Gene editing human babies

[liv]

I'll probably get round to making a post noodling about different social media platforms and gathering links about the Tumblr purge and so on. But lots of people have interesting views about that. For now, something to make my journal interesting to all my potential new readers where I have a specific contribution to make:

Who has been following this real-life horror story where a scientist claims to have gene edited human babies? It's absolutely all over the technical press but I haven't seen much lay discussion. If any of this story is true, it's a massive first with extremely far ranging implications. Let me break this down: He Jiankui is claiming that he has edited the genes of human embryos and that twin girls were born a couple of weeks ago with their genes deliberately altered. That means every cell in their bodies will express the edited sequence. If they go on to have children of their own, their offspring and all future generations of their descendants will have this alteration. (The account is slightly confusing because the romanization of the scientist's name, He, is identical to the male singular pronoun. Sorry about that.)

If it's real. If it's real it's as big as human cloning, and it's also a complete ethical disaster with bad science at every stage.

What He claims to have done is that took some human embryos created by IVF from a father who is HIV+ and a mother who is seronegative. He then used a very recently developed technique called CRISPR/Cas9 to change the sequence of a gene which encodes a protein that the HIV virus would normally bind to. In theory, this change means that the cells can not be infected by HIV. Then. Then he took these edited embryos and implanted them back into the mother and, we are told, she went through a normal healthy pregnancy and gave birth to twins. There is some vague evidence from similar approaches in animals that these babies may be protected against at least some strains of HIV. NB not "immune to" HIV as a lot of popular reports are saying, it's not that they can mount an immune defence against AIDS, it's that their cells can't be infected with the virus in the first place.

This decade has seen a major revolution in molecular biology: until really very recently it was just accepted as fact that you can't directly edit mammalian genes. You could alter genes, but it was always a scattershot approach, you never had precise control over what new sequence you'd get. So it was a bit rubbish even for doing experiments, and certainly not adequate for medical use. Around five years ago, there was a new technology developed, based on modifying some really obscure enzymes from bacteria, so now we can, in principle at least, change any genetic sequence at will. Nowadays, any molecular biologist can buy a kit from mainstream scientific suppliers and edit genes in most types of cells, including human cells.

Up until now, people have been cautiously improving the safety and reliability of this technique and holding off on using it in actual living humans. There are particularly strict prohibitions against editing embryos, because of the multi-generational implications and potential for eugenics. Until this rogue scientist, He, went on unpaid leave from his university, and secretly and with a very sketchy and hardly documented ethical approval process, recruited a bunch of Chinese couples who wanted to have children but the potential father was HIV+. He doesn't seem to have clearly explained to these parent volunteers what he was doing, some vague talk about an HIV vaccine, which this certainly isn't. It's possible that he talked his subjects into avoiding the normal precautions you would take if trying to conceive when the father is HIV+. Even if he didn't deliberately increase the babies' risk of contracting HIV, he at very least ignored the existing and effective ways of preventing transmission from an HIV+ sperm producing parent, which are both safer and more certain to work than this highly experimental gene editing thing.

He then made a big fanfare of a press release. Nothing's published in peer reviewed journals, but based on He's own report, most scientists who work in the field believe that the precautions taken to check that the editing had worked correctly were entirely inadequate. Part of the reason why normal, ethical scientists don't try to gene edit humans is that CRISPR/Cas9 technique isn't completely reliable yet, there's still a high chance of what's called "off-target effects", namely editing unrelated genes by mistake, which is going to have entirely unpredictable consequences. And it seems like He has done some cursory checks to make sure this didn't happen, but not proper, rigorous ones.

Then He went into hiding, occasionally popping up to claim to the media that he's a great maverick who has advanced the medical field by ignoring all the petty rules that Hold Science Back. Every statement I've seen from him makes him come across more like the most clichéd of cartoonish "mad scientists".

Because of this pattern of behaviour, I think there's quite a high chance that the whole thing was a fraud. Remember Woo Suk Hwang, the Korean scientist who in the early 2000s claimed to have made a cloned human, using the techniques applied to breed Dolly the Sheep? Yeah, he made the whole thing up. When something this big is published in a press release and skips the peer review stage, I am very skeptical that any of this human gene editing actually occurred. (Of course, now He has grown a conscience and tells us he can't reveal the real identity of his supposedly engineered babies because he wants to protect their privacy.)

