University Place

The Ethical Landscape of Human Genome Editing

– Welcome everybody, to this special Wednesday night here. I’m Tom Zinnen, I usually host Wednesday Nite @ the Lab, down at the Biotech Center. I wanna thank Rob Streiffer for inviting me to have this be part, this special, UW Philosophers at Work presentation be co-promoted and part of Wednesday Nite @ the Lab. I think it’s great that so many people are here. And now I wanna introduce Professor Michael Titelbaum, who is the Chairman of Philosophy. [audience applauding]

– All right, thank you all for coming out. I am Mike Titelbaum. I’m the Chair of the Philosophy Department here at the University of Wisconsin-Madison. It is my pleasure to welcome you to this year’s installment of our UW Philosophers at Work series. And I wanted to thank Wednesday Nite @ the Lab for co-sponsoring this event.

Also wanted to thank a couple people who were involved in organizing it. So a huge amount of work was done by Rita Richter, who is in our Philosophy Department office. Thank you, Rita. [all applauding] And also, Christy Horstmeyer came down and helped us set up tonight. So thanks, Christy. [audience applauding] Our speaker tonight is Robert Streiffer. He is an associate professor in the Department of Medical History and Bioethics and also in the Department of Philosophy. He is also the Philosophy Department’s

– Robert: Full!

– Sorry, full professor, my fault. [audience laughing] I would like to point out, he sent me that piece of information. I was just repeating what he said. So you did that! Right, full professor. I was about to say, almost more importantly, he is our semi-official, semi-professional photographer in the department. And that’s really what he spends most of his time doing. Right, Full Professor Streiffer, received his PhD in Philosophy from MIT in 1999. His research focuses on ethical and policy issues arising from the use of modern biotechnology. He has served on numerous local oversight committees, and that’s really important work that gets done around our university and around our community. These include UW-Madison’s Health Science Human Subjects Committee, the Hospital Ethics Committee, the Human Genomics ELSI Committee, the Biotechnology Advisory Committee, and the Embryonic Stem Cell Research Oversight Committee.

– Robert: Not all at the same time though.

– Right, yeah. He has also served as the Chair of UW’s Letters and Science Animal Care and Use Committee. And as a member of UW campus’s Animal Planning Advisory Committee, he has presented to the National Academy’s Human Embryonic Stem Cell Research Advisory Committee, and to the German Ethics Council on the ethics of creating part human, part animal chimeras for research purposes. His scholarship on informed consent and human embryonic stem cell research led to multiple academic institutions, as well as the U. K. Stem Cell Bank, modifying their rules governing how human embryonic stem cell research is conducted. He also effected the National Institute of Health’s guidelines on human embryonic stem cell research. This is all a long-winded way of saying that we are very fortunate to have a true expert speaking to us tonight, on the ethical landscape of human genome editing. The way this is gonna work is Professor Streiffer is going to give his presentation, and then he will field questions at the end. So please join me in welcoming him. [all applauding]

– Good evening. So this is Dr. He Jiankui. He’s a researcher at the Southern University of Science and Technology in China. In November last year, November 25th, he made an announcement that he probably hoped would be received as a tremendous scientific and clinical breakthrough. He announced that he had assisted in the creation of the first genome-edited human babies, who had just been born a few weeks before. He probably didn’t get the reaction that he was hoping for. Within days, he had multiple investigations by multiple groups: his university, local health officials, Chinese authorities investigating his work. Statements started being produced by professional societies dealing with genetics and genome editing coming out against his work. By January, a preliminary investigation had been concluded and found that he had broken multiple laws in China and quote, that he would be quote, “dealt with seriously. ” And once that report was filed, he was fired from the university. So perhaps not the reaction he was hoping for.

We’re gonna talk about that, but it’s also an opportunity for us to talk about some more general ways of thinking about the editing of human genomes. And so I’m gonna give you some arguments and concepts and distinctions that have played a prominent role in this debate, and I’m gonna argue for a pretty modest and wise and reasonable policy position at the end. I’m gonna give you a brief background on CRISPR. This is not a science talk, this is an ethics talk. But it’s good to have a common ground to start with on what we’re talking about. CRISPR is one of the latest tools, continues to be developed and refined, that can be used to edit genomes. I’m gonna introduce two standard distinctions that have played a, sort of overpowering role almost, in some of the discussions about human genome editing. And I’m gonna argue that certain germline treatments and even germline enhancements are sometimes ethical. So I’ll give you an argument for that and I’ll explain some of the terminology. And at the end, we’ll go back to Dr. He and talk about why he got into so much hot water. So a very abbreviated timeline.

Go back about 8,000 years ago and you wanted to modify the genes of an individual, pretty much the only thing you had at your disposal was selective breeding, right? You look at the parents, you look for traits that the parents have that you think are gonna be advantageous, you breed them, and you hope they’re passed on to the offspring. And this can, of course, lead to tremendous advances over long periods of time. But if we take a quick step forward to 1987, we get the discovery of an odd genetic phenomenon. So these are basic researchers looking at DNA in bacteria and other kinds of microorganisms, and they sort of see something odd. And they dub it CRISPR. We’ll talk about what that is in a minute. But they identify these in lots of different bacteria, but they didn’t know what they did, they didn’t know what their function was. 2012, lot of effort being put in to sort of figuring out what these stretches of DNA do, and researchers at UC Berkeley figured out that they could use this DNA to actually make targeted alterations to DNA.

