When Will We Have Designer Babies?

Within 20 to 40 years, sex will no longer be the preferred method of reproduction. Instead, half the population with decent health care will - not kidding you - have eggs grown from human skin and fertilised with sperm, then have the entire genome of about 100 embryo samples sequenced, peruse the highlights, and pick the best model to implant. At least that's what Stanford law professor and bioethicist Hank Greely predicts in The End of Sex and the Future of Human Reproduction. But skin-grown humans aside, how long until we have "designer babies"?

Illustration: Chelsea Beck/GMG

Here's where we are: A gene-editing tool called CRISPR/Cas9 has produced a variety of terrifying wonders over the past few years. In August, a group of scientists announced that they had successfully edited a human embryo to eradicate a heart condition. (The validity of their paper, published in the journal Nature, is now disputed, but Chinese scientists also claimed to have edited embryos in 2015, though with less success.) And "designer pets" are already within reach; mice have been turned green. Beagles have been doubled in muscle mass. Pigs have been shrunk to the size of cocker spaniels with "designer fur". Woolly mammoths are being attempted.

Greely expects human selection to be less like the Build-a-Bear workshop which "designer pets" suggests, in which you start from scratch with a base animal and pick features from a menu, but rather in which prospective parents select from a pot of about 100 embryos made by two people for preferences such as gender and health, and maybe tweak out the hereditary diseases with CRISPR/Cas9.

It's debatable when and how CRISPR-edited embryos will be approved for implantation (US Congress has currently banned the FDA from even considering it), and scientists are currently wrestling over putting a moratorium on its use on humans. CRISPR's own co-founder Jennifer Doudna fears what eugenic nightmare her technology could bring; in 2015, she co-signed a letter calling for thorough investigation into potential risks "before any attempts at human engineering are sanctioned, if ever, for clinical testing".

This week on Giz Asks, we asked geneticists, bioethicists, and biotechnology experts and sceptics whether a future of "designer babies" is as crazy as it sounds.

Hank Greely

Director, Center for Law and the Biosciences at Stanford University, author of The End of Sex and the Future of Human Reproduction

I think that when average readers hear the term "gene editing," our minds jump to Gattaca or a build-a-bear workshop for babies where you walk into a lab and select desirable traits (strength, beauty, intelligence, and so on) from a pamphlet.

I think GATTACA was much more about selection than about editing, but, yes, most people have an exaggerated view of what is possible.

We don't know anything, really, about genes that give increased IQ. I suspect - and in my book, The End of Sex, I predict - that in 20 to 40 years we'll know a little, but not much - maybe enough to say "this embryo has a 60% chance of being in the top 50%" or "a 13% chance of being in the top 10%." Intelligence is just too complicated and despite decades of work only genes associated with very low intelligence have been found.

Note, though, that you said "genetically select" - selection is more about preimplantation genetic diagnosis [PGD, pre-screening for genetic diseases in embryos] and picking among embryos randomly created by a couple. Editing is intentionally changing DNA away from what the couple created. Neither, though, works worth a damn unless you know what the relationship between the DNA and the trait is and not only aren't we close with intelligence, we may never be very good at it: too many genes are involved, along with too much environment and too much luck!

...On eye colour, hair colour, skin colour, etc., we've got some clues now and will be pretty good in ten to 20 years - though I doubt CRISPRing embryos will be shown to be safe and ready for clinical use in less than 20 to 30 years. On IQ, maths ability, sports ability, music ability, personality type we'll have some information but probably not very much: 60%, 70%, maybe 80% chances of being in the top half but not 90 to 100%. ... Right now, though, we have made almost no progress on those traits or on common complicated diseases like asthma, type 2 diabetes, or depression.