And because the whole story is so ethically horrifying, with the parents not properly consenting to be in such a revolutionary trial, and the babies having an unknown risk of contracting HIV (might be lower than unmodified babies, might not be), and an unknown risk of basically random side effects, I am really hoping it is, in fact, a fraud. I hope the terrible experiments never happened and He is just trying to hog the limelight. I was expecting at some point in the next five to ten years that there would be a big breakthrough in human gene editing, because now we have the tools there are so many huge medical problems we could tackle. But I wasn't expecting it to be like this.

Please feel free to ask any questions. I haven't done CRISPR/Cas9 gene editing myself but I have a 20-year career of doing that sort of experiment but with worse tools, and I'm happy to explain this stuff at whatever level is most helpful.

Comments

[emperor]

FWIW, this story did the rounds at work, but I've not seen it much discussed elsewhere. As you, say, there is so much ethically wrong that "fraud" is probably the less bad outcome.

[liv]

Yeah, I'd be pretty surprised if your workplace hadn't noticed this! I think there's a pretty high chance of fraud because of the working in secret and bypassing both his institution and the peer review process. But I suppose, if He had carried out and submitted this work in a normal way he would never have got approval for it, so horribly unethical experiment actually happening is about equally likely as fraud :-/

[redbird]

I was thinking of this as "being talked about" because I saw it in at least three places--but two of those are science and medicine blogs (the second of which I looked at because of a "more info here" link on the first). He's announcement and some of the criticism was covered on the CBC news webpage (one of my regular news sources nowadays), which is probably where I came across it first.

I agree, the best case here is that he faked the whole thing, rather than that he did something inappropriate that might have unrelated harmful effects on these infants, and lied to the parents to get "consent."

[liv]

Yeah, it's not that the news isn't in the "mainstream media" at all, it's that it isn't being talked about as much as I would expect for something this huge. In some ways that's a good thing: if opinion-formers are holding fire until this has been thoroughly checked out and replicated in properly controlled conditions, that will discourage people like He from trying for a shot at fame in this extremely damaging way.

[jjhunter]

I've been seeing considerable discussion about it in various science journals & science-related news streams, and if it was a successful edit it increases the risk of another disease, not to mention potential life-long health implications that we don't even know about yet.

It's such a damn arrogant thing to do, and the utter lack of proper informed consent for the parents, IRB oversight, or exceedingly thorough animal testing first is infuriating and utterly against 'do not harm' or at least don't screw up more than what you're trying to fix.

Also the sketchy evidence that this even offers anything more than reduced risk of HIV infection - which is something that you can get by taking daily oral doses of one of the new HIV 'vaccine' -type meds these days - is such a minor benefit to gain for the long, terrifying set of potential off-target side effects, not to mention the known vulnerability the on-target effect induces. Some protection against HIV for increased vulnerability to another disease is a devil's bargain. (If it were complete immunity to any and all variations of the common flu forever more - that might be worth the risks of someday maybe making building up to editing in, but just partial against some current strains of HIV? Seriously? That might be outdated / irrelevant before they reach five years old.)

tl;dr this fraud of a 'scientist' fucked with two kids' lives and lifelong health as a PR stunt and that is absolutely unacceptable.

[liv]

Yes. I share your anger, the lack of consent is the biggest problem, but the inexplicable choice of gene for the first trial is nearly as bad. Even if this works perfectly it's worse than existing anti-HIV measures! I'm starting to hate He nearly as much as Wakefield, honestly.

[silveradept]

I haven't seen much around in my spaces, but from what you're describing, it sounds like someone has done a thing that should get them barred forever, license revoked, from doing anything at all scientific for the rest of their life. If this turns out not to be fraudulent, at which point they should probably still be sanctioned heavily.

Is there legitimate research into gene editing on human cells for various reasons? Is there an idea in mind to make this a regular medical practice in the same way that we spend monies on drug research?

[liv]

I completely agree that He should face career-ending sanctions over either committing this scale of fraud, or conducting this hugely unethical experiment.