So they could go to specific points in the genetic sequence, right, in the cell of an organism, and make changes at that particular point. So we’ve been able to do alterations in DNA for a long time with the development of genetic engineering in the 1970s. And we have some ways of doing targeted changes, but they’re very inefficient, and I’ll give you some numbers a little bit later. And this was a pretty remarkable breakthrough. The people who collaborated on this and some other developments here, are commonly talked about as in the running for Nobel Prizes. 2013, the year after the UC Berkeley researchers made their discovery, researchers at the Broad Institute in Boston and Cambridge showed that CRISPR can in fact be used to edit the genomes of mouse and human cells, so doing work in actual cells. 2015, we get researchers who use CRISPR to correct a gene associated with a genetic disorder called beta thalassemia, which is a blood disorder, in non-viable human embryos. So they do alter the genetics of a human embryo, but these are embryos that would have never been implanted, they were created through in vitro fertilization in the course of fertility treatment, but these were embryos that had been fertilized by two sperm, so they have an extra nucleus running around and they’re not gonna develop. And so they use this as an opportunity to try to figure out how to do this kind of technology in human genomes, but not with embryos that were gonna be implanted, right? So still just research in the lab. Still generated a fair bit of controversy.

And then in 2017, you have another group of researchers who use CRISPR to fix some genetic defects in viable human embryos, but were then destroyed. So not implanted, not brought to term, not for reproductive purposes, but now we’re actually able to show we can do this with some efficiency, it illustrated all kinds of problems as well, but at least sort of proof of principle that we could use this in viable human embryos. And then in 2018, Antonio Regalado, a science writer for MIT Tech Review, and reporters at the Associated Press, broke the news about Dr. He and his announcement. And so we actually have CRISPR-edited babies who were brought to term and born. So what’s CRISPR? CRISPR is a family of tools; the distinctions are not of much concern to us. The most common one is something called CRISPR-Cas9, but I’m just gonna talk about CRISPR. CRISPR’s an acronym, stands for Clustered Regularly Interspersed Short Palindromic Repeats, which describes the DNA that those researchers in 1987 found in the bacteria. And then Cas, of Cas9, stands for CRISPR associated proteins. And it took awhile to figure out, but researchers now think that this is complex of molecules that provide bacteria with an adaptive immune system to help defend themselves against viruses.

So viruses that can eat DNA latch on, they insert their viral DNA into the bacteria to attack it, the bacteria have various defenses that they can use to try to defend themself against that attack from the viral DNA. When those defenses start cutting up the viral DNA that’s been inserted, the bacteria grab some of that DNA and saves it. And basically, puts it in a little DNA-like database and saves it forever. And now, if the same kind of virus attacks this individual bacteria later, it matches the viral DNA that it’s being attacked with, with this copy that it has saved, and it basically says, “Oh, I gotta act now,” and it’s essentially immunized against that viral DNA, gets rid of it very quickly, because it uses that DNA as a guide, or it uses RNA that’s constructed complementary to that DNA as a guide to go and find that DNA in the cell, and then uses these Cas9 enzymes to cut it up into pieces rendering it dysfunctional, okay? So it’s a way for the bacteria to learn, over time, who the invaders are and to have a more active and prompt response to those invaders, okay? So by modifying the guide DNA that’s used to determine where the Cas9 endonucleoleases cleave DNA, CRISPR-Cas9 can be used to insert or delete single nucleotides, insert or delete genes, and even make epigenetic changes to gene expression. And here’s a nice slide from Ellen Jorgensen. And if you want more detail about the science, there’s a huge amount of educational resources available online. So she describes Cas9 as essentially a warhead that’s on a leash, so it can basically destroy DNA. And then the guide RNA is a targeting mechanism that is holding the leash, and so the guide RNA travels around in the cell until it finds the match for it, and then it uses the Cas9 warhead to obliterate the viral DNA, okay? And the sort of key here that makes this such a revolutionary development, was that you can change where these Cas9 systems, these CRISPR-Cas9 systems target, merely by rewriting this short little stretch of RNA. Like, 20 base pairs long. And it used to be that we could do that, in terms of targeting changes, but it required a huge amount of work and a lot of effort and time to sort of develop a tool for each and every different target that we wanted to do.

Now we just come up with this short RNA guide and it does the rest. Obviously, I’m oversimplifying a bit, but that’s the rough idea, okay? If one imagines a DNA of a cell as an instruction manual, a CRISPR system can be imagined to sort of go to any particular line in that instruction manual and make a tear in the page, okay? So it’s cutting the DNA, it’s tearing the page, right, the DNA is like an instruction manual for the cell. The cell has natural repair mechanisms that try to fix that break in the DNA; you can imagine the instruction manual sort of trying to be fixed by matching up the two sides of the tear again, and putting some Scotch Tape on, but it doesn’t always line up just right, right? Any of you book lovers know that, right, you get that tear and you can’t, and it may be that in the end, the tear actually covers up a line, right? You’ve deleted an instruction out of the instruction manual. So similarly, Cas9 can go in and disrupt a gene, because when those repair mechanisms come in and it tries to reconstruction things, it doesn’t reconstruct that gene again. So that gene is now knocked out. Or, they figured out that if you put template repair DNA sequences in high concentrations into the cell, then the natural repair mechanisms of the cell, when they’re trying to fix that break, will actually use those DNA sequences to fix that repair, but those DNA sequences are ones that we’ve constructed, right, so we put in whatever gene we’re interested in studying, and now the natural repair mechanisms of the cell will actually put that gene into the DNA of the organism that’s being edited. So we can also use this to add genes. So knock out stuff, knock in stuff. What’s different about it? Well from a sort to philosophical perspective, there’s not that much different. But from a scientific perspective, it’s huge.

