CRISPR Babies — Episode #1

Show Notes

Corey and Steve discuss news of gene edited babies in China, and the future of human genetic engineering.

▶️ Watch: CRISPR Babies — Episode #1

Resources

• Generation CRISPR?
• He did it: He Jiankui talk at HKU conference on gene editing
• Economist Radio podcast interview on Genomic Prediction
• Transcript
  
  

Creators & Guests

Host

Stephen Hsu

Steve Hsu is Professor of Theoretical Physics and of Computational Mathematics, Science, and Engineering at Michigan State University.

What is Manifold?

Steve Hsu is Professor of Theoretical Physics and Computational Mathematics, Science, and Engineering at Michigan State University. Join him for wide-ranging conversations with leading writers, scientists, technologists, academics, entrepreneurs, investors, and more.

Corey: Yeah, let’s begin to move into the topic of today’s discussion which is these CRISPR babies that recently generated in China.

Steve: CRISPR babies, designer babies.

Corey: And so this touches on your interest in genomics.

Steve: It does.

Corey: And so let’s get a little background about this. Let’s talk a little about CRISPR first. And I have to say, first of all, these babies aren’t in some sense a surprise. We’ve been talking quite a lot about when this was going to happen.

Steve: Absolutely. We’ve been anticipating it for several years.

Corey: There’s a theory which is, in technology anything that’s possible will eventually happen. And so you can’t be surprise by this. But give a little background about what you know about this particular case and what happened. Many people [inaudible 00:00:45] gene editing to some degree.

Steve: So recently, I’d say what?, five years ago, big breakthrough in gene editing technology. Something called CRISPR. I think you actually deserve credit for making Michigan state institutionally aware of CRISPR, running some big meetings on it, helping set up a gene editing lab here. So you probably actually, in a sense, know more about CRISPR than I do. But you could just say we now have the ability to very selectively edit a genome with relatively low, whether it’s exactly zero or not, low, off-target mistaken edit rate. And so there are now many possible applications. One of the applications is, if you’re going through in vitro fertilization and you have an embryo, you might edit the genome of that embryo. And you could edit it, edit the genome in a way, making a change that isn’t present in the mother or the father. So a de novo change in the child that makes it inevitably very different, genetically very different from the parents. And so what was done in this first experiment was they actually… Because many embryos had been edited before and testing had been done to see how effective the edit was. This is the first time the embryo was implanted and actually became, in this case, twin baby girls.

Corey: And so this has caused enormous controversy.

Steve: Yes.

Corey: It was announced at a meeting on genome editing in Taiwan.

Steve: Hong Kong, Hong Kong.

Corey: Hong Kong. Yeah, Hong Kong. And it’s interesting because I guess there… I saw an article in MIT Technology Review. It seems like there were signals early on this might be happening.

Steve: The reporter at MIT Technology Review, whom I actually know, he broke the story through some actual gumshoe reporting. He was kind of aware that stuff like this was going on, and he found the record, online record, of the experiment at this hospital website. So he actually broke the story.

Corey: So give us a little background. What was edited and what was the researcher trying to accomplish by doing it?

Steve: Right. So this researcher was a very interesting guy. His background’s actually physics. And he did his post doc at a very famous lab at Stanford, who’s run by another physics guy who went into genomics and molecular biology. So he’s a little bit of an outsider to the mainstream genomics world. He’s actually more a bioengineer. He has that physicist view of things like, “Oh, great thing. Let’s do it.” So this guy, who’s a professor at a university in China, he ran this experiment where he recruited couples to be part of the experiment. They gave them free IVF. And his goal, I think, all along was to produce the first gene-edited baby. And the edit that he made was an edit to a particular gene, CCR5, that actually, if it’s done correctly, and it’s not clear is was actually done correctly in these two, that it actually worked properly in these two girls, gives you resistance against HIV. And there’s actually about, I think the percentage might be 10% in Europe.

Corey: Something like that, yeah. Right.

