Would selecting embryos from genetic predispositions make the world any better?

The prestigious journal Nature Medicine has just published in its March issue an article in which researchers from California and New York analyze the genome of more than one hundred embryos and try to predict what risk they would have of suffering from common diseases throughout their lives. Its proposal is to apply this technology to pre-implantation genetic diagnosis in order to select embryos that are less likely to have these diseases. But the issue is complex and not without serious problems, as other researchers argue in a commentary that appears in the same issue of the journal.

Pre-implantation genetic diagnosis (PGD) has been offered for years as part of in vitro fertilization techniques , with the aim of implanting only embryos free of mutations that cause some serious genetic diseases. It consists of extracting a cell from the embryo at its earliest stages, when it is formed by no more than eight cells, and studying the presence of mutations in its genetic material, presuming that all other cells in the embryo will have the same genetic constitution.

Given the low frequency of these diseases, few couples resort to PGD. On the other hand, its use to detect chromosomal defects in the young embryo is the subject of intense debate in the scientific community, due to different issues that have to do with the first cell divisions in the embryo. But this situation could change, if the utility of the DGP were demonstrated to predict the risk of having much more common diseases, such as cancer, Alzheimer’s disease and diabetes. This is precisely the line of work that concerns us.

Genetic factors

Much of the genetic research of the last twenty years has tried to to decipher the genetic factors that are associated with several common diseases, or even with physical and psychological traits. For this, the letters of the genome that change from one human to another are studied. Of the three billion nucleotides in our genome, most are – of course – the same in all people, but there are several million positions in which different people carry a different letter. By comparing these genetic variants in healthy people and in patients with, for example, type 1 diabetes, it is possible to identify all the variants that are associated with a high risk of having the disease. The effect of each one is very small, but as each disease has many (dozens or even hundreds), information from all of them can be combined to calculate the so-called “polygenic risks”.

The authors of this research studied one hundred and ten embryos, coming from ten couples who had resorted to in vitro fertilization at the time when they had only five cells. Next, they read about six million variants in the genome of the cell extracted from each embryo and compared it to the complete genome of each of the progenitors, which allowed them to reconstruct the embryo’s genome. As ten of these embryos had been implanted, they were able to compare the predictions with their respective genomes, since they had already been born, and they saw that they were very good.

Having validated the methodology, they analyzed the data from the UK Biobank (a collection of nearly a million people for whom very detailed genetic and medical information is available). They calculated the polygenic risks of each Biobank donor for a total of twelve common diseases that included various types of cancer, heart disease, or diabetes; as expected, they proved that people with higher risks had a higher incidence of these diseases.

Probability is not certainty

The problem here is the magnitude of these effects. For example, people with a very high polygenic risk of coronary heart disease were three to four times more likely to have it than those with lower genetic risks; only in the case of type 1 diabetes, a probability ten times higher was reached. This illustrates very well something fundamental: these risks apply to population groups, but they never predict what will happen to specific people. Is it scientifically correct to use this information to decide the fate of an embryo in vitro? It seems not: a probability is not a certainty, especially if we are talking about low or moderate probabilities.

There are other limitations, as acknowledged by the authors of the work. For example, the predictive power of these risks is greatly reduced when they apply to populations of different ancestry than the population originally analyzed; Given that most studies have been done on white individuals from developed countries, these genetic variants are not useful for calculating risks in people of African or South American descent. The authors are also concerned that access to these technologies is restricted to the wealthy classes, but what really matters is that the public correctly understands that these calculations do not provide a crystal ball, but a probability that only works with large numbers but not it works for specific cases.

Eugenicist mentality

The worrying thing is that this type of risk can also be calculated for behavioral traits , such as mathematical intelligence or reading ability. In all these situations, genetic factors influence. But how much? A few weeks ago, for example, a huge study was published that analyzes the genetic variants associated with the educational level achieved in three million participants, calculating a polygenic risk that explains 15 % of variation in educational levels. The rest of this variation is due to environmental, family, social or educational factors, but its importance tends to be overlooked by the emphasis on polygenic risks. Not to mention psychiatric illnesses or personality traits. Will embryos at high risk for bipolar disorder or hyperactivity be eliminated? In an increasingly eugenicist mindset, it is evident that the very perception of disability will be greatly affected negatively.

We must not minimize the importance of genetic association studies and the calculation of polygenic risks: they are are crucial to understanding the molecular causes of many common diseases, and identify a high-risk population group in which early detection campaigns will be especially effective. But, misapplied, they can be very harmful; some are already calling for the implementation of a “personalized education” based on the genetic makeup of children. And, of course, companies already offer in vitro fertilization users to select embryos based on genetic variants that predict “low intelligence” or short stature. That is why it is urgent to adequately inform society so that everyone understands that a more or less high probability of having a disease depending on the genetic constitution is never the last word and should not be the criterion for making important decisions about embryos, children or flesh-and-blood adults.

Javier Novo is a professor of genetics.

©2022 ACEPRENSA. Published with permission. Original in Spanish.
Back to top button