How many people have fantasized about using technology to get a healthier, smarter, better-looking child designed according to their wishes?
In fact, this may not just be a rhapsody, someone has already put it into action. This man, Rafal, is a neurologist from Novant Health Forsyth Medical Center in North Carolina, who created the world's first child born through "polygenic score" screening.
Over fifty, he had to calculate the "perfect" child for himself
Rafal and his current wife, both over fifty, have reorganized their current families with their respective children after their divorces. Despite being conceived at an advanced age, they decided to have another child belonging to two people through assisted reproductive technology.
Considering that having children at an advanced age may bring a higher risk of genetic diseases to children, Rafal looked at a lot of research materials, and an article published in the MIT Technology Review caught his attention.
Source: Literature 1
The article is titled "Eugenics 2.0: We're at the Dawn of Choosing Embryos by Health, Height, and More." The article mentions that Embryo Selection Based on Polygenic Scores (ESPS) technology can screen for disease and select embryos [1].
ESPS is an application direction of polygenic risk scores (PRSs), which is to screen embryos for polygenic related traits through scoring in the process of assisted reproduction, so as to select embryos that are most likely to meet the needs (such as the lowest risk of a certain disease, the highest probability of high intelligence, etc.), and transfer them into the mother for fertility. At present, in addition to scientific research, the main potential of this technology lies in the primary prevention of polygenic genetic diseases [2].
Disease-related embryonic genetic scoring Source: Literature 3
Although Rafal practiced neurology clinically, he also obtained a doctorate in genetics in his early years. From ESPC, he saw an opportunity to choose the perfect gene for his child.
Although the jury is still out on whether the technology is effective, safe, or should be used for eugenics, Rafal is adamant. He found an IVF technology provider and another company willing to provide embryo testing services to work with IVF doctors to extract DNA samples, perform sequencing analysis, build genetic maps, and interpret genetic information from his own IVF embryonic cells to predict the likely health and intelligence level of the child after birth.
In the end, they selected 5 of the 16 embryos for genetic testing. Of these 5 embryos, one was discarded due to the detection of chromosomal abnormalities, and the remaining 4 received an overall "embryo score" [4].
Scientific fortune telling: This embryo "can make money without disease and disaster"?
Thanks to Rafal's efforts, his youngest daughter, Aurea Yenmai Smigrodzki, became the world's first "perfect baby" born through ESPC. At just two years old, she didn't know that she, as the world's first ESPC baby, was likely to live longer than others and have a lower risk of disease in her lifetime.
So how did this probability come about, and is it really reliable? To answer this question, let's first look at why the Polygenic Score can be used to "fortune-tell" embryos.
In the population, each person carries most of the same genetic information, but the genetic variation unique to the individual determines the differences between people in appearance, personality, physical function, disease risk and other aspects.
For two decades, based on the results of the Human Genome Project, a large number of studies have revealed the association between trait differences and genetic variants by comparing the differences in genetic variation between people with different traits (such as sick and unsick, high and low intelligence).
However, the impact of genetic variation on traits is not simply a one-to-one correspondence.
Taking diseases as an example, sometimes a single gene variant can lead to serious genetic diseases, such as thalassemia, phenylketonuria, etc., which are typically inherited in families. And the more diseases, the more complex the way of inheritance. For example, diabetes, cancer, high blood pressure, etc., there are many variants associated with them, and the impact of each risk variant is also large or small. At this time, a single variant has little effect on the risk of disease, but many variants accumulate a significant impact on the probability of final disease.
So suppose we want to evaluate the probability of a person being obese, what would the "polygenic score" do? First, we select genetic variants that are clearly associated with obesity from a large amount of literature data, and then determine what the weight of each variant is. Then, according to the genetic test results of the person, we list which of the above variants the person carries, and finally through complex weighted calculations, we can get a comprehensive score to visually reflect the person's risk of obesity.
At present, there are many personal genetic testing services, and the interpretation report will tell us such conclusions, such as "the risk of diabetes is × times that of the average population" and "the risk of obesity is × times that of the average population", most of which are obtained through polygenic risk assessment. These results are mainly calculated based on the academic literature related to each trait, and their reliability depends on the development of existing research results and database capacity, and the reference value of the reported results and the scope of applicable groups need to be further clarified.
Example of obesity risk assessment of a genetic testing company in China
In summary, polygenic scores in humans, although still in the research stage, already provide clinical or commercial testing services for the genetic risk of certain diseases [3].
On the other hand, embryo selection is actually not new. For parents who have difficulty conceiving, or have serious genetic diseases (such as chromosomal diseases, single-gene diseases) with high risk of death, teratogenicity, and disability, collecting sperm and eggs to grow into several embryos at a time in vitro, and selecting the best growth to implant the uterus, "IVF" technology has long been a mature application [3].
Since polygenic scores can predict traits and the probability of disease, and IVF technology has long been mature, shouldn't the combined ESPC also be effective?
But the fact is that although "scoring" embryos does not change the genetic sequence of embryos in the same way as gene editing, the problem is far from being solved whether considering efficacy or safety.
In its special report "Problems with Using Polygenic Scores to Select Embryos", NEJM pointed out that ESPS must at least be able to give a moderately accurate prediction of a desired feature in order to determine it to be effective.[5]
However, while there were indeed differences between people with high and low scores when producing polygenic scores from genome-wide association analysis, the populations that analyzed for these differences were not related. When ESPC is used, the object of analysis is several embryos of the same parent, so the predictive power is difficult to replicate perfectly.