Steve: There’s some percentage of the European population actually has this mutation and is very resistant to HIV. The father of the girls is HIV positive and was very, I guess, according to what was said, extremely traumatized by becoming HIV positive. And so he had a very strong desire that his daughters would be HIV resistant. Now, a lot of what’s reported in the press doesn’t take into account all the information that we know. Because this guy, the researcher’s named He, he gave a talk at this meeting in Hong Kong. And he went over his description of the experiment and what had been done. And a lot of the facts that… Now, again, he might have lying. But if he was telling the truth about what he did, those are important facts which are seldom reported by the press.

Corey: True.

Steve: So one of the things he said was both parents are highly educated. This process takes time because you have to go through the IVF process and collect the embryos, and then do the editing, and then sequence the embryos to make sure the off-targets aren’t there, and then communicate with the parents. So there’s a lot of time over which he was communicating with these parents. And so he claimed that they had a good understanding of the science and the technology behind all this. And what he told them at this point, when the embryos had been edited, was, “We found what might be an off-target in one of the embryos.” But when you’re sequencing an embryo, you’re only taking, I think in his case they just took one cell off the embryo. And so it’s a noisy method of sequencing.

Corey: How many cells are there at his point in time?

Steve: 50 to 100 typically.

Corey: Okay.

Steve: And so they take as few as possible. Because you don’t want to damage the embryo. So they took one cell off and they sequenced it. And he said to the parents, “We found some evidence of an off-target. We’re not sure, it could be an error in just the sequencing. We won’t know until the kid is actually born whether that is really an off-target mutation.” And he said that the parents were offered the choice of using, to complete the IVF cycle, either the edited pair or some other unedited embryos. There were unedited embryos that came from this mother and father. And they had the option of using those. So at that moment it was parental choice to implant. And the editing of embryos is not controversial generally. So that scientifically has been done many places around the world.

Steve: So the key decision here was actually made by the parents, not by the researcher. If the parents had just said, “Okay, let’s stop. Let’s just use the unedited embryos to finish the pregnancy,” you would’ve never heard about this. This would just be one of dozens of, “Well, okay, they’re testing the CRISPR technology. It works. We checked the embryo.” So the parents went ahead and had the kid. And it turns out the off-target isn’t really there. It was actually an artifact of the sequencing. And so that aspect of it, I have not seen reported anywhere. So it kind of changes the tenor of the whole thing.

Corey: Now I think it’s been really interesting to watch the response to this. Because there’s been uproar within the scientific community. And the researcher’s been denounced as having done something unethical. And I think there are an enormous number of complexities here, but I have to say I think the scientific position may not be tenable actually, the idea that you should be allowed to edit an embryo but it’s somehow immoral to implant the embryo. As if this embryo somehow needs to be obligatorily destroyed.

Steve: If you’re a Catholic… What’s totally legal in I think all of the United States, to go and abort something that actually looks like a baby. It’s not a bundle, a soccer ball of 50 cells, but a baby or what some people would call a baby. You’re allowed to abort that. How can there be moral equivalency between the question of, “Oh, I want to make one tiny change to the genome of this embryo, or implant or not implant this embryo,” those decisions seem very minor to me compared to decisions made every day by potential mothers about abortion, right? So I think people are just doing what they’re… So the thing they’re accustomed to, they don’t question. The new thing, they question a lot in this context.

Corey: I think it’s very common. One interesting issue is going to be how this will evolve as this practice probably becomes more common, whether people’s attitudes will change. There’s a Pew survey actually, a couple years ago, about attitudes towards gene editing. It turns out the US public is morally split right down the middle as to whether they would support editing to bring about exactly what they did in this case, which is disease resistance. People more religious were more opposed while people who knew more about CRISPR were more supportive.

Steve: Right.

Corey: But it seems like it’s an issue that [crosstalk 00:08:40].