There absolutely is legit research into gene editing. Masses of it. People were trying it even before we had efficient tools, because there a lot of genetic diseases that really can't be treated with drugs. I don't know if it will ever be a regular practice, it would take a lot for gene editing to be anything like comparably safe and cheap with drug-based medicine. But if we had a way to treat otherwise incurable fatal diseases, there's a huge incentive to try it.

It's an open ethical question whether this should ever be done on embryos, though. That's what's known as "germ-line gene therapy" - editing every cell in the body and causing changes that will be passed on to offspring. If you could edit only the cells that are already affected, that has much fewer unknown and very long term consequences. Scientists were already working on altering the genes of particular cells in order to treat or cure genetic diseases, even when all they could do was delete or add a whole gene, not edit the sequence. With CRISPR/Cas9 there's a whole lot more scope, and I think we'll see legitimate clinical uses fairly soon.

[silveradept]

Thanks for the additional information and context. It can be difficult to see between the breathless press releases to what the actual information is, and sometimes further difficult to read the technical say to know whether or not I should be impressed.

[alatefeline]

That is ... horrifyingly unethical and mad-science-y. I certainly HOPE it is a fraud.

But also, I expect that at some point in the next few years there WILL be a confirmed case of CRISPR gene-editing of humans that is *not* a fraud, because the tools have become good enough and accessible enough that it is at least PLAUSIBLE for that one person *willing* to go there (out of many more with the knowledge and access, but better standards) to do this. I can believe that most people are trying to be and do good and don't have shitty standards, I can be hopeful about humanity, while still being cynical about the collective performance of large numbers of humans combined at in upholding ethical standards or predicting consequences.

I think the technology to edit genes is amazing and I actually am hopeful that it will be able to help humans be healthier/more adaptable at some point in the future; I am not inherently opposed to deliberate genetic alteration per se or to research into means and effects. But I WILL NOT accept that it's remotely safe or okay to IGNORE precautions and ethics, when the consequences of charging ahead and ignoring the warnings, of taking brutal advantage and harming innocents and exploiting vulnerable populations, are so visible all over the history of human experimentation and of the abuse of genetics-related pseudoscience. This is horrifying.

[liv]

It really is horrifying, isn't it? Even He's own account of what he's done, carefully massaged to make himself look good, is just awful.

I think you're right that with the sheer number of people out there, it was inevitable that someone like He would try something like this sooner or later. I'm really hoping that legitimate, ethical uses are developed and adopted first, though.

I'm very much on the same page that gene editing has the potential to drive some amazing breakthroughs. There are also huge risks to the technology, and also risks about the way humans keep trying to use eugenics to promote a master race at the expense of vulnerable groups. Both the medical and the social risks need to be properly mitigated before this technology can be used in a medically beneficial way.

[crystalpyramid]

This went around my Facebook friends accompanied by appropriate levels of horror. The idea that it might be a fraud seems like the most promising thing I've heard.

[liv]

Your friends are apparently well informed, then. I really really hope it's fraud, and I think the chances are quite high, precisely because the research was done in secret and published without peer review.

[crystalpyramid]

What's even the point of that kind of hoax? Dramatic public career suicide?

[liv]

From He's public statements, I get the impression he genuinely doesn't realize what a terrible thing he's done. He thinks he's going to go down in history as the First! Scientist! Ever! to gene edit a human being, and people will be so impressed with his breakthrough they will overlook his cavalier approach to ethics and the normal scientific process.

He also keeps comparing himself to the mostly-American biohacker community, and doesn't seem to understand the difference between adults attempting risky gene editing on themselves, and what he's done to babies with inadequate parental consent. Basically he thinks ethics is just a bunch of annoying bureaucracy getting in the way of advancing science, AFAICT.

Thanks for the context.

[jesse_the_k]

Adults attempting risky gene editing on themselves is a grand tradition in science at the borders, yes? No IRB required if you're the one harmed?

Could you point me to more info on biohackers? (I'm assuming this isn't at the level of the Prolethians).

[liv]

It's not an area I know very much about. I think a lot of the people who are making videos of injecting themselves with gene editing enzymes are complete cranks, but some of them are desperate because they have serious diseases with no developed cure, and some of them are genuine activists who are trying to democratize genetic medicine. This article on Biohacking CRISPR stunts is probably a decent starting place. But anyway, it's not at all equivalent to genetically editing babies with inadequate consent.

