PGT-P Guidance

ASRM

The ASRM published a joint Ethics and Practice Committees opinion in late 2025 (formally appearing in Fertility and Sterility 2026) concluding that PGT-P should not be offered as a clinical service at this time and should only be performed under IRB supervision until safety, efficacy, ethical, and societal concerns are addressed. The concerns fall into several distinct categories. American Society for Reproductive Medicine

Scientific and predictive limitations

The committee's foundational concern is that the underlying technology simply isn't accurate enough yet. Polygenic risk scores provide probabilistic rather than deterministic predictions of disease risk, and there is uncertainty about the accuracy and reliability of these scores. Crucially, polygenic risk scoring only accounts for 5–10% of clinical variation in the conditions tested — meaning the vast majority of disease risk comes from factors the test doesn't see. 

Several specific problems compound this:

  • Environment and epigenetics are ignored. Simply knowing an individual's genetic makeup is insufficient to predict future disease development in a given individual, as both environmental factors and epigenetics influence whether and to what extent a given disease will manifest over time. 

  • Temporal mismatch. The scores are derived from today's adults, but the children born would develop disease decades from now under different environmental conditions. The genetic risks an individual faces today may not accurately reflect risks when the resulting children would be expected to develop these diseases many decades into the future, and the long-term predictive validity of these tests cannot be meaningfully assessed until those future decades arrive. 

  • Pleiotropy. Pleiotropy, defined as the condition in which one gene influences two or more seemingly unrelated phenotypic traits, complicates the interpretation and prediction of genetic findings. Selecting embryos on the basis of one desired trait may inadvertently impact other traits influenced by partly overlapping or the same genes. 

  • Population bias in reference data. The databases used to create algorithms for PGT-P derive from samples that are most often collected from adults of Western European Caucasian origin, thereby potentially limiting the generalizability of the data. 

The absolute-versus-relative risk problem

One of the most pointed critiques is that PGT-P results are easy to misinterpret. Relative risk can sometimes exaggerate the perception of risk, especially if the general population risk is low. Small reductions in absolute risk correspond to large reductions in relative risk when the risk of a particular clinical outcome is low to begin with. The committee gives a concrete example: although the relative risk reduction of developing type 1 diabetes in offspring of biological parents with European ancestry is seemingly substantial at 35%, the absolute risk reduction for this disease is only 0.12% points, primarily because of the condition's low prevalence in the US population (0.34%). A "35% reduction" sounds impressive but corresponds to barely more than a tenth of a percentage point in actual risk. 

Risks intrinsic to IVF itself

PGT-P requires IVF, which carries real costs and risks even for people who could conceive naturally. The committee notes that pursuing PGT-P can lead patients into multiple cycles of IVF with the goal of trying to obtain embryos with the lowest risk, raising concerns that include possible embryo damage, misdiagnosis, mosaicism, imprinting, monozygotic twinning, and a false sense of security regarding the health of resulting children. There's also a practical problem: many IVF cycles do not result in sufficient number of embryos for embryo rankings to meaningfully impact the outcomes. 

Embryo selection dilemmas

If too many embryos get flagged as "high risk," patients may end up with nothing acceptable to transfer. The reduction in the number of embryos deemed suitable for transfer could potentially limit the chances of successful pregnancy and live birth, adding to the complexity and stress of the IVF process. The committee also flags concerns about how the technology is marketed — the practice of ranking embryos for transfer on the basis of polygenic risk estimates raises ethical concerns, particularly if presented in a way that could be perceived as directive, potentially undermining patient autonomy in reproductive decision making. 

Disease hierarchy, disability, and equity

PGT-P implicitly elevates some conditions over others. Polygenic embryo risk scores may inadvertently create a nonbinary hierarchy of disease by prioritizing the prevention or mitigation of certain conditions over others. This occurs in part because only select conditions are included in testing panels, meaning some diseases are assessed whereas others are excluded. There are knock-on effects: the emphasis on screening for specific diseases or traits could stigmatize individuals and families affected by those conditions, and there is a risk that PGT-P could contribute to discrimination against individuals with disabilities by deselecting embryos affected by certain diseases, thereby limiting research funding and clinical experience with these conditions. 

Access disparities are also flagged: Preimplantation genetic testing for polygenic disorders can be expensive, is not covered by insurance, and is available only to those who can afford it, thereby exacerbating socioeconomic disparities in access to advanced reproductive technologies, particularly among marginalized and underserved populations. 