Steve: At the very same conference in Hong Kong, the Dean of Harvard Medical School, who is himself I think a stem cell or, anyway, some kind of molecular biology researcher, he gave a talk. And he was on a panel in 2015 that set forth a bunch of guidelines for when CRISPR could be used in a clinical setting. And 2015 was far enough away that they could be rational as opposed to what people are doing now, talking about what happened. So he gave his talk and he actually just said, “Well, these are the things that we thought in 2015 would have to be in place before we could do it. And frankly, we’re pretty close. We’re pretty sure that it works well. It’s effective. It’s safe. We can carefully monitor what’s going on. We can make sure the parents understand and give informed consent.” So he actually said something which was quite supportive, not of this particular researcher, He, who everybody says is a villain now, but of the general trend. He was quite supportive. And then it was announced. I think it was revealed that some Harvard researchers are doing CRISPR on sperm right now. So it’s going to happen. There’s no question it’s going to happen.

Corey: It’s interesting, a lot of our casual conversations are [inaudible 00:09:51]. Again, the cat’s out of the bag. People are going to do this. We’re thinking, is it going to be Russia? Is it going to be China?

Steve: Right, right.

Corey: And those seem to be the two leading cases.

Steve: Right.

Corey: And it looks like it was China.

Steve: It’s funny because the Chinese scientists who are all sort of cowardly conformists, they were the most vociferous in denouncing He when this happened. And the most supportive guys were the Dean of Harvard Medical School and George Church, who’s also at Harvard Medical School.

Corey: That’s right. Yeah, George Church came [inaudible 00:10:17] criticism over this.

Steve: Yeah. So it’s funny. And the most amusing thing is if you go on YouTube and you watch the video of… This guy had a video all set up of just an interview of him in his own lab talking about what had been done, that he himself put on YouTube, when the story broke. And you read, there must be like 5,000 comments on YouTube under this video. And every comment from a Chinese person is like, “I am a life scientist and He has disgraced, has shamed the Chinese people, and he should never have done this because, well, the international people don’t like it.”

Corey: Is the last part key to that comment? [crosstalk 00:10:53]

Steve: Yeah, it is. It is. It’s all about, “Oh, well, if the international thinks this is bad, this is…” Anyway. This is YouTube, right? So then there are thousands of comments by people, mostly Americans and Europeans, saying, “Hey, it’s awesome man. Technology is going to happen. What are you coming down on this guy so hard?” So it’s amazing the dichotomy of those comments.

Corey: Of course.

Steve: Yeah. But in the US press, of course, “The Chinese, they don’t have morals like we do and they’re going to edit all their babies.”

Corey: People should just look at this Pew survey of Americans, where essentially half of them said that they thought that this was acceptable. So in this same survey, people were very resistant to things like editing to increase intelligence. And we’re going to get into this too because this is actually something I think that you’ve written about, that you are supportive of.

Steve: Right. Well,…

Corey: Okay, okay.

Steve: Let’s be careful about that.

Corey: Okay, we’ll be careful about that.

Steve: All right.

Corey: What’s interesting is this gene in particular has other effects.

Steve: Yes.

Corey: And this gene looks like it’s also involved in contextual memory, at least in mice.

Steve: Yes.

Corey: Even heterozygotes, where one copy of the gene is disabled and one is normal, those mice look like they had better memories than regular mice do.

Steve: Yep.

Corey: And the question is how will this work out in these children? It’s quite possible that these edits may give some cognitive enhancement in these children.

Steve: Yep.

Corey: Do you have any thoughts about that?

Steve: Well, so I’m pretty sure the parents didn’t do it for cognitive enhancement. I think the story about the father being really traumatized about by accident becoming HIV positive and wanting his kids to be resistant, it has the ring of truth to me. And it didn’t seem like the parents were trying to enhance. This is a weird way to enhance the intelligence of your kids.

Corey: Sure.

Steve: Because the effect is not that big. But whether we’ll see people making active edits to accomplish goals which are not related to disease, yeah, we might eventually see that. I think the biggest barrier, which people do not understand… So people are very focused on CRISPR, the technology, how well can we edit, are there off-target edits made. The thing people don’t understand is that the more informatical problem of actually figuring out what pieces of the genome are affecting which traits and in what way, that’s a huge machine-learning, AI problem. That’s what I work on. And that is not a solve problem. If you gave me today a super CRISPR, a vector, that could edit 1,000 places in the genome instantaneously with no error, we wouldn’t know how to use it.