[jesse_the_k]

Thanks for the pointer.

[ewt]

I have been vaguely aware of it, in a "well, that's horrifying if true, but there isn't a lot I can do about it either way and I guess we'll find out eventually" way.

[liv]

That makes sense. I suppose I can't do very much about it, but I can at least make a lot of noise about why this is horrifying and unethical and completely the wrong way to advance medical science.

There's a scenario where the babies really were edited, but the editing didn't have very much effect, and their lives are basically normal. I think that's reasonably likely; it's still awful that they had such a drastic experiment performed on them that wasn't at all in their best interest, but if we're lucky they may not suffer much direct harm. Like, maybe they're slightly more resistant to AIDS and slightly less resistant to West Nile Virus than an average human, but hopefully they won't get either of those diseases anyway, and hopefully they won't have any other side effects.

[ephemera]

*reads with interest*

I heard something about this on Radio 4, but you lay out all the *many* ways in which this is horrifying so much more clearly than their brief piece.

[liv]

Thank you. Much of the credit goes to Yong who is one of the best science journalists currently active. But the last couple of weeks my pro Twitter feed has been full of all the ways it's horrifying. It may still be a huge breakthrough if it turns out to be true, a proof of concept that this can actually be done in humans. The unethical and sloppy approach will set research back rather than advancing it, though :-/

[lilacsigil]

Something I don't understand and would love to ask a scientist: why would someone who wants to be seen as a maverick genius pick this particular project for his gene editing? Surely he could have got more credit and less horror for editing out a really terrible hereditary disease rather than maybe lessening the risk of HIV in a way he can't prove worked and has better odds with pharmaceutical treatments anyway?

[liv]

The least awful reason I can imagine is that He went along with something he was already working on. I mean, it's not uncommon for people working in gene therapy to try the gene they already study, partly from familiarity, and partly because the reason they chose to work on that gene in the first place is because they believe it is medically important.

But more cynically, I think the reason he chose the HIV receptor is, firstly that the stigma about AIDS made it easier for him to persuade the parents to "volunteer". They were scared about their children's future and so agreed to put them through a procedure that would supposedly make them resistant to HIV.

And secondly, I think the not being able to prove it worked is a kind of advantage, because he also can't prove it didn't work. If his rickety badly designed experiment doesn't work, it's going to take much longer for the failure to become obvious. Like, if he'd claimed he was going to cure a terrible hereditary disease, and tinkered with some embryos and then a child was anyway born with the terrible hereditary disease he offered to prevent, well, there goes his big scoop. This way he can show genetic readout saying that these babies might be HIV resistant, and maybe they are, but who knows? Most people don't get HIV, including most people who have HIV+ fathers.

[lilacsigil]

Oh, yes, that's a good point - I was thinking primarily from the "go down in history" POV, where "can't prove he didn't do it" is perfect if he's fraudulent from the start!

[worlds_of_smoke]

Honestly, as a disabled person, I am really, really hoping it's a fraud.

But the problem is that, eventually, this will really happen (if it hasn't already). :/

[liv]

Yeah, the disability and eugenics angles are really scary. There's a very good reason why multi-generational genetic alteration of embryos is banned everywhere.

I think that if the technique were used, with proper consent, to allow people with genes for severe life-limiting conditions to conceive genetically related offspring, it wouldn't be "this happening". I still don't know whether that would be morally acceptable. I want to live in a non-ableist society that doesn't practise eugenics, but in the society we have, I'm not sure if gene editing is worse than the current practice of genetic screening to prevent the birth of babies with genetic conditions.

[bagpuss]

Wandered in from miss_sb

There are situations where gene editing people is might be a good idea, severe cancers which aren't amenable to other treatment is an example

In rarer circumstances, there might be reasons why gene editing embryos is a good idea such as both parents have the same severe life-limiting (or just plain life preventing) mutations and this is the only way they can have children that are genetically related to them, though the "requirement" for that seems less clear

Whatever is true, before regular gene editing of people, embryonic or post-natal is a thing there needs to be a lot of transparent discussion about when it is appropriate, what level of certainty/safety needs to be confirmed in animals first and what oversight is appropriate before it happens and not what may or may not have happened here (I am with you on hoping it is fraud)

[liv]

Hi, welcome. I can believe in scenarios where human gene editing might be desirable, and I entirely agree we need far more robust animal evidence and procedures before it can be a safe option.

