SO, - WHAT’S UP WITH POLYGENIC TESTING OF EMBRYOS (PGT-M)?
CONTROVERSIES IN FERTILITY & INFERTILITY.
David H Barad, MD, MS, is Director of Clinical IVF and Research and a Senior Scientist at the CHR. He can be reached through us at Hello@ReproducitveTimes.com.
Briefing: In contrast to preimplantation genetic testing for monogenic diseases (PGT-M), where the status of a single gene is determined, and preimplantation genetic diagnosis for aneuploidy (PGT-A), where allegedly the status of an embryo’s 46 chromones is evaluated, polygenic testing of embryos (PGT-P) allegedly assesses the risk of developing so-called polygenic diseases, which are diseases which occur if several different gene constellations work together to establish disease risk. We here describe PGT-A and PGT-M in their “alleged” purpose, while in regard to PGT-M the word is not used, - and this is by full intent. And the reason is that PGT-M is very accurate in achieving its goal, while PGT-A clearly is not (as the literature has demonstrated), while the accuracy of PGT-P is still completely undetermined as this article points out.
In J.D. Salinger’s book “The Catcher in the Rye” Holden Caulfield envisions himself as someone who saves children for falling off a cliff, a metaphor for his desire to protect children from the hardships of growing up. So too, do some of us, as modern catchers, imagine ourselves as protecting future generations by means of genetic screening of embryos.
Over the past two decades, the process of embryo selection in IVF has shifted from simple morphological assessments—based on factors like cell number and symmetry—to chromosomal screening aimed at detecting aneuploidy as a means of improving IVF outcomes by deselecting embryos with abnormal chromosomal complement before transfer into the uterus (called PGT-A). However, despite widespread clinical adoption in U.S. IVF clinics over the last 20 years, PGT-A has failed in enhancing live birth rates, as a recent combined Committee Opinions of the two most relevant professional U.S. societies in the IVF field recently finally made clear (1).
Now, the field is, however, poised to take yet another, even more daring step into genetic preimplantation embryo selection, moving from targeted single gene (PGT-M) and chromosomal (PGT-A) genetic assessments to whole-genome screening (PGT-P) using so-called polygenic risk scores (PRSs). This approach aims to estimate an embryo’s predisposition to polygenic traits and diseases, such as eye color, heart disease, or diabetes.
Though the introduction of PGT-P to IVF is in marketing campaigns usually presented as “progress,” this new practice offered by a small number of laboratories and in vitro fertilization (IVF) clinics in the U.S. has raised significant medical as well as ethical concerns. Among the former is an almost complete lack of scientific validation of PRS as a legitimate embryo selection method, while ethical concerns should be very obvious, - like the risk of (as happened with PGT-A) overpromising benefits, the potential for exacerbating societal inequities between poorer and richer segments of society, and – of course- a concern about eugenics. As such, it remains highly questionable whether these emerging technologies represent meaningful clinical advancements or merely add new complexities and cost to an already challenging and too expensive IVF process.
PGT-P involves the use of PRSs derived from large-scale observational studies in affected adult populations to estimate the potential future risk of disease in a child born from a given embryo. Unlike traditional genetic testing that identifies single-gene mutations (PGT-M) or chromosomal abnormalities (PGT-A), PGT-P assesses the cumulative effect of multiple genetic variants across the genome, each contributing a small amount to the overall risk of complex, polygenic diseases such as heart disease, diabetes, or certain cancers. By analyzing these polygenic profiles, the goal is to provide prospective parents with additional information about the relative risk associated for each embryo.
The predictive value of these scores is, however, inherently probabilistic and influenced by numerous factors, including population-specific genetic variations, gene-environment interactions, and the limitations of current genomic databases, which are, of course, based on already born individuals and, therefore, may not fully capture the diversity of human genetic backgrounds in preimplantation-stage embryos, as was learned in association with PGT-A, when it was discovered that embryos with chromosomal abnormalities at blastocyst-stage often can self-correct downstream from that stage (2,3).
Since PRSs provide a relative risk estimate, they do not offer definitive predictions. Consequently, patients – for example - may misunderstand a 20% increased risk for a disease as a 20% chance of developing the disease, leading to either overconfidence in the results or unnecessary anxiety. The determination of single gene mutations (PGT-M), such as sickle cell disease or Tay Sachs disease, in contrast, have clear and unequivocal outcomes; an embryo either is or is not affected.
Explaining the distinction between absolute and relative risk to patients can be particularly challenging, especially given that the predictive accuracy of polygenic risk scores (PRSs) varies widely depending on the trait in question and the nearly limitless variables involved. For example, PRSs predictions of adult height or an embryo’s future BMI are far more reliable than PRSs attempts to predict future autism or schizophrenia.