And that has some sort of practical ethical implications that we’ll talk about. It’s much more efficient that older methods. It’s more accurate; there are fewer unintended effects. You don’t get tears at places in the instruction manual that you don’t want them, as often. You can modify the genome at several points at once. Alterations that would cost thousands of dollars using old technology can be done in under a hundred dollars now. Things that would have taken months can now be done in a few hours. So tremendously rapid tools for people who are studying genetics, developmental biology, all kinds of things. And here’s a quote from the National Academy of Sciences 2017 report on genome editing. They say, “When conducted carefully “and with proper oversight, “gene therapy research has enjoyed support “from many stakeholder groups.

“But because such technologies as CRISPR-Cas9 “have made genome editing so efficient and precise, “they’ve opened possible applications that have until now “been viewed as largely theoretical. ” So as a philosopher and a bioethicist, I like thinking about futuristic scenarios and science fiction scenarios. And we’ve been talking about genetically engineering human beings since it was a spark in somebody’s imagination. But now we actually have some babies, right, who’ve been genetically engineered. It also continues to be developed, so it’s an incredibly rapid area of research. A few weeks ago, people who are interested in the science here, probably saw a series of articles and publications about something called prime editing, which can change any DNA base into any other, and can insert or delete any stretch of DNA more efficiently and precisely, and with fewer unintended consequences than even old CRISPR. So CRISPR’s already showing its teeth in its original form. And we’ve got some interesting speculations from some of the scientists involved. “Prime editing,” they say, “substantially expands “the scope and capabilities of genome editing, “and in principle could correct about 89% of known “pathogenic human variants. ” So those are genetic variants that have genetic causes that they feel pretty confident that prime editing could actually fix.

And the refinements keep coming. According to a report by Vox, there were 100 papers indexed by Google Scholar on CRISPR in 2011. In 2018, there were 17,000. And I checked the other day again, and we’re up to 35,300 hits in Google Scholar. So a very fast-moving area. It’s a very broad tool; it can be applied to lots of different situations: isolated DNA, microorganisms, cells, plants, animals, insects, human beings. And if you plug CRISPR into your Google News search or whatever you prefer to do, you’ll find stuff that’s going on. And it’s heavily used in areas of basic research. People are doing genetics work. But we’re mostly gonna talk about actual or potential clinical applications with humans.

So we’re gonna talk about somatic cell editing, which is using CRISPR in this context to edit the genomes of somatic cells, so these are non-reproductive cells, so cells like skin cells. And then germline editing, so reproductive cells: eggs, embryos, sperm, or the progenitors of sperm and egg, from which the sperm and egg are derived. So somatic cell and germline editing is what we’re gonna be focusing on. I’m gonna use the term intervention as kind of a broad umbrella term, covering basically any kind of medical intervention, and then I’m gonna draw some distinctions within that category of interventions. And in particular, as I mentioned, there are two distinctions here that have been discussed extensively in this literature on CRISPR, and also going back many, many decades in the general discussion about genetic engineering, and these kinds of distinctions are used in the National Academy of Sciences recent report and in science opinion and commentary pieces. So these are sort of all over the place. So one is the distinction that I just mentioned between somatic cell editing, where you’re editing a body cell. And then germline editing, where you’re editing a reproductive cell. And obviously, the important difference there, both scientifically and ethically, is that if you modify somebody’s somatic cell, it’s not going to affect the reproductive cells, unless you’re in some sort of special circumstances, it wouldn’t normally. And so the change doesn’t really have any significant risk of being passed on to offspring.

It’s not gonna be passed on to future generations. Whereas if I make a change to a sperm or an egg or an embryo, then that is going to affect future offspring, so not just the individual who is edited, but also any offspring that they might have and any offspring that they might have. So that’s gonna be an important distinction. And then the other distinction is between treatment and enhancement. And this is subject to a lot of definitional argumentation, which I’m gonna try to avoid, just by sticking to what I take to be kind of clear examples. But there’s a lot of discussion about how to draw this distinction and what the metaphysics are, and what the ethical implications of that are. I’m gonna focus on the ethical implications. So these two distinctions then give us a possible, right, 2×2 grid, so we can talk about somatic cell treatments, and we can talk about somatic cell enhancements. And we can talk about germline treatments and we can talk about germline enhancements. And we can start asking what the standards should be for a kind of research that falls into one of these boxes, what standards does it have to meet in order for it to be ethical? And I’m gonna introduce a technical term here, and I apologize ’cause it is a term in ordinary use, but I intend to be using it sort of stipulatively.

And this is gonna be a term, responsible interventions. So as a general rule, routine medical interventions of almost any kind, right? It could be anything from a liver transplant, to resetting a broken bone, to a crown. Routine medical interventions are ethical just in case they meet certain, very run-of-the-mill, ethical standards. And in the United States and most other countries that have similar medical systems, these aren’t really under dispute, although, of course, we can quibble about some of the edges of these concepts. But things like, intervention should be done with informed consent, treatment should only be done if they have an acceptable risk/benefit ratio for the patient, treatment should be provided to people on an equitable basis, and so on, right? So I don’t know how to give you a complete list of what the standards are, but whatever you think they sort of are for regular treatment, call that, call a treatment that meets those standards a responsible treatment, okay? So we’ll say that an intervention is responsible, just in case it meets those run-of-the-mill, ethical standards, and there are lots of others that we could list, but time is limited, okay? So let’s think about our first box here, somatic cell treatment. I’m gonna propose that a somatic cell treatment just has to meet the exact same standards that all the other medical treatments have for it to be ethical. So if somatic cell treatment is responsible, in that it meets those standards, then it is ethical, and there’s nothing more to argue about. We can argue about whether it meets those standards, but if it meets those standards, then it’s ethical. And here’s an example to try to motivate that. So in the past year or so, we’ve had four early stage clinical trails that have started in the United States using CRISPR to do somatic cell treatments.

