The risk of developing many common conditions could be drastically reduced by making dozens of edits to people鈥檚 genomes at the embryo stage, claims a team of biologists and ethicists. We don鈥檛 yet have the technology to do this safely, but we should start thinking about whether to use it when we do, they say.
However, their paper has come under fire from other biologists, who say we currently know too little to predict the effects of this kind of genome editing with any certainty and that making 鈥渋ncredibly speculative鈥� claims is irresponsible.
鈥淚t鈥檚 going to be taken up by people who are pushing a eugenics agenda in an unsophisticated way,鈥� says at Trinity College Dublin in Ireland.
Advertisement
Genomes continually mutate, so there is a huge amount of variation in the human population. A few variants or mutations are clearly harmful, causing inherited conditions such as cystic fibrosis.
What effect most variants have is much less clear, however. To find out, biologists trawl through medical data looking for associations between particular variants and the risk of, say, heart disease.
These studies have revealed variants that appear to reduce the risk of certain conditions. The most common of these protective variants only have a tiny effect, but some rare protective variants have a bigger impact. For instance, some people have much lower cholesterol levels due to variants in a gene called PCSK9.
at the University of Queensland in Australia and his colleagues have now estimated what the potential benefits would be if genome editing were used to give individuals dozens or even hundreds of rare variants that each has a relatively large protective effect. Making lots of edits to a single cell at once is known as polygenic editing.
The idea is that the editing would be done at the embryo stage so every cell in the body inherits the edits and they are passed down to any children, known as heritable or germline editing.
According to their estimates, making around 10 edits could reduce the prevalence of type 2 diabetes up to 60-fold. A different 10 edits could reduce the prevalence of heart disease up to 30-fold, Alzheimer鈥檚 disease nearly tenfold, and so on. The smallest benefit was with major depressive disorder, whose prevalence would be halved.
Making 40 changes at once could reduce an individual鈥檚 lifetime risk of developing Alzheimer鈥檚, schizophrenia, type 2 diabetes and heart disease to less than 0.2 per cent for each condition, the team concludes. 鈥淧eople tend to underestimate the potential power of polygenic gene editing,鈥� says Visscher.
Because even the most advanced forms of CRISPR gene editing can lead to unwanted changes to genomes, it isn鈥檛 yet possible to safely make so many changes at once. But Visscher and his colleagues think this will become possible within three or so decades, so we should start thinking about whether, and how, to use heritable polygenic editing now.
鈥淭his is not an advocacy piece,鈥� says team member , a bioethicist at the National University of Singapore. 鈥淚t鈥檚 about raising the issue and beginning a conversation.鈥�
But there are numerous problems with the team鈥檚 estimates, says Mitchell. Firstly, these rare variants might be rare because they protect against one condition but increase the risk of others. 鈥淚t overlooks the glaring fact that these may be rare for a reason,鈥� he says.
Secondly, association studies cannot reveal which variants are the direct cause of the effects they are linked to 鈥� it is like being able to link a gang to a series of murders but not being sure which individual pulled the trigger.
Thirdly, the effects don鈥檛 necessarily add up. If two variants work the same way, having both may be no more beneficial than having one.
Fourthly, the effects of variants can depend on which other variants people have and the environment they live in. A variant that is protective in one person in one environment might be harmful in another person or environment.
Visscher and his colleagues do acknowledge these potential limitations in their paper and also discuss the many ethical issues surrounding heritable polygenic editing, but Mitchell thinks they don鈥檛 go far enough.
The science of understanding how genetic variations affect individuals is still in its infancy, he says.
鈥淭hese sort of simplistic models are simply not warranted when it comes to making predictions about individuals where you鈥檙e going to make an intervention and you have a responsibility to do no harm,鈥� says Mitchell, who, with others, has of the paper. 鈥淚 think we need to be humble about where we are now and not think that we can predict the outcomes of these things with any certainty.鈥�
at Harvard Medical School, who has estimated how polygenic editing could extend lifespan, also has an issue with the idea that such editing could reduce the prevalence or lifetime risk of degenerative ageing conditions. People will still get these eventually, just at a later age, he believes. 鈥淚f we live long enough, all of us would get Alzheimer鈥檚.鈥�
Nature