Psychosocial impacts on parents and children

The committee worries about effects that ripple beyond the testing itself. Learning about potential risks or conditions may cause anxiety, stress, or guilt, particularly if there are limited options for addressing or mitigating those risks. There are concerns about birth order — if earlier transfers are based on perceived lower risk, siblings born later may be viewed as inherently less healthy, which could have unintended social or psychological implications within families. And there's a worry about the child's "open future": parents might encourage or restrict access to opportunities or environments on the basis of PGT-P scores that conflict with the child's desires, or children might engage in risky behaviour if they believe their genetic risk is low. 

Informed consent failures

The committee is sceptical that meaningful informed consent is currently achievable for PGT-P. Patients may hope for comprehensive solutions and definitive answers regarding their reproductive choices. However, the current clinical offerings may only provide limited efficacy and safety information, making it challenging to meet patients' expectations. And: the complexity of polygenic inheritance means that not all data can be provided confidently, creating a gap between desires for complete information and the limitations of what can be accurately communicated. PGT-P may also surface incidental findings with implications for relatives, with neither research nor guidelines on how such information might impact clinical management. 

The committee also explicitly limits the role of patient autonomy as a justification: autonomy alone is not a sufficient justification for the widespread clinical implementation of a technology that remains scientifically and ethically unsettled. 

What ASRM recommends instead

The committee concludes that research on PGT-P should only be undertaken under an institutional review board protocol, focused on validating clinical utility and addressing the ethical concerns before any clinical rollout. They also note this aligns with positions taken by the American College of Medical Genetics, the European Society of Human Genetics, and the European Society of Human Reproduction and Embryology, all of which have reached similar conclusions. 

One final point worth highlighting — the document explicitly carves out trait selection (height, intelligence, eye colour) from its scope, stating that PGT-P used for trait selection (e.g., height, intelligence, eye color) is not within the scope of reproductive medicine at all. So the document's qualified "not ready yet" framing applies only to disease-risk use; trait selection isn't considered a legitimate medical question even in principle.

PET

The Progress Educational Trust (PET) is a UK charity focused on fertility and genomics — it publishes BioNews.

The 2025 PET Annual Conference in London ("What Does Genomics Mean for Fertility Treatment?"), was used by the HFEA to publicly clarify the UK's legal and regulatory position. 

PGT-P is unlawful in the UK. 

Julia Chain – chair of the Human Fertilisation and Embryology Authority (HFEA) – said in her concluding presentation that 'PGT-P is unlawful for use in the UK', because 'it does not meet a permitted purpose for testing set out in the Human Fertilisation and Embryology Act'. Progress

This applies whether you're trying to screen for non-medical traits or for disease risk. Screening embryos for non-medical traits is not permissible in UK law. According to the HFEA, using PGT-P to screen for risk of conditions such as cancer, diabetes, or Alzheimer's disease is not permissible either. 

The legal logic — important because it distinguishes UK law from the US ASRM ethics framing — is that the HFE Act permits testing for the presence of a known condition, not for statistical estimation of future risk. As Chain put it: 'PGT-P is not the search for the presence of a condition, but instead the assessment of its likelihood'. 

She also stated the scientific case alongside the legal one: 'Current evidence suggests that PGT-P is not sufficiently advanced for application in embryo testing, with scores giving information relevant to populations rather than individuals'. 

The enforcement loophole PET has been highlighting

This is what makes the PET coverage particularly newsworthy rather than just restating the law. UK couples have found a workaround:

  • They have IVF and PGT-A (which is legal) in the UK.

  • They request the raw genomic sequencing data from the PGT-A lab.

  • They send that data to a US PGT-P provider for polygenic scoring.

  • They return to the UK clinic and request a specific embryo for transfer — without disclosing why.

The HFEA's chief executive Peter Thompson addressed this directly: 'There is nothing to stop a UK-based couple seeking such testing, and indeed treatment, overseas, but a UK licensed clinic should not then make decisions on what embryo to put back using that information'. 

How this contrasts with the ASRM position

It's worth seeing these side by side because the framing is quite different:

  • ASRM (US): PGT-P is not recommended for clinical use at this time, on a mix of scientific and ethical grounds — but it isn't illegal. US clinics offer it; ASRM tells the profession not to. The opinion explicitly carves out trait selection as outside reproductive medicine but allows that disease-risk PGT-P could eventually become acceptable if evidence improves.

  • HFEA (UK, via PET): PGT-P is unlawful, full stop, regardless of whether it's for traits or disease risk — because the HFE Act 1990 only permits testing for the presence of a known condition, not probabilistic risk estimation. Changing this would require parliamentary action, not just shifting professional opinion.