So it’s now at the stage where the FDA and researchers and clinicians think that it’s safe enough, not to actually use in a clinic yet, but at least to start clinical trials to test for safety and efficacy and figure out how to improve it. This is Victoria Gray; she has sickle cell disease, which is a recessive condition caused by a single base pair mutation. So a single mutation in her entire genome results in this condition. This condition effects hemoglobin, which helps red blood cells transport oxygen through the body. It makes it so that the hemoglobin clumps up in a certain way, so that it doesn’t live as long, so it reduces the lifespan of the red blood cells that it’s being used by. And it makes it so that the red blood cells can’t actually move through the blood vessels smoothly. So you end up getting blockages, and can cause sort of intense pain, blockages in blood vessels, and you end up with a shortage of red blood cells. And so you can end up with very severe anemia. Researchers propose to take cells out of the patient’s blood and edit them in a Petri dish using CRISPR so that they produce a slightly different kind of hemoglobin that doesn’t have the problems that their natural gene has. And this significantly ameliorates the actual symptoms of the sickle cell anemia.

So the cells don’t clump up in the wrong way, they travel more smoothly. So they do that change outside of the patient’s body. So the genetic alteration is actual done in vitro, not in vivo, and then they inject those cells back into the patient, where they can do their work if all goes well, all right? So we could have a long discussion about whether or not this is responsible or when it would be responsible to use in a clinic. There are various risks associated with this treatment. There are various risks associated with CRISPR. So it’s much better than the older methods, but it’s still not perfect. It generates off-target effects, so it causes changes at places other than where you intend. And even if it were to only cause a change where you intend and to cause the change that you intend, we’re not perfect at figuring out what the implications of that change are, right? We might want it to change one thing, and it might do that, but it might also change 15 other things, right? So genes can have lots of complicated interactions. So we can have lots of discussion about at what point this becomes responsible, but I would submit that for this kind of thing, once it is responsible and it meets the same standards that other medical interventions have to meet, then it’s ethical. The mere fact that it uses CRISPR does not sort of raise any additional obstacles to this being an ethical intervention, all right? So proposal, for somatic cell treatment, it’s ethical if responsible, and a little bit more practical importance, it sometimes is responsible.

So we will reach a point where this can be done in a way that meets the standards necessary for a medical intervention, and when we’re at that point, it will be ethical, all right? So it’s not just sort of a hypothetical, if it ever meets those conditions. No, it actually will meet those conditions, okay? So that’s the first box. What about germline treatments? So now we’re going to initiate a change that’s going to affect the reproductive cells. So it’s going to affect the developing individual, but it’s also gonna be passed on to that individual’s offspring if they choose to have any, okay? Well, what would your view sort of look like, if you agreed with me so far, and thought that a somatic cell treatment, right, as long as it meets the normal standards is ethical, and of course it can meet those standards, but you wanna sort of draw a bright red line and say, “No germline treatment, that’s too much,” okay? “We’re not gonna allow that. ” Well, it seems like you’ve got sort of two options, if that’s what you’re trying to do. It could either be that the standards for somatic cell treatment and germline treatment are the same, but for practical, empirical, contingent reasons, it might be harder, perhaps even impossible, for germline treatment to meet those same standards, okay? So you’re not changing your sort of ethical perspective. It’s just that when you apply that ethical perspective to somatic cell therapies, it turns out that they can meet them, maybe soon. But with germline therapies, for sort of practical reasons, not gonna happen. That’s one option. Another option would be that in fact, the standards shouldn’t be the same, all right? We’ve got one set of standard for interventions and for somatic cell therapies.

But when it comes to germline, we actually need to add some additional ethical principles to our arsenal. And that’s gonna make things a little bit stricter, right? It doesn’t just have to be responsible, it has to be responsible, as well as meet these new requirements before it’s ethical. And if those requirements are hard to meet, it might justify sort of drawing a sharp line. Even though you agree, I’m supposing, that somatic cell treatment can be responsible and ethical, but nonetheless, the standards for germline treatment are different, harder to meet, maybe we’ll never be able to meet them, okay? Let’s look at the idea that the standards are the same, but the reason why we’d wanna draw a red line here are that they’re impossible to meet in the circumstances of germline therapy. Well, what are some of the standards? Informed consent. Well, almost the first point in any ethics discussion about germline engineering is you can’t get informed consent from the offspring of the embryo that you’re editing. And since you’re editing an embryo, you can’t get informed consent from the very individual upon whom you’re performing the intervention, right? Can’t get informed consent from an embryo. And can’t get informed consent from subsequent generations. So if informed consent is one of the standards, you’ll never be able to meet that standard, right? We’re not saying there’s different standard, it’s exactly the same standard, but it can never be met in germline therapy. Don’t think this works, because for regular medical interventions that are performed on, for example, babies and young children, other individuals who aren’t capable of giving informed consent, we actually don’t require informed consent from the patient.