For example, in a polygenic scoring study on future educational attainment, the college graduation rate was about 10% in the lowest 1/5 of the population with PRS scores, and about 45~60% in the top 1/5 population, and the difference was significant [6]. However, when verifying between the same family members, because the genomic differences in the embryos produced by one pair of parents are too small, the probability that the two surviving embryos of the parents will have the highest and lowest PRS scores as low as 3%. This is obviously almost impossible to make an effective ESPC [7].
At the same time, the current research on gene function is far from perfect. Genes are pleiotropic, and polygenic scores are likely to lead to another unfavorable trait when selecting one trait we want, such as choosing to make children smarter based on a gene, but the aspects that make this gene more likely to have psychiatric problems have not been revealed, which is likely to cause irreversible consequences [5].
While both effectiveness and safety are being questioned, ESPC adoption is spreading.
Currently, many companies around the world are offering ESPC services. To help customers visualize the data, the company shows intended parents how the score compares to the demographic average to form a comprehensible report.
Source: Orchid Health's official website
It's worth noting that while these companies all offer this service, they claim to have different characteristics. For example, Genomic Prediction primarily screens for diabetes, breast, prostate and testicular cancer, malignant melanoma, coronary artery disease, hypercholesterolemia, hypertension and schizophrenia, idiopathic short stature and intellectual disability. Orchid Health adds inflammatory bowel disease and Alzheimer's disease to the above conditions [5].
At the same time, these companies don't want to stop at preventing disease. MyOme claims to provide embryonic polygenic scores for education, household income, cognitive ability, and subjective well-being, and one of the founders of Genomic Prediction has suggested that screening for "above-average cognitive ability" and even skin color may be available in the future [5].
This embryo "can make money without disease and disaster", which looks really beautiful. But the risks, the unknown, the discrimination, and the potentially irreversible consequences behind it have yet been answered.
The child is about to be two years old, but the boundaries of embryo selection are far from clear
The company that provides Rafal with IVF technology, when learning about his needs, said: "Embryo screening is a gray area, and although it is not prohibited, we still consider it unethical to screen embryos."
But as the father of the world's first ESPC child, Rafal is convinced he is right. Not only did he give Aurella the perfect start, he also bought her a home test kit, hoping to use that information to guide her health and lifestyle throughout her life.
Different services can be performed using polygenic scoring: Genomic Prediction website
"If parents choose a high IQ for their children through genes, children should be grateful." Rafal insisted on his opinion, "When a doctor tells you that you can't use life- and health-saving technology because they think it's unethical to use it in this way, I think he's wrong."
The technical questioning of ESPC is indeed just the tip of the iceberg of opposition. Due to its high cost and the limitations of the databases included in the studies, the technique is largely only suitable for the wealthy, and the most accurate predictions are Europeans. And, because the more developed regions will have more resources for population genomic research, the gap is likely to widen over time, which will undoubtedly exacerbate inequities.
Moreover, the characteristics that are abandoned by the technology will be considered a "low-value" feature, which in turn will lead to prejudice and discrimination. For the smallest example, if many people choose double eyelids, then people with single eyelids may be discriminated against.
Professor Daniel J. Benjamin of UCLA said that the reproductive decisions of many individuals accumulated over generations can have far-reaching social consequences. These decisions can change demographics, exacerbate social inequities, bring discrimination, and more[5].
To this end, some scholars have called for the development of a policy and guidance document for ESPC as soon as possible, requiring companies providing these services to certify the completeness and accuracy of the information provided, and ensure that customers understand the meaning of the information.
But even so, the ESPC service continues. In the eyes of Rafal, the genetic screening of embryos is not crazy, and it is even the next step for human beings, who are just the first to do it. "In 10 years, this polygenic test [on embryos] will be completely uncontroversial, and people will do it for granted."
Acknowledgements: This article has been professionally reviewed by Wang Jizheng, researcher of the State Key Laboratory of Cardiovascular Diseases, Fuwai Hospital, Chinese Academy of Medical Sciences
The author of this article: Yunye, planning: Yunye | Executive Producer: Gyouza
Source: Visual China
Resources:
[1] https://www.technologyreview.com/2017/11/01/105176/eugenics-20-were-at-the-dawn-of-choosing-embryos-by-health-height-and-more/
[2] Lencz T, Sabatello M, Docherty A, et al. Concerns about the use of polygenic embryo screening for psychiatric and cognitive traits. Lancet Psychiatry. 2022;9(10):838-844.
[3] https://www.technologyreview.com/2019/11/08/132018/polygenic-score-ivf-embryo-dna-tests-genomic-prediction-gattaca/
[4] https://www.wired.com/story/genetic-screening-ivf-healthiest-embryos/
[5] Turley P, Meyer MN, Wang N, et al. Problems with the use of polygenic scores in embryo selection. N Engl J Med 2021;385:78-86.
[6] Lee JJ, Wedow R, Okbay A, et al. Gene discovery and polygenic prediction from a genome-wide association study of educational attainment in 1.1 million individuals. Nat Genet 2018;50:1112-1121.
[7] https://www.nejm.org/doi/suppl/10.1056/NEJMsr2105065/suppl_file/nejmsr2105065_appendix.pdf