Tightening Gene Editing Rules for Humans

Dr. He Jiankui has ensured that his name will forever be linked to the history of gene editing in humans. In 2018 He announced that he had edited the genes of twin girls using CRISPR during IVF. Importantly, he edited them at the single cell stage, which means their germline was affected, which further means that the changes would be passed down to their offspring. He removed a single gene, the CCR5 gene, to reduce their chance of contracting HIV from their father.

He was widely criticized for doing human germ-line gene editing without proper regulatory oversight or approval. He has since been convicted and spent three years in prison, but now seems unrepentant and plans on doing further gene-editing research in humans. Specifically he wants to research how to cure the genetic neuromuscular disease, Duchenne muscular dystrophy.

At the same time China announced it is tightening its regulations regarding human gene editing. Part of the He controversy was that he was allowed to conduct such research in the first place. Regulations only limited human gene editing in hospitals. China has now expanded the ban to all human gene editing research, regardless of setting. Similarly, in the US germline human gene editing is banned, although non-germline genetic engineering is not (although there are further regulations about federal funding of such research). The EU position is that gene editing in humans: “may only be undertaken for preventive, diagnostic or therapeutic purposes and only if its aim is not to introduce any modification in the genome of any descendants.”

Cells in the body are either germline, which means they affect the germ cells (sperm and ova) and therefore can be passed down to offspring. Somatic cells, part of the body, are essentially everything else. If the cells in your liver, for example, were gene-edited those changes would only affect you, and could not be passed down. The big controversy with He’s research is that he altered all cells, including the germline, which can be passed down.

The obvious concern with germline gene editing is that it can potentially introduce genetic changes into the general human population. If such changes were not regulated, once this technology is widely available there would be a steady stream of such changes, having unpredictable effects on the human population. There is also concern that such changes could create different genetic populations of humans, even potentially genetically mutually exclusive populations. That would require significant genetic alterations, but is technically possible.

Often the term “designer babies” is brought up in such discussions. These are genetic alterations that are not intended to treat a disease, but to make either cosmetic changes (such as eye color) or to enhance function. This is the “Gattaca” scenario, as in the 1997 film of a dystopian biopunk future with extreme genetic profiling and discrimination (although in the film gene selection, but not editing, was apparently used).

Ethically there is a clear difference between somatic and germline gene editing. One can argue that competent adults have a right to make whatever changes they want to themselves. Germline changes potentially affect others, and this effect crosses national boundaries. So there is a motivation and justification for international rules regarding human germline gene editing.

However, I think that a simple ban is not the way to go long term. Dr. He is one demonstration of why this is – the goal of using gene editing to prevent Duchenne muscular dystrophy is a reasonable one. Pressure for such technology will become extreme, once the public gets the idea that this technology is possible and it works. A total ban is almost a guarantee of creating a black market, and medical tourism in less well-regulated parts of the world.

Also – using germline gene editing to prevent horrible genetic diseases is, I think, the right thing to do. Obviously the research needs to be highly regulated and supervised, but not completely banned. That is an overreaction. There are over 10,000 single gene disorders known, collectively affecting about 1% of the population. This is a huge disease burden. This is the low-hanging fruit for gene editing as these are generally single mutation disorders (although some are more complex such as repeat expansions). The knowledge of what changes we should make and their effects is pretty well known for most of these disorders.

The limiting factor now is simply the effectiveness, precision, and safety of the technology to use some version of CRISPR during in-vitro fertilization in order to make the desired genetic change. This process can be combined with something already in use – preimplantation genetic testing. Fertilized eggs without an undesired mutation can be selected for implantation. However, this won’t work for a dominant homozygous parent. l

But in practice PGT (selection) can obviate the need for actual gene editing in most cases, if the only goal is to prevent genetic disorders. But gene-editing can potentially do things that selection alone cannot. We can go beyond genetic disorders to genetic predispositions. What if a parent has two copies of the genetic allele that confers an increased risk of developing Alzheimer’s disease, for example? Similarly, there are some gene variants (alleles) that may be protective, reducing the risk of heart disease or stroke.

If we expand the list of acceptable targets from not only genetic disorders but genes that have a significant effect on disease risk then we can rapidly go beyond what selection can do. If, for example, there are 20 or 30 targets for gene selection in each parent, the probability of having a fertilized egg with all the desired alleles is negligible. Even with selection at the sperm level, with millions of sperm, getting all the desired traits will be difficult – and not possible with eggs from the mother, which are much more limited in number.

Even if we do not get to more extreme uses of this technology, such as physical enhancements, and just stick with disease modification, it’s easy to see the tremendous potential and risk of this technology. The utopian version is that single-gene genetic diseases can be virtually eliminated. There is always a spontaneous mutation rate, so new mutations are always appearing, even after fertilization. But if known mutations for dominant and recessive diseases were altered at the fertilization stage, the burden of such diseases would be dramatically reduced. At the same time we can reduce the burden of major diseases, such as neurogenerative disease, vascular disease, and diabetes.

The dystopian version is essentially the Gattaca scenario, where the technology is available to the relatively wealthy who not only disproportionately benefit from the technology but also use it to screen for potential partners. Even if this selection were not explicit, there is already a problem of selective sorting along economic lines when it comes to couples. It is not hard to imagine at all a dating app limited to people who can verify they have been “genetically cleansed”.

As if often the case with technology, the difference between incredibly beneficial and dystopian abuse is regulation and how we implement the technology. Usually, something in the middle emerges. Using gene editing will likely disproportionately benefit the wealthy, but the technology may become low cost enough that it is actually cost-effective for societies to subsidize the procedure. Remember – the whole concern with germline editing is that these edited genes get into the general population. But this cuts both ways – this also is a great potential benefit. Everyone potentially benefits from fewer disease causing alleles circulating in the population. A poor child may benefit from an upper-middle class grandparent who had gene editing.

The health-care cost savings to society could massively outweigh the upfront cost of selection and gene editing, especially for those with known genetic risks. Perhaps prospective parents may routinely get an exome screen, and then may qualify for state-sponsored gene editing (or at least insurance will cover part of it, or there is a tax-break or some assistance).

This world is very likely coming, sooner or later, even if it makes you feel squeamish. Rather than pushing this technology underground, regulations should focus on developing the technology in a responsible way, in order to protect individual patients and have the maximal benefit for society. A simple ban will not accomplish this.