We’re happy enough with informed consent from the parents or the guardian. And in the case of germline therapy, we can get consent from the guardian or parent, depending on how you wanna understand the relationship between gamete providers and the embryos that are created with their gametes. We can get informed consent from the parent or guardian in that case, they’re the ones who, presumably, would be requesting the germline therapy on behalf of their embryo or their future child, right? So if we properly understand the standards, germline therapy can meet the informed consent requirement, all right? Now there’s a little wrinkle here, because we’ve got the parents and we’ve got the embryo, well I’ve just pointed out that you can get consent from the parents, and when you’ve got an individual who isn’t competent to give their own consent, then going to a guardian is fine, what about when this embryo turns into a baby and then grows up and then has a kid of their own? Who’s the parent now? Well, it’s the first generation of the modified individuals. Well, we didn’t get consent from them to modify their offspring, so once we’re past the first generation, we actually can’t get consent, either from the embryo, ’cause you can never get consent from the embryo, or from the embryo’s parent, because it’s something that we already did before we could have their consent. So there is an interesting wrinkle here, but I think this still doesn’t ground a good objection to germline therapy treatment, because in cases where you can’t get consent even from the parent, we still allow treatment, we just decide on the basis of the interest of the individual in question. So if you have a child who comes into the ER and the parents aren’t around and we don’t know who they are, and there’s some harm that’s gonna happen if this kid doesn’t get treated right away, we don’t stand around and say, “Well, we don’t have informed consent, so we’re gonna “let this individual die for want of treatment. ” No, we make it based on a judgment about what’s in the interest of the individual. We’re imperfect, we could get it wrong, but that’s what we do and we do it as well as we can, right? So in this case, where you’re talking about the second generation of engineered individuals, well if it’s actually in their interest to have this alteration, then the fact that we don’t have consent is no more a deal breaker than it would be to say, “Well, we don’t have the consent “of the parents,” in an emergency situation, right? But now that does beg the question, right, “Is this a alteration that is in the interest “of that second-generation child?” So in the absence of consent from the parents, we look to this sort of best interest standard. Well, is it in the best interest of the child? But that then takes us to the next standard, right? Is there an acceptable risk/benefit ratio for the patient or the individual upon whom this intervention is being practiced? Well, if there was an acceptable risk/benefit ratio in the case of an analogous somatic cell treatment, it seems like in principle there could be an acceptable risk/benefit ratio in the case of a germline treatment. In the somatic cell case it’s got risks, right? Might have these off-target effects, but it might help with the sickle cell disease, right? Well, that may be also true of the second-generation individual, right, if we do a germline intervention instead.

So they would have had sickle cell, but now they have this other gene that minimizes the effects. So while it’s certainly true, that when you make a decision to genetically alter an embryo, or a sperm, or an egg, you’re gonna get changes that get propagated down the generations, and if there are any downsides, well those should be taken into consideration. But if there are any upsides, those also have to get taken into consideration. And the mere fact that you’re iterating it doesn’t mean that you can’t have an acceptable risk/benefit ratio, because the costs get iterated, but so do the benefits, right? So yeah, we’re putting more individuals at risk if something bad happens, but we’re also benefiting more individuals if something good happens. So the ratio could still be acceptable, okay? So it seems to me that we could have an acceptable risk/benefit ratio in the case of germline engineering, in cases where you would expect to have an acceptable risk/benefit ratio in the case of a somatic cell therapy. I’m gonna skip over some stuff for time. Equitable access to these kinds of things. These are gonna be expensive. Some of the genetically engineered treatments, some of the treatments involving genetic engineering run half a million dollars, and this is gonna be, maybe cheaper, but maybe not. Gonna be very expensive, that’s a huge issue.

I have colleagues who spend their entire careers looking at access issues and inequalities and justice. But that’s not something that’s peculiar to germline edits, right, germline treatments. That’s also true of somatic cell treatments. And so if you think that kind of obstacle can be overcome for somatic cell therapy, then it seems like it could be overcome for germline therapy, as well. It’s a problem, we should deal with it, but it’s not gonna justify drawing a sharp line between somatic and germline treatment, right? And in fact, it’s not clear that it provides much of a reason to withhold care from anybody, right? If you’re worried about access, it’s sort of the last option to say, “Well, let’s cut access for everybody,” right? What you wanna try to do is improve access for everybody and you wanna try to make sociopolitical changes that make the inequalities that were limiting access less grave to begin with, right? You don’t wanna just say, “Well, if everybody can’t have it, “then nobody can,” right? That doesn’t seem like a good solution. So same standards seems like there’s no reason why, in principle, germline treatments couldn’t meet those standards just like somatic cell treatments do, okay? What if they’re not impossible to meet, but they’re more difficult to meet, all right? So it’s not, sort of, in principle, impossible to meet the standards, but it is gonna be harder. I think is this probably true. So we have more experience with somatic cell treatments than we do with germline treatments. We have more people working on it. We have clinical trials underway, we have a lot more animal work and in vitro work and all kinds of stuff.

