A theoretical analysis by Australian authors and the newly created Uehiro Institute in Oxford has just been published in Nature, which we consider merits special attention. The article discusses gene editing in germ (gamete-generating) cells or human embryos using the CRISPR-Cas9 technique [1]. The study, whose first author is Professor Peter Visscher, is co-authored by the transhumanist philosopher Julian Savulescu, attached to the aforementioned Oxford institute funded by the Uehiro Foundation on Ethics and Education.
The paper analyzes the possibilities of applying “hereditary polygenic editing” (HPE) in germ tissue to reduce the risk of transmitting complex diseases to offspring, determined by systems of multiple genes with additive effects, known as polygenic systems. Although HPE remains in the experimental stage and it may take decades before it can be applied, this study anticipates its likely benefits and challenges for application in regions of the genome involved in the risk of transmitting complex diseases, such as coronary artery disease, Alzheimer disease, depressive disorder, diabetes, schizophrenia, hypercholesterolemia, cancer, etc.
The controversial technology applied to the germline proposed in this article has ignited a robust discussion on both its technical and bioethical aspects, which we are joining. Before doing so, it is important to clarify the general concepts and biological principles on which the modification proposed by the authors is based.
Experimentation in the germline and human embryos
When we talk about the germline, we are referring to the special tissue intended to produce the gametic cells in higher organisms. In human embryonic-fetal development, this tissue does not begin to develop until the seventh week after fertilization. It is important to know that, in most developed countries, experimentation with embryos generated from in vitro fertilization is only authorized up to day 14 after fertilization, when the primordial germ cells in the gonadal primordia have not yet been organized. Therefore, any genetic modifications made in that period of early embryonic development would affect the embryo itself, including its future germ tissue and, therefore, the gametes of the developing organism, which would transcend to the following generations.
When analyzing the technology proposed in this study, it should be borne in mind that it applies to embryos created through in vitro fertilization, whose biological nature and human dignity are indisputable.
From this perspective, any technology that is not aimed at benefitting that life, or that discriminates against it or puts it at risk, whether or not 14 days have passed since fertilization, must be considered unethical. Similarly, any genetic manipulation that affects the germline in adults carries the risk of genetic alterations appearing in their offspring.
The theoretical model of Hereditary Polygenic Editing (HPE)
The term edit means to correct, and refers to the ability to repair errors in DNA sequences. Since its implementation, this can be done using CRISPR-Cas9 technology, a molecular tool that can recognize and act on any known DNA nucleotide base sequence.
In its application in Medicine, clinical ends are pursued, aimed at treating or preventing a disease of genetic origin in somatic, non-reproductive cells, or for research use in order to learn about how genes function (for more detailed information, see the papers presented at the conference “Genetic research, its possibilities and risks“, organized by the Bioethics Observatory of the UCV in July 2024). However, it is not the same if it involves applying this technology for purposes other than health promotion, or when it seeks to modify germ tissue cells, which could transcend to future generations, as explained.
Although embryo selection after pre-implantation genetic diagnosis (PGD) is authorized in many countries to avoid certain monogenic diseases or chromosomal alterations, the difficulties are greater in its application for complex traits involving polygenic systems.
In a previous article, Liberal eugenics and preimplantation genetic diagnosis, we discussed the limited usefulness of PGD for determining the “polygenic risk”, which actually offers a confusing estimate of the probabilistic risk, in comparison to the certainty offered by this type of analysis in monogenic diseases.
In the mathematical model applied by Visscher et al. for the use of HPE in several disorders related to polygenic systems, it was concluded that editing only a small number of variants could significantly reduce the individual’s risk of disease, which would only be possible when heritable genome editing is feasible, safe, and applicable on a large scale in germ cells and embryos. According to the authors of the article, this scenario will require at least three decades.
Identification and evaluation of genomic differences in polygenic systems
Determining and evaluating the genetic variants that are statistically associated with the type of disease or polygenic trait to be modified starts with the analysis of a type of genome molecular marker: single nucleotide polymorphisms (SNPs). The analysis is facilitated by comparing the genomes of thousands of previously studied whole genome samples (genome-wide association studies, GWAS) available in databases.