Corey: Of course not.

Steve: Because we actually do not know the specific edits that we want to make. And so people misunderstand what is the gap, the barrier that needs to be crossed before all the nightmare scenarios of blond, six foot-five, super-genius decathlete babies can be produces. We haven’t solved the problem yet actually.

Corey: I want to say that I think part of the story is that this is new technology. But let’s be honest, people are already in the process of trying to produce blond, six foot tall babies by simply having Scandinavians be perhaps the largest sperm donors in the world.

Steve: Sure, right. Right.

Corey: [crosstalk 00:14:10] which is people get all upset about CRISPR. People are already making choice about their genome. They’re already seeking out particular kinds of mates. This is only using a very different technology. But the fact is, these kinds of decisions are being made by most women all the time.

Steve: Absolutely, yeah.

Corey: And men actually, if they’re looking for an egg donor.

Steve: Egg donation in the US is somewhat… I don’t know if it’s regulated. But statistics are gathered. And so, for example, there’s a famous plot, I think I put it on my blog a few years ago, but the data was gathered by researchers. On the horizontal axis is the SAT score of the egg donor. And then the vertical access is the price paid by the parents for the egg. And it’s like this. So, yes, people are already doing this kind of thing.

Corey: I willing to bet height is actually a similar [crosstalk 00:14:58].

Steve: Yeah. And if you look at the ads, if you go to the Crimson or the Yale Daily, whatever it is, the ad will usually say, “We want a smart, Yale student who’s also an athlete to donate eggs. And we’ll pay.”

Corey: There’s a famous case, I think maybe 10 or 15 years ago, of a young woman who was I think six foot tall, athlete, and she was selling her eggs for $50,000.

Steve: There’s you go, yeah.

Corey: So just to say, I think-

Steve: The free market, man.

Corey: Yeah, it’s interesting. I think people have gotten used to it so they don’t say anything about it. But I personally don’t see any principle difference between people making these kind of choices. It’s a matter of degree, it’s a matter of perplexity.

Steve: Principled, I agree. I totally agree with you. I know some of the people who were the early pioneers in IVF, IVF itself, which is not 30+, 40 years old I think. You can go back and read what were the reactions. What did The New York Times say? What did bioethicists say when the first test tube… I remember this being on TV.

Corey: Of course.

Steve: I was eating dinner, Dan Rather was on TV.

Corey: It was ’78.

Steve: Yeah, with my parents.

Corey: You were 12.

Steve: Yeah, I was a little kid and was like, “Dad, test tube babies.” And so I remember how freaked out people were then. Now, you might be surprised that a million test tube babies are born each year. In many European countries, 5% of all births are through IVF. In Denmark, it’s 10%.

Corey: Wow.

Steve: In Japan, it’s 5%. And it’s just a consequence of women having more professional, more career options and tending to have the kids later.

Corey: Sure.

Steve: And then there’s maybe a fertility issue. And so they go through IVF. But what was reviled and thought to be a milestone in…

Corey: [crosstalk 00:16:38]

Steve: Yeah, evil progress of science, cold technology, now has become a way to produce 5% or 10% of all babies in France or in Denmark.

Corey: It’s interesting. And in my state, Massachusetts, I think right around the turn of this millennium became the first state where more women over 30 had children than women under 30.

Steve: It’s exactly it.

Corey: And that’s clearly driving this.

Steve: I think all young women should be told that fertility decline in women happens between the ages of 30 and 40, for almost everybody. And if you’re lucky, it’s in your late 30s. And if you’re unlucky, it’s in your early 30s. And so there are many women who, “Wow, you look still pretty young and you’re looking for a husband. And you’ve got a great career going, you’re a lawyer,” you’ve already had fertility decline. You just don’t know it.

Corey: I think we have friends where we’ve seen it become an issue.

Steve: Exactly, yep.

Corey: But you’re involved in a startup actually that’s…

Steve: I’m a founder of a startup that does advanced genetic testing of embryos.

Corey: And how does that work?