And there’s a wider array of situations in which germline treatments would probably be unnecessary. But it is worth noting that there is kind of a wrinkle here, so this was noted by George Church, who’s a geneticist at the Harvard Medical School. So if you’re doing somatic cell treatments, you’re basically editing the genomes of hundreds of thousands or millions of cells outside the patient’s body and then injecting them back in. Well, every time you edit a cell, you run some risk of these mistakes and these off-target edits. So you’re doing something hundreds of thousands, millions of times, and each time you do it, you actually increase the risk that you’re gonna, for example, introduce a cancer-causing gene or get rid of a tumor-suppressing gene, which can be just as bad, right? Whereas, if all you’re doing is editing a single-celled embryo or a single sperm or a single egg, you just have to do it once, so lower risk, other things being equal, at least, right? It depends on what you’re editing for and all those kinds of things. But so there is sort of a wrinkle here. But I’m happy to say that yeah, maybe this will be a little bit harder to, it’ll take a little bit longer, be a little bit harder to make sure that this research or this treatment, when it gets to that point, will actually meet the relevant standards. But it’s not sort of a red line. That’s basically saying, “Look, this research is still “ethical if it’s responsible. “Germline treatment is ethical if it’s responsible.

” And it can be responsible, sometimes it is. It’s just not responsible as often as somatic cell treatment ’cause it’s harder to meet those standards. So conclusion here’s gonna be that germline treatment sometimes ethical, but perhaps not as often as somatic cell treatment. But no sort of bright red line here. No justification for a long prohibition, okay? What if the standards are actually different? What if we think the standards, the ethical principles that we use to judge the acceptability of germline treatments should actually be stricter and different from the ethical principles that we use to judge somatic cell treatments? Certainly some commentators seem to be invoking different standards. Some of these seem to be relatively minor tweaks on the standards for routine medical interventions. So people talk about how we should be more rigorous in our informed consent procedures because this is new and people may not understand it and it’s controversial. People talk about how maybe we should have, sort of a more thorough understanding of the risk/benefit trade-offs that are involved in this area than we might for a new dental treatment or something like that. People who are expert in the science and bioethics and clinical medicine who have talked about this seem to agree that we shouldn’t proceed in this area using germline therapy until we have sort of an especially thorough understanding of the kinds of unintended or harmful effects that CRISPR might have, and a better grasp on how to minimize those effects before we proceed with genome engineering of the germline. I’m willing to concede that.

Whether those really count as different standards, as opposed to just arguing that we should have better standards across the board, ’cause I think you could make that case pretty easily, it’s not clear. Maybe it’s just that they disagree and they think the standards for everything should be a little bit better. But no need to quibble here; these seem sort of plausible enough. So I think that even with somewhat stricter standards, we’re still gonna get not a red line here, but just a cautionary, take things slowly kind of approach. So even if this is responsible, it could still be unethical until it meets this somewhat stricter standards. But that could probably be done in the near future. So no sort of red line. But other commentators seem to be invoking standards that are sort of radically different from the ones we use when we’re assessing normal medical interventions. And the ones that seem to be at play in somatic cell treatments. And to put my cards out on the table, they seem implausible to me.

So this is part of a statement from the Center of Genetics in Society, in response to Dr. He’s announcement. So, “Claim of genetically modified babies: if true, “a grave abuse of human rights. ” As part of their argument in this press release for a moratorium on germline editing using CRISPR, they point out that Dr. He has reportedly applied for patents on the procedures that he used, and is chairman and co-founder of a DNA sequencing company. Well, I have a bias here. I work at UW, we have WARF. If we were gonna say that a clinical researcher here, their involvement in patenting or a company that they started up or something like that meant that that area of research should be banned, we wouldn’t have an clinical researchers [laughing] left on campus. So this is an incredibly strict standard. I submit, implausibly strict.

So this would get to a stronger line between somatic cell treatment and germline treatment, if these were the standards for germline and not somatic cell, but it’s just not palatable, right? We don’t think that the mere fact that somebody has a conflict of interest or is interested in patents means that the research itself should be banned. So we’ll get pink here, okay? Much stricter standards would make it harder for this work to meet the relevant standards, because it’s gonna take a long time to root out all these financial conflicts of interests and it’s gonna take a long time to extricate clinical research from the patenting world. So we’re talking about a pretty significant delay on these much stricter standards, but I don’t think those much stricter standards are very plausible. We don’t apply them in other cases that seem to me to be analogous, not that they don’t raise significant moral issues. So I think patenting does raise significant issues. I think conflicts of interest raise huge issues. But we don’t think that their mere presence justifies a ban on research until that stuff gets removed, right? That’s too strong, okay? Another group, this is a prominent group of bioethicists and CRISPR researchers, including one of the three main developers. They called for moratorium on heritable genoming in March of 2019. As part of their justification, they suggested the following standard. They said part of the reason why there should be this moratorium on heritable genome editing is that many religious groups and others are likely to find the idea of redesigning the fundamental biology of humans morally troubling.

So apparently now it’s a standard for science that we can’t do anything that religious groups find morally troubling. I don’t find that standard plausible. It’s not a standard we do adhere to, and it’s not a standard that we should adhere to. Respectful engagement and discussion with all interested parties, absolutely, of course, right? We wanna have those discussions. And if those discussions bring to light legitimate reasons for the state to restrict this kind of research, then, right, we should implement appropriate policy that’s responsive to those legitimate reasons. But the mere fact that some religious group, even many, even all religious groups find some kind of academic and scientific research, especially around reproduction, morally troubling, we don’t need to wait for that before this can be ethical. That is not a relevant standard. So public policy in general, and especially public policy that restricts academic and scientific research surrounding reproduction, should not have to meet this standard, it doesn’t have to meet this standard. They also note that unequal access to technology can increase inequality, right? A worry that I mentioned a minute ago. It certainly can and often does.