It is based on comparing the genome of those who share a phenotypic manifestation with those who do not. In other words, the method looks for common SNPs in people with a specific disease or trait, under the assumption that the differences detected are related to the phenotype in question.
Since polygenic systems are analyzed, i.e., multiple “loci” distributed throughout the genome, the SNPs detected provide a map of regions of the genome related to the phenotypic difference. However, that does not mean that one or more specific candidate genes have been located; these are only regions of the genome that influence the total phenotypic manifestation of the polygenic trait concerned to a greater or lesser extent.
Although a polygenic risk score (PRS) is used, a statistical calculation that assesses the likelihood of a person having or developing a particular medical condition or phenotype, it is not possible to assess the relative importance of the different genomic variants detected, nor are these data sufficient. They are based only on the presence or absence of small variants distributed throughout the genome, but do not take into account environmental or other factors that are particularly involved in polygenic systems.
GWAS thus allows the discovery of areas of the genome or signaling pathways that are likely associated with a particular phenotype.
Nevertheless, it is important to note that the majority of the statistically more significant differences in SNPs are found in non-coding regions (introns or intergenic regions), making it difficult to demonstrate their causal role. Moreover, there may be polygenic systems without large effect variants, so that if the goal is to apply HPE to modify the genome, the change would only be achieved by editing many “loci”, which is not technically possible in the short term.
There are also further problems related to the evaluation of SNP variants found in polygenic traits, on which the genomic editing proposed by Visscher et al. in germ or embryonic cells is based. The authors themselves highlight these difficulties:
- It is not possible to predict the degree of variation in the prevalence of a disease that would be achieved after applying HPE to each region of the genome involved. As stated above, these studies do not identify causal alleles, but only common genetic variants that are linked to the disease or trait to be modified, For example, according to the authors’ model, the application of genomic editing to 10 SNP variants presumably involved in Alzheimer’s disease would only reduce disease prevalence by less than 0.6% of the population of modified embryos. Even if genomic editing were completely effective, given that each individual may differ from another in the SNP variants present in their genome, not everyone will benefit equally from polygenic editing.
- Although gene editing using the CRISPR-Cas9 technique offers a very active and promising field of research, its application is still unsafe and it can be a source of errors, especially due to its off-target effects, i.e., unwanted and unpredictable in its consequences. These can be particularly serious if it involves modifying genes in germ cells or embryos, both in individuals born from them and in their offspring.
- Many variants associated with polygenic diseases can have “pleiotropic effects.” “Pleiotropy” or “polyphenia” is the production of phenotypic effects that are seemingly unrelated to a particular genetic variant, and is due to the gene’s multiple effects on metabolism during embryonic development [2]. Under these conditions, it is impossible to predict the side effects of gene editing on embryos or the germline.
- Similarly, “epistatic effects” may occur [2], i.e., effects of the interaction of one gene on the phenotypic expression of another gene or other non-allelic genes. It is not guaranteed either that the positive effects of different variants involved in a polygenic trait or disease will be added together.
- The environmental factor is not considered. Alleles that are favorable or “protective” in one environment may not be so in other settings, today or in the future. In the absence of data on people carrying several rare protective variants, additivity is not guaranteed for all environments. In fact, there are risk variants that have very variable effects in different environments and circumstances in life.
Despite all the drawbacks mentioned, Visscher et al. insist on their mathematical model as a basis for evaluating the application of HPE technology when this is feasible and safe on a large scale in embryos or in the germline.
Rare protective alleles
Thus, for the latter, there are alleles of particular genes that, despite their low prevalence in the general population, are known to offer protection against certain diseases in carriers [3]. Their acquisition in non-carriers through CRISPR-Cas9 editing would confer a kind of “genomic prophylaxis” upon the embryos in which it was applied.