I'm not sure gene editing is likely to be a useful approach to cancer, because cancer is a multi-gene disease. Changing a single gene, assuming all the practical difficulties could be surmounted, would probably not halt the growth of a tumour, let alone prevent the spread of cancer once metastatic (spread through the body, a feature of what we'd usually define as severe cancer). Cancers are highly mutation prone and are under selection pressure. There have been a number of attempts to block driver genes using traditional methods and what always happens is that a few cells of the cancer find a way to work round the block, and keep growing anyway. If someone had a genetic problem with their liver or bone marrow, and we could fix it in 3/4 of the cells, fine, the problem would be basically fixed. But if we only fixed 3/4 of the cells of a tumour, the remaining 1/4 would just grow even faster.

Diseases caused by a single mutation which mean people can't have genetically related children are a much more likely candidate at least for early applications of the technology. But I really don't know whether I could be ethically comfortable with fixing those mutations by using gene editing. Certainly it would need to be very tightly legally restricted to prevent all kinds of ethically horrible processes.

[siderea]

I'm not sure gene editing is likely to be a useful approach to cancer, because cancer is a multi-gene disease. Changing a single gene, assuming all the practical difficulties could be surmounted, would probably not halt the growth of a tumour,

BRCA1.

[liv]

This... probably needs to be a separate post. Yes, BRCA1 is a good example of a driver mutation where a single gene change has major implications for cancer genesis and progression. But I don't believe that BRCA1 gene therapy is ever likely to be clinically useful.

Are you talking about gene editing embryos where there's a family history of BRCA1 mutations, to reduce susceptibility to breast cancer? Or treatment of cancer that is already severe and not treatable with conventional chemotherapy? The first case might have potential, if we collectively decide that gene editing embryos is ethically acceptable. The second is I would say entirely unfeasible.

[siderea]

I don't even know how you got onto the topic of the latter case. Post-natal gene therapies weren't even under discussion, were they? The controversy under discussion is precisely gene editing embryos.

[liv]

Ah, sorry, threading fail. I was responding to bagpuss who mentioned severe cancers which aren't amenable to other treatment – I took that to mean post-natal, as in treating somebody who already has advanced cancer, rather than susceptibility genes like BRCA1. Gene editing to prevent familial cancers is absolutely plausible, despite the involvement of multiple genes. Good point.

[bagpuss]

Most cancers are indeed complex diseases but there tends to be smaller number of driver mutations and larger numbers of passenger mutations. With the use of single-cell sequencing, we are also starting to better understand the heterogeneity across tumour microenvironments so relatively small edits might end up having large impacts

This isn't simple but what I failed to find when I posted the first response was this story where it already has happened

https://bmcbiotechnol.biomedcentral.com/articles/10.1186/s12896-018-0455-9

Here talens rather than crisprs have been used to treat acute lymphoid leukaemia. These treatments tend to be last resort at the moment but we will get more effective at this

As far as treatment of Mendelian genetic disorders, you are correct there will need to be a lot of oversight and lots of testing before it happens but that doesn't mean it shouldn't be a possible treatment in the rare circumstances where it is the best solution

[liv]

Thanks for that link, that's really great stuff and I hadn't seen it. Apologies for writing my first reply to you at the wrong level; I should have checked how much you already know about the technical side first.

I do agree that driver mutations are important, even with heterogeneity. And yes, I can very much see the potential of using gene editing to improve stem cell / immune therapy treatment for haematological cancers. I can't see the Qasim paper cited in that fascinating review, but from the summary of it it looks like a fantastic first step in actually using gene editing clinically.

[ayebydan]

I have seen it here and there and just hoped it was....wrong? In that it did not happen because while I see science heading that way for good or ill this is NOT the way to start.

[liv]

Fair enough! I am completely with you in hoping it's wrong. The fact that everything was done in secret without normal scientific collaboration or review does seem to increase the chances that the claimed experiment never happened, I think.