And that’s a moral problem, as I readily concede. But again, the solution there is to increase access and to try and resolve the inequalities that make those limitations on access so problematic. Very rarely, I won’t say never, but very rarely is the solution there gonna be, “Let’s ban the technology,” right, let’s let the technology do what it can do. Useful, beneficial, but try to resolve the socioeconomic and political institutions that are really the underlying problem, all right? Other commentators, not these guys, they’re a little bit more careful here, but other commentators seem to be operating on the assumption that germline editing should only take place if it’s known to be perfectly safe and risk-free. So they say, “Off-target edits, they happen. “Therefore we can’t do it. ” Well, that’s not a good argument, all right? What the level of acceptable risk is for these technologies, or any technology, is very difficult, perhaps impossible to pin down with any precision. But we certainly know that it’s more than zero, right? Any request for perfect safety and zero risk, in any area of technology is unreasonable, doesn’t have to be met. We can have good-faith discussions about what the acceptable level is, but it’s not zero. It’s higher than that.

And interestingly, George Church, the geneticist that I mentioned a minute ago, notes that the spontaneous error rate on natural mutations is on the order of . 1 to 10 mutations per cell. So as your somatic cells right now are dividing, they’re introducing mutations, off-target, unregulated, undesigned, right, mutations that are happening randomly, while we actually have CRISPR techniques that are orders of magnitude better than that already, and they’re only gonna get better. So if you think, problems here, but if you think safe enough is, well, it’s as safe as what’s already going on, then we’ve kind of met that threshold. Now I’m not sure that that’s a good baseline, right? Because you wanna ask, “Yeah, “maybe that’s how things work in nature, “but maybe we can do better at very little cost,” right? So we always wanna be looking for ways to try and improve things and to make sure that the risk/benefit ratio is as good as we can get, even if we’re well beyond what nature would provide, right? So but it is an interesting question there. Finally, there seems to be a consensus among the commentators in this area that there should be a delay on germline editing until there’s a broad societal consensus about the appropriateness of the proposed applications. So this has been endorsed by the National Academy of Sciences and several other prestigious groups. I’m worried about this the same way I’m worried about the religious groups that might be upset here, all right? Might not have a social consensus. We should work to get that, we should try to work together and come up with good compromises that satisfy everybody and leave everybody feeling respected, have their values taken seriously and incorporated to the extent that they can be, but if the lack of a consensus is due to irrationality and lack of information or religious belief that shouldn’t play a legitimate role in setting public policy, then the fact that we don’t have a consensus shouldn’t hold up the research, at least not for principled grounds, I mean, there might be good practical grounds for making sure that we don’t do something that pisses so many people off that they end up, right, defunding the NIH or something, right? That would be bad. So there are practical strategic considerations here, but at least from an ethical, principled perspective, I don’t think this is important, okay? So these kind of standards that are much, much stricter, right, perfect safety, social consensus, satisfy all these religious groups, that would draw a pretty sharp red line here between somatic cell and germline treatments.

So this would sort of justify the view that even if this stuff is, it meets these standards of responsibility, it’s still unethical until it meets these much, much stricter standards, and that’s gonna be practically impossible to do, so maybe we don’t sort of ban it forever, but in effect, it’s gonna be a ban that will never be lifted, ’cause we’ll never satisfy those standards. But I think you only get that, if you adopt these implausibly strict ethical standards, okay? So I’ve been suggesting that we should end up somewhere close to this yellow box in the lower left on germline treatments. So even if responsible, even if it meets those standards for normal medical interventions, it will be responsible perhaps less often than somatic cell treatments, maybe it’s a little more risky, harder to do well, and it might still be unethical unless it meets somewhat stricter standards, right? So maybe we have to have a better understanding of the risk/benefit trade-offs in this than we might insist in another area, but these are things that could be handled in the near future. So a cautionary approach, not a sort of prohibition approach, okay? So if you’ve agreed with me about the green top left box, then I think you should agree with me about the yellow bottom left box too, okay? Now let’s go for the top row, thinking about treatment versus enhancement. And this can go pretty quick, ’cause I think this doesn’t have a lot of merit. What would make it the case that a somatic cell treatment could be ethical, whereas a somatic cell enhancement can’t be ethical, right? We’re gonna put a green box in the somatic cell treatment, where we’re gonna have red for the somatic cell enhancement. Well, I submit that this would only be plausible if the goods that were promoted by somatic cell treatment were always sort of substantially more important than the goods promoted by a somatic cell enhancement. In fact, more specifically than that, it would have to be that not only are those goods much more valuable in the case of the treatment than in the enhancement, but it would also have to be the case that the goods in the enhancement case are not all that important. ‘Cause even if they were less important than the goods that are secured by treatment, if the goods secured by enhancement are still very important, well then let’s start doing those, right? And some of the discussion does seem to suggest that this kind of view is underlying some people’s distinction between treatment and enhancement. So you often get people comparing a treatment for somebody’s life-threatening illness.

