Should we edit human embryos' genomes, and if so then how?

I recently received a thoughtful letter from a college student I do not know – I'll call her Claudia* – who took the time to reach out to me with her perspective on heritable genome editing in embryos. The promise of this new technology is that it could fix disease-causing mutations in all cells of the body before damage occurs, and the correction would be passed on to future generations as well.

My laboratory at Columbia University, New York performs basic research to better understand early human development with the goal of benefiting patients with heritable diseases. Last fall, we published an article in Cell about using CRISPR/Cas9 to potentially correct a blindness-causing mutation in embryos; we discovered that the use of this technology could lead to unintended loss of additional genetic material. Our findings were discussed in national and international media, and had apparently caught Claudia's attention. I reference Claudia's letter here, as I believe that many people share the hopes and concerns she expressed, and because they also in fact align with my own scientific views.

My group's primary intent is to answer fundamental questions about the biology of the early embryo. This includes identifying the risks and benefits of changing genes early in development. When Cas9 is used to edit genes, it introduces a cut in the chromosome. In our study that Claudia read about, we showed that almost half of the time, instead of simply correcting the targeted mutation, the cut actually resulted in the loss of the entire chromosome or large parts of that chromosome. Such detrimental changes to the DNA will lead to the loss of the embryo. Similar findings were also published in the journal PNAS by the group of Professor Kathy Niakan, at the Francis Crick Institute in London, providing further weight to concerns about the potential impact of using this technology in humans. Basic research is critical because it prioritises insight into basic biology, unbiased by hoped-for clinical outcomes. Support for basic research enables scientists to be invested in finding answers, rather than requiring them to make inventions that generate a financial return. Knowledge about negative outcomes is critical information for regulators, and defines and narrows the path of what are meaningful interventions. It points out risks that need to be weighed against medical benefit, in comparison to other potential approaches.

In a commentary in the Wall Street Journal on our article identifying risks of genome editing, a longstanding critic of genetic intervention in the human embryo, Marcy Darnovsky of the Centre for Genetics and Society, used our findings about the potential damage CRISPR/Cas9 editing can do to chromosomes to support her argument. It is to be anticipated, however, that the risks we identified could be reduced, and clinical application might eventually become more realistic. This raises the question of what CRISPR/Cas9 should be used for, if it can be done safely?

In Claudia's letter, she stated: 'I know that the main goal... is to be able to prevent rare fatal genetic diseases, which I think is a good idea. But where do we draw the lines as far as morals and ethics? What if we accidentally edit out important genes that could help people survive a new disease in the future? As far as… choosing the qualities of children, I think that having the opportunity to design a child with appealing features should not be an option.'

Though the distinction between therapeutic use and design has yet to be fully defined, the prevention of a life-threatening disease such as cystic fibrosis, or the prevention of diabetes would likely fall into the first group, while engineering intellectual skill or athletic abilities would fall into the latter.

Editing the human embryo (or human germ cells) comes with unique potential consequences as any changes will be inherited by subsequent generations. Some scientists, including Professor Fyodor Urnov, of University of California Berkeley see little rationale in heritable genome editing, as he outlined in a comment article in Nature. He notes that there is still much more potential to treat disease by editing somatic cells without affecting germ cells. Others, like Harvard geneticist Professor George Church voiced support in the Wall Street Journal for broader use of heritable genome editing, including for immunity against infectious disease.

One of Claudia's important points above is that genetic diversity is the single biggest asset of human reproduction, preparing us for what is to come. If, in the future, genome editing were to be used to engineer qualities, this diversity might be compromised through what was being viewed as a desirable or advantageous genetic feature at the time.

Claudia also identified that genome editing of offspring driven by parents' wishes might confer a sense of 'entitlement' and expectation about how the child 'should turn out.' The use of a technology with non-zero risks to confer a genetic change when the person would be born healthy without any intervention, has no medical basis. Adding to that, neither the child nor subsequent generations could consent to such intervention that is not designed to address an unmet medical need. These are some of the fundamental flaws with design – it appears unlikely that any regulatory body would endorse such application.

The reports of the National Academies of Medicine and the Royal Society state that germline genome editing might be appropriate to prevent severe genetic disease for cases where other attempts, including preimplantation genetic diagnosis, do not succeed. At the time of writing, the World Health Organisation is due to publish its position paper on human genome editing on Monday 12 July, along with its 'Framework for Governance' report and a 'Recommendations' report (see BioNews 1103). It will be interesting to see how commentary from these bodies impacts public confidence in human genome editing technology. The current status in the USA, where the Food and Drug Administration is prohibited from considering any application involving genetic modification of embryos, suggests that there is not sufficient confidence in the ability of regulatory bodies to allow medical use while preventing non-medical use. And yet the idea of correcting mutations in embryos and germ cells is here to stay because of the compelling reason to prevent disease. If heritable genome editing is to become part of established medical practice, there is an urgent need for continued basic research to inform further debate and regulation.

* Not her real name, to protect her identity.