[halojedha]

Thank you for unpacking the implications and ethics of this so clearly - I hadn't heard the story, and the more I read about the cavalier approach and lack of consent the more I shared your horror.

I think the ethics of gene editing are fascinating and complicated - as is the social tension between improving health, ableism and eugenics. I can see this technology contributing to a widening wealth and wellness gap if only rich parents can afford to make their kids free of genetic diseases. But I am very interested in the potential of medical technology to improve our quality of life and capacity as humans - our health, memory, lifespan, intelligence - and I don't know how to weigh that against the inevitable realities of developing such technologies within our current capitalist and racist world. Not to mention the potential for technology like this to undo important social work being done to reduce stigma around disability, and increase social support and resilience.

[liv]

Yes, I do agree entirely with your second paragraph. I am pro-technology, I think we can achieve a lot of amazing benefit to humanity with genetic technology. But we can also do a lot of harm if we apply the technology in a racist, eugenicist way.

[naath]

Predictably I heard that at work, also on some science focussed podcasts, but not the news. It smells like fraud, if not I hope those kids are OK, because who knows what that'll do... AFAIK crispr/cas9 has the common side effect of 'cancer', on account of off target effects (and that editing my hair to be purple is apparently impossible)

[liv]

Yes, cancer risk from off-target effects is one of the biggest reasons why we aren't actually ready to be gene-editing humans :-(

The big problem with purple hair is that no mammals have purple fur, so where would we start modifying a gene to make purple pigment? I bet lots of people would change their hair genetically if they could, though. If we see gene editing getting to the point where it's (approximately) cheap and safe that will be a huge market.

[hairyears]

This was one of the lead articles on Reuters a couple of weeks ago and I'm fairly sure that the Financial Times picked it up, too.

It was a topic of discussion among traders and geeks in banks; the cat really is out of the bag.

The questions and opinions among my fellow bankers range from 'probably fake' to 'someone in Beijing wants this' by way of a shower of derogatory remarks about medical and academic ethics in China - which I sincerely hope are untrue.

I think we're heading for the 'Black clinics' future from Neuromancer and I recall that this was not, in any way, a utopian society.

The point I've picked up from your summary today is that it's germ-line editing. I hadn't realised that and it's horrifying.

The authorities will almost certainly order that these children be sterilised; and that will be the least of the evils that will follow from this monstrous act.

If He isn't convicted and sent to a labour camp, or shot, then someone will find a use for his technical abilities and his amoral irresponsibility.

[liv]

It's really fascinating when I make a post and ask, who's talking about this and learn about how far a story has penetrated into different groups. Interesting that financial world thinks it's worth paying attention to.

I think emphasising that He is Chinese is not really relevant here and it's basically rooted in racism. Like, both his institution and the Chinese authorities have credibly denied any involvement in this grotesque experiment. It's not impossible that the Chinese government in fact did encourage this and the whole thing about He doing the work in secret and the public condemnation are just propaganda. Based on what we've seen, I'm disinclined to believe that it was caused by "China" in general having weak medical ethics.

Whether he'll be useful, I don't know. By his own account he doesn't seem to be a very competent scientist, and if some shady governmental or organized crime folks wanted a geneticist with no conscience they could find a better one. Also the whole thing about CRISPR/Cas9 gene editing is that it's not very hard at all. Anyone with a few weeks' training in general molecular lab skills could probably do it. IVF takes more specialist skills than that, but it's not like there's a desperate shortage of people trained in the 40-year-old technique of making test-tube babies.

[hairyears]

What needs to happen in molecular biology to prevent the toxic and irresponsible 'brogrammer' culture taking root?

We really dropped the ball on that one, here in the IT professions.


I am reminded that 'computing' once took place in dedicated buildings with 'Laboratory' in the title; now it's an industry, not a science, and there really is a computer on every desk, in every office and in every home. And in quite a lot of lightbulbs, microwave ovens and doorbells, with the password set to 'Admin'.

This was unimaginable forty years ago. Ridiculous, even: so I'm not at all sure that editing DNA won't be within reach of thousands of dangerous idiots in a decade, even if it doesn't reach millions of consumers.