increased of genetic disease
(The use of PGD for HLA matching may appear to fall outside of these auspices, but is still relevant as it involves testing a known genetic variant for a known purpose.)
Array-type technologies and the amplification of whole genomes are already used clinically pre- and postnatally. The test is performed within a clinical context, and is focused on generating information that is relevant to the question being asked. Unanticipated findings are minimised.
It has always been the case that unexpected clinically relevant information may be discovered. The issue of feeding back is a clinical responsibility, and is by no means new to genetics community.
The challenge with use of these new technologies is how to maximise clinically relevant information (the 'signal') and minimise other information (the 'noise'). We have only a limited understanding of the clinical consequences of much genetic variation. Interpreting this variation is an immense challenge, which must not be underestimated.
Contrary to what is implied by the HFEA's email, in PGD it is already possible to test for two or more conditions if there is a clinical indication. Licences are issued on a condition-by-condition basis, and therefore to screen for multiple conditions with no prior risk would be a significant variation to current licensing processes.
As stated above, if the suggestion is that array or sequencing technologies could be used for a more general screen, this may be considered similar to the use of prenatal arrays. However, prenatal arrays are not generally used in a low-risk pregnancy. Guidelines on this area are currently being produced by the Joint Committee on Genomics in Medicine, but have not yet been finalised.
A key point, that must be recognised and clearly communicated, is that increasing the number of possible findings that are reported will reduce the number of embryos for implantation. This in turn will reduce the chance of a successful outcome for patients.