PRSs are, moreover, derived from studies of adult populations where the traits or diseases of interest have already manifested. Their applicability to embryos is, therefore, fraught with challenges, as practically all polygenic traits and diseases are influenced by developmental, environmental, and epigenetic factors that accumulate over a lifetime. An embryo's genetic predisposition, therefore, may not directly translate into the same risk levels observed in adults.
These limitations significantly complicate the interpretation of adult-derived risk scores when applied to embryos and may, in fact, call into question the entire concept of PGT-P. The practical impossibility of conducting prospective studies that span the developmental journey from the preimplantation-stage embryo, through pregnancy (a period where epigenetic modifications play a crucial role), and into adulthood undermines the scientific validity of these predictions. Without such studies, the utility of PGT-P remains speculative at best.
Further adding to the concerns about the already offered clinical utilization of PGT-P is the limited predictive accuracy of PRSs even in adults for whom they were developed because they often explain only a small fraction of the variance in disease risk. Adult PRSs, moreover, are frequently based on genome-wide association studies (GWAS) conducted in cohorts predominantly of European ancestry. The data generated in European populations are, however, not equally applicable to other populations, like for example Chinese Han. This is one reason why, even in adults, PRSs, are still considered experimental. Their clinical utilization in embryos, therefore, has been questioned (4,5), and even declared unethical by some professional genetic societies (6).
Ethical concern also arises from the possibility to apply PRSs to selection of non-health related traits, such as height, eye color, intelligence, or athletic ability. Eye color selection is, indeed, already offered in the marketplace (7). Information about non-medical traits could, of course, also reinforce certain societal stereotypes and prioritize cosmetic traits over choice of embryos more likely to lead to successful pregnancy.
That PGT-P at this point is not ready for primetime, should, therefore be obvious. That it, indeed, even in the foreseeable future will become a clinically and ethically viable offering as part of an IVF cycle, also appears unlikely, - but can, of course, not be ruled out. Any integration of polygenic risk scores (PRS) into reproductive decision-making forces us, however, to confront profound ethical, scientific, and societal questions about our capacity to use this information responsibly. We must ask ourselves if we possess the wisdom to navigate the inherent uncertainties and limitations of PRSs, while ensuring that its application truly aligns with the well-being of our patients and their future children. PRSs offer probabilistic data rather than definitive predictions, raising concerns about how this information might be misinterpreted or misapplied in the emotionally charged context of IVF. Are we prepared to communicate the nuances of relative risk in a way that empowers patients without overwhelming or misleading them?
Finally, as clinicians and scientists, we must openly confront the gaps in our current scientific understanding of polygenic traits, the limitations of risk prediction models, and the ethical boundaries of our interventions. Are we truly ready to wield such complex information in ways that prioritize the long-term well-being of patients over the allure of technological innovation?
As guardians and stewards of this field, we bear a profound responsibility to approach the questions surrounding PRSs with careful consideration and humility. The potential to apply this technology in ways that shape the genetic future of humanity places us in a delicate and morally charged position. Are we the proverbial “Catcher in the Rye,” standing at the edge of the metaphorical cliff, striving to protect others from unknowingly stepping into harm’s way?
Like Holden Caulfield, who wrestled with the desire to preserve innocence and shield others from an uncertain future, we too face a similar internal struggle. Our role is to balance the promise of innovation against the risk of unintended consequences, safeguarding the patients and families who entrust us with their hopes for the next generation. The complexities here are staggering. PRS data, probabilistic and incomplete, forces us to tread a thin line between empowerment and coercion, between advancing medicine and overstepping ethical bounds. We must ask ourselves whether we are truly equipped—not just with knowledge, but with the ethical clarity—to guide others safely through this uncharted terrain.
To be the “Catcher” in this context is not to prevent progress but to ensure that it is pursued thoughtfully, ethically, and with a clear-eyed understanding of both its potential and its limits. If we lose sight of this, we risk falling into the very abyss we seek to protect others from.
References
1. Practice Committees of ASRM and SART. Fertil Steril 2024;122(3):421-434
2. Bolton et al., Nat Commun 2916;7:11165
3. Yang et al., Nat Cell Biol 2021;23:314-321; CORRECTION: Idem 2021;23:1212
4. Gleicher et al., Nat Med 2022;28(3):442-444
5. Sierman et al., Social Scie Med 2024;343:116599
6. Forzano et al for the European Society for Human Genetics. Eur J Hum Genet 2022;30:493-495
7. The Fertility Institute. https://www.fertility-docs.com/programs-and-services/pgd-screening/choose-your-babys-eye-color. Accessed December 21, 2024