Steve: Well, it’s very similar to the He, Megele story. I’m just kidding. So it’s standard practice now to let the embryo grow to 50 or 100 cells. And then you can take a few cells off for genotyping. And then they freeze it. They actually freeze it in liquid nitrogen. And the freezing and thawing doesn’t seem to harm the embryo. And so there are a lot of women who are actually advised, if they’re high-powered career women, to freeze embryos so that they can use them later. That little sample that’s taken off now, the DNA from those few calls can be amplified and we can get a full, relatively inexpensively, we can get a full 23andMe level genotype for each of the embryos. And then you can predict certain things like disease risk. So you can detect mutations that are dangerous and you can also predict polygenic disease risk. So, for example, the probability that someone gets breast cancer or the probability that someone had heart disease is actually controlled by many genes. But the machine learning is getting good enough that we can, from this measurement of the genome, actually make meaningful predictions about whether a particular embryo is going to be an outlier for a particular health risk.

Steve: And so if you have multiple embryos and you’re only going to use maybe one or two of these multiple embryos, you can now make a more informed decision than you could just a year or two ago.

Corey: And it’s known that taking these cells off of the embryo has no negative health effects.

Steve: So no detectable ones yet. So so far there have been no statistically significant results shown about the viability of the embryo post-biopsy. It could turn out there’s some slight effect from the biopsy. But so far it’s not been detected.

Corey: And there’s no negative effects to the freezing long term?

Steve: Actually, the freezing thing is actually positive, possibly for reasons that embryos that look okay but there’s some problem with the way that they’re developing, those, when they thaw out, are not viable. And the ones that are able to go through the thawing process are viable. And so the actual success rates of frozen and thawed embryos is actually significantly higher…

Corey: That’s interesting.

Steve: … than, yeah, than just fresh [crosstalk 00:19:49]

Corey: So it’s almost like an evolutionary test of fitness.

Steve: It is. That’s exactly what it is. It’s a filter for fitness. But it doesn’t actually impact most of the embryos at all.

Corey: And so where do you see this story going? We’re this far, these babies are now born.

Steve: And there’s another pregnant woman too, I’m pretty sure.

Corey: What’s your expectation for the next year or two?

Steve: Well I think the bigger story is actually selection, not editing, for the reason that I stated, which is that we don’t really… Even if I gave you a super CRISPR, at this moment in our scientific knowledge, we don’t know exactly what edits we would make. The predictors rely on correlations. If you’re going to edit stuff, you have to know about causation. And so you can know ahead of time that, “Okay, this distribution of genes in this embryo predisposes it to, it has a higher probability of having breast cancer,” but you don’t know how to change the genome necessarily to make it resistant to breast cancer. So that science I think is going to take at least another five or 10 years, if not more, to develop before we can, if you wanted to, go hog wild on editing. Because you need to know what to edit.

Steve: However, the strength of these predictors is increasing very rapidly because machine learning just needs data. And there’s more and more data. And so the number of traits that we can predict from genome alone is going up and up and up. And we don’t really even need to fully predict the trait, we just need to figure out who’s an outlier. So if I’m just trying to warn you about, “Hey, embryo four may have much worse health, much decreased expected longevity than the other ones. Maybe you should use one of the other ones,” that kind of differentiation is already possible, and it’ll just get better and better.

Corey: So how much does this process cost now via your startup?

Steve: I think the price… I’m not involved in the day to day of the startup. I’m one of the founders and I’m on the board of directors, but I’m not really directly involved. So I might say something that’s wrong. But I think the pricing is $400 an embryo for the advanced testing. It’s very affordable.

Corey: How does that compare to the price of the IVF [crosstalk 00:22:04]?

Steve: I think an IVF cycle, it really depends on where you do it. If you do it in Korea or Taiwan, it’s actually pretty inexpensive. It could be like a few thousand dollars for an IVF cycle, I think. But in the US, maybe it’s more like $5,000, $10,000. But it’s a [inaudible 00:22:17].

Corey: Sure.

Steve: It’s a small correction to the overall cost.

Corey: That’d be fabulous.

Steve: Okay, so let’s just cut it there. And until the next time.

Corey: Fabulous.

Steve: All right.