... The more simple genetic diseases like Huntington's disease turn out to be quite rare - common diseases are usually a mix of heavily genetic, somewhat genetic, and not at all genetic. For example, 1% of people with Alzheimer disease (and so about 1 person in a 1000 in the population) have an early onset form that is very strongly genetic. If you have the gene variation involved, the only way you won't get the disease in your 40s or 50s is to die first from something else. About 4% of the population have two copies of a genetic variation that gives them a 50 to 80% chance of getting AD, compared with the overall average 10% or so. About 20% of the population has one copy of that variant and has about a 20 to 40 per cent chance. About 2% of the population has two copies of a different version of the same gene and seem to have zero chance of getting AD. And most people with Alzheimer disease have no copies of any of the known, strong genetic risk factors for it. That's what most common diseases are like; most behavioural traits will likely be even worse.

Glenn Cohen

Harvard Law School Faculty Director, Petrie-Flom Center for Health Law Policy, Biotechnology & Bioethics, leading expert in bioethics and law

I do not love the term "designer babies" because it is imprecise. [In one sense], they are already here and have been for a long time. When [individuals] consider need to use a sperm donor for artificial insemination... they engage in a form of trait selection. The catalogues that feature sperm "donors" (donors in scare quotes because they are paid) recruited for sperm banks have already excluded 99% of applicants, and who is left tends to have very desirable health, intelligence, and beauty traits. The same is true for egg "donors." So already parents using these technologies are "designing" their babies. And there is nothing to stop someone who wanted to from buying both sperm and egg from donors to maximise their ability to engage in trait selection. For individuals using IVF [in vitro fertilization], there is also the possibility of using preimplantation genetic diagnosis (PGD) to examine the embryos and screen out those predicted to have diseases or other problems and engage in sex selection. The next step on the horizon, written about recently in the MIT Technology Review is combining big data from whole genome sequencing and population databases to make predictions on non-disease traits as part of the PGD technique.

Some companies are either directly or indirectly building this into their pitch to investors as that report suggests, though also [the MIT Technology Review reports that some scientists are sceptical]: "Some experts contacted by MIT Technology Review said they believed it's premature to introduce polygenic scoring technology into IVF clinics - though perhaps not by very much. Matthew Rabinowitz, CEO of the prenatal-testing company Natera, based in California, says he thinks predictions obtained today could be "largely misleading" because DNA models don't function well enough. But Rabinowitz agrees that the technology is coming."

Even if perfected this will only be available to individuals willing to go through IVF and PGD, costly in both time, money, and health risks. So I don't know how much uptake there will be in the near future. The more radical change would come if In Vitro Gametogenesis became cheap, safe, approved, and easy such that we could generate a large number of eggs without having to make women go through egg retrieval. I am a bit more sceptical it will be [as quick as Hank Greely suggests], but do think that such technology will be used in the future. Finally, if CRISPR gene editing becomes available as a therapeutic technique in the US, and we get much much better at targeting what to edit, it may be possible to do all this with fewer embryos generated and have more options since we can alter existing embryos instead of waiting to find an embryo that has the desired traits created through natural means. I am sceptical though that the regulatory system in the US will allow human use of this on embryos in the foreseeable future.

So the bottom line is on some definitions of designer babies they are here already and have been for some time. On other definitions I think the estimate is somewhere between 20 to 100 years depending on what you think about the progress of technology and the regulatory system in our country.

David Liu

Harvard professor in Chemistry and Chemical Biology, founder of five biotechnology companies and lead researcher and innovator in gene-editing technologies

When will 'designer babies' become a reality - if so, how many years from now, and what would that look like? For example, will there be a time when people will be able to edit their future child's genes to lead to greater strength, musical talent, mathematical genius?

Probably never. There's a common misconception that allows people to imagine designer babies as you describe them. The misconception is that traits such as the ones you list above behave in a "Mendelian" manner - that is, like the traits of Gregor Mendel's famous pea plants: tall vs short plant, green vs yellow seeds, etc. It turns out that the vast majority of human traits are not "monogenic," meaning that they are not determined by the DNA sequence of a single gene. Instead, traits such as mathematical or athletic or musical ability reflect thousands of genes -- recent research suggests perhaps even all of our genes! - as well as many environmental factors.