This was the goal of the Chinese researcher He Jiankui from the University of Shenzhen, who, disregarding regulatory standards and abandoning all caution in this type of research, announced the application of genome editing in human embryos. According to him, the embryos were implanted and resulted in the birth of twin girls Lulu and Nana in November 2018.
His unproven idea was to modify the CCR5 gene into a ‘protective allele’ against the AIDS virus. We must take into account the fact that, even if the father was HIV-positive, the girls were not at risk of acquiring an infectious disease unless exposed, so performing gene editing was completely reckless and unnecessary. It was a foolish idea with unpredictable consequences for girls whose development will have to be followed throughout their lives.
Bioethical considerations
The article discussed herein serves at the very least to spark a debate on a type of technology, heritable polygenic editing. While not yet ready for practical implementation, it raises ethical concerns alongside the host of technical difficulties and repercussions mentioned, rendering its proposal irresponsible and premature, presented with undue confidence in the article under consideration.
The first of these is the problem of the use of embryos obtained by in vitro fertilization, which, as stated above, are human lives in the initial phase of their development. This implies a high level of reproductive intensity due to the selection and discarding of thousands of embryos resulting from these techniques, either because they do not carry the desired genetic combination or because of technological manipulation.
In addition, a phenotypic assessment is implemented, classifying individuals as better or worse based on their heritable characteristics, which, in every respect, is a eugenic procedure similar to the “social eugenics” of the early twentieth century.
It is unrealistic to think that gene editing can be applied to polygenic systems related to complex diseases when the conditions and guarantees for affordable use to treat simple monogenic diseases are not even in place.
Indeed, the authors recognize the risk of recreating a eugenic era similar to the one that occurred in the late nineteenth and early twentieth centuries, for which they propose promoting respect for individual liberty, safeguarding privacy, preventing discrimination or stigmatization for genetic reasons, and encouraging societal values such as diversity and equality.
Similarly, they advocate that states should neither impose their vision of a good life, nor use coercive measures to encourage the application of HPE outside of Medicine.
That may be all well and good, but in practical terms, it would not be necessary, given the unrealistic nature of its present and anticipated future applications. The proposed HPE represents an ideal scenario, in the sense that would only apply to embryos created by in vitro fertilization. Consequently, even if it has some advantage for improving health, it would benefit just a small percentage of the human population.
The European Society of Human Reproduction and Embryology (ESHRE) estimates that, in 2023, around 12 million children were born worldwide using this technology, and that by 2100, 400 million will be born, which will account for about 3% of the world population. It is assumed that the remaining 97%, which reflects the population derived from natural fertilization, would be excluded from a technology whose advantages are only seen by its proponents.
Another problem that arises is the establishment of limits in terms of the type of modifications that are to be carried out in human populations using these gene editing techniques, since some will not be satisfied with the eradication of diseases due to monogenic or polygenic systems; rather, their goals for enhancement will encompass other traits related to physical health or well-being, such as intelligence, height, longevity, etc., which are in line with transhumanist goals.
It should be emphasized that, for these types of traits, the desired levels of enhancement or well-being could be achieved through actions outside genetic editing, geared towards improving the environment, an extremely important factor in human well-being and, especially, in the phenotypic manifestations of polygenic traits. That is to say, through actions aimed at improving living conditions, habits, education, food and opportunities for physical and cultural development.
Nicolás Jouve
Professor Emeritus of Genetics at the University of Alcalá
Member of the Bioethics Observatory of the Catholic University of Valencia
References
[1] Visscher, P.M., Gyngell, C., Yengo, L. et al. Heritable polygenic editing: the next frontier in genomic medicine? Nature 637, 637–645 (2025)
[2] Rieger, R., Michaelis, A., Green M.M. Glossary of Genetics and Cytogenetics. Classical and Molecular”. Springer Verlag, Berlin, Heidelberg, New York, 1976.
[3] Rutter, L.A., MacKay, M.J., Cope, H. et al. Protective alleles and precision healthcare in crewed spaceflight. Nat Commun 15, 6158 (2024).