So even if one imagines a future in which (A) genome editing capabilities are truly like word processing (note: we aren't there yet!), and (B) editing of human embryos, currently quite controversial and not practiced in most countries, is commonplace, it's still unlikely that designer babies with traits such as higher intelligence or superior athletic skills will ever be possible. Human genetics simply doesn't work that way. Finally, it's also worth pointing out that the screening of human embryos to avoid certain genes strongly associated with devastating genetic diseases prior to implantation and pregnancy is already taking place in a number of countries around the work, and has been for a number of years.

...Even cosmetic traits such as hair and eye colour, while genetically simpler than mathematical ability, aren't really monogenic. Also, one has to consider the cost, risks, and ethics associated with such a procedure. The risks and costs and ethical challenges are easier to bear if the outcome is to avoid an incurable, fatal genetic disease that causes great suffering. For changing eye colour, it's not so compelling.

Dietrich M. Egli

Assistant Professor of Developmental Cell Biology at Columbia, prominent sceptic of the recent "breakthrough" claims of successful embryonic editing by CRISPR technology

The technique is not there yet. Nobody has shown that they can modify the genome of the human germ line and reliably avoid adverse effects. Nobody knows when that hurdle will be overcome. The most recent paper that appeared in Nature reporting they could [successfully edit the DNA of a human embryo] is most likely wrong.

Does that imply that we are not even at a place where we can use gene-editing at all, for anything? CRISPR is not an effective tool yet? And what kind of adverse effects? Leading to mutations of some kind?

[CRISPR] works very well in cultured cells, where you can afford to have a less than 100% efficiency, and then simply select those that worked. It may also work for gene therapy in somatic cells.

And, its an amazing research tool. No clinical applications yet. It's just a few years since it has been discovered.

Do you think there's a natural balance of how many super-traits a person can possess, in a similar way to how when homo sapiens' jaws shrank, it made more room in the skull for a larger brain? For example, would it offset the natural balance if maths wizards were also born football players and possessed the musical talents of Mozart?

I don't think there is a good answer to that. There are tradeoffs and constraints in the human body. There is also a huge gap of knowledge what a specific variant could accomplish.

For instance, there are some variants that protect against malaria, but cause [sickle cell] disease as well. The most meaningful thing to do is to further study the human genome to enable improvements in human health.

Dr Robert Green

Medical geneticist and director of the Genomes2People Research Program at Brigham and Women's Hospital; Professor of Medicine, Harvard Medical School

If we're talking about the next five to ten years, I think what's really exciting about these technologies is that they may allow us to edit genetic errors. Genetic errors like a mutation that would cause a disease in an infant. I think that that's really the focus, it's the disease focus, not the designer baby focus. That's point number one.

Point number two is most of the traits that you can imagine someone trying to enhance, which is intelligence, height, athletic ability, musical ability and so forth, are probably very multi-genic traits. They are traits that many, many genes, many regulatory regions and so forth. They are very complex. Even if there were not ethical rules in place, which I think there probably will be ... I think that it would be very, very difficult to [change those things].

If, say, the United States outlaws that kinds of thing, do you think other countries might start offering services where you can edit for things like hair and eye colour? Do you foresee that as an inevitable consumer service?

I guess, I think that things like that are possible, for sure. I think that the hope is that most countries will try to agree on guidelines for which most of humanity will agree to and abide by because of the potential to do so much harm when messing with DNA to lose embryos, to create diseased human beings. I do hope that such actions as selecting for specific cosmetic differences might not be encouraged or permitted, but I am aware that there are reproductive clinics already in which gender selection is done. Embryos are created, they are looked at for a potential gender and then implanted. It is going to be difficult to regulate. Yes, all that can take place.

How quickly do you think this could happen, if the technology made it to the market, for consumers to use those services?

I guess theoretically, I would say, within a decade. But again, even things like hair colour may be driven to certain traits. An eye colour may be driven by a few genes, but you've got to imagine most people do not want to put their embryos at risk for anything less than avoiding a really serious disease. The notion that you would put an embryo, this child, at great potential risk - we really aren't going to know the long-term effects of mocking around in the DNA of a human embryo and to put an embryo at risk for what you perceive to be a cosmetic improvement, I think would be against the principals of most responsible parents.