And they then say, “Well, that’s obviously worth doing,” right? Somatic cell, maybe even germline, right? If we can save somebody’s life, that’s important. Whereas, when it comes to enhancements, we’re talking about adding a little bit of height to somebody, right, who’d rather be taller so maybe they can play basketball a little better. Changing somebody’s eye color so that it’s more attractive in the particular society that they live in, all right? So very trivial kinds of things. Well, if the only treatments you’re looking at, right, can save lives, and the only enhancements that you’re looking at are totally trivial, then yeah, it’s gonna seem plausible that there’s an important distinction between treatment and enhancement. But that is not an accurate way of mapping the value of these things onto these categories of treatment and enhancement. So treating somebody’s premature baldness, not a big deal, versus enhancing somebody’s immune system to prevent AIDS or cancer, right, that’s an enhancement. Incredibly important, much more important than the treatment that I’m comparing it with. So the distinction between treatment and enhancement doesn’t cleanly map onto an analogous distinction between things that are really worth running some risk for and working hard to achieve, versus things that are so trivial, we shouldn’t ever bother with them, right? And there are lots of other examples here that have kind of been beaten to death in the bioethics literature. So vaccines are enhancements. These are wonderful things, right? We should be having more vaccinations.

They’re enhancements, but they’re incredibly valuable. Much more valuable than some treatments. Immunization programs, they’re not all medical. Exercise, good diet, education, right? These enhance us, they make us better than we would have been otherwise. They’re not treating a disease, but they’re making us better. So that’s, I think, what we should really be looking at. Not whether it fits into this box of treatment or box of enhancement, but how important are the goods that are gonna be produced, all right? Let’s just go straight to the ethically relevant category here. So I think that enhancement, somatic cell enhancement should get a green here, right? If it’s responsible, right? Informed content, good cost/benefit ratio, et cetera, et cetera, then go for it. And I think it can meet those standards. And now the last bit of the argument, well if you’ve given the yellow in the lower left and green in the upper right, how you gonna refuse, right? [audience laughing] We at least have to have yellow in the lower right, right? Even if it’s responsible, you might still think that it’s not gonna be responsible as often as somatic cell treatments, ’cause we are dealing with germline stuff and maybe that’s more risky and harder to do, we have less experience.

Maybe it has to meet more exacting standards or something like that, and maybe because it’s not all that important, right, it’d have to have a better cost/benefit ratio or something like that. But these are all things that could be managed in the near future, which I’m leaving vague here, right? So I think you’ve gotta give me a yellow in the bottom right as well, okay? So let’s return to our case. This is the headline from MIT Tech Review. Exclusive: Chinese scientists are creating CRISPR babies. Antonio Regalado is looking at Chinese clinical trials registration database that’s available online, sort of analogous to our clinicaltrials. gov here in the United States, and he found a description of Dr. He’s research where it talked about genetically altering these babies. And he wrote this up a few days before the Second International Genome Editing Summit that was taking place, and he scooped the Associated Press by a day; they had been working on this, they’d actually had interviews with Dr. He, but they hadn’t published it yet because He hadn’t gone public, but then Regalado scooped them and so then they published, I think within a couple hours. And so the idea here is an interesting one.

And it’s one that had been discussed in the scientific literature, so if you go back and look at sort of predictions and suggestions about what kinds of things we might be using CRISPR to do in the near future, what Dr. He did wasn’t all that far off from what very respectable and responsible scientists had been thinking about happening in the future. But sort of way in the future, right? Not yet. So scientists have realized for a while now, that there appears to be an association between a particular deletion mutation and something called the CCR gene that confers resistance to some strains of HIV infection. And you actually have a case of a patient who got a blood transfusion from somebody who had this deletion. And they ended up having their infection cured. So hey, this is very interesting, right? So He basically used CRISPR to edit two embryos to delete the CCR5 gene, in the hopes that that would confer resistance to at least some strains of HIV. Did it, had seven couples, some of them backed out, some of the embryos sort of didn’t work, resulted in two embryos being actually implanted, so pregnancies, and then brought to term with these twins. So fall of 2018, probably mid-November, early November, Lulu and Nana, who we assume are pseudonyms, were born. So interesting proposal.

Some scientific backing here. He’d been doing a lot of work on his own with animal embryos and with human embryos that weren’t implanted, so he had a little bit of experience here, but the scientific community came down on him like a bolt of lightning. And looked over sort of every aspect of the study that they could get their hands on, which was tricky because he presented slides at the Second International Summit, but he didn’t actually have a paper that had been presented anywhere or that had been accepted for publication. So you had a situation where you’ve got people in the room, basically every single cell phone was out taking pictures of every slide, ’cause they didn’t know if they were ever gonna see this data ever again. So you can get all that online, people have put it online and you can go and look at it. So here’s some of the problems that were identified. One, it seems pretty agreed that he didn’t really have the pre-clinical data that we expect researchers to have or the community to have as a whole, in order to know that the risk/benefit ratio was acceptable. So we require a lot of pre-clinical work, a lot of animal work, before we’re gonna start with human trials on things. And he had some, but he didn’t really have the kind of robust data that people would like. He himself had no experience running a clinical trial.

So he’s a physicist by training. So he’s obviously an incredibly intelligent scientist. But he had never done a clinical trial before, he wasn’t a doctor, he didn’t have experience with patients. Informed consent was, I think, generally considered to be inadequate, you can go online and find at least an English translation of the consent form. Sort of didn’t describe the trial very well. This is a wonderful little tidbit here. “If a participating couple “who had had a successful implantation of an embryo “with no identified genetic defects or serious diseases,” right, so they’ve had a implantation, they’ve got a pregnancy with one of these embryos that was generated through IVF and then CRISPR-edited, now we’ve got a pregnancy, they do amniocentesis and genetic testing and it turns out everything looks fine from a genetic perspective, if the couple decides to back out at that point ’cause they don’t wanna go through with it, they have to return all of the cost of their IVF treatment that the research team had covered so far, which amounted to about 40,000 U. S. dollars, and if they didn’t do that within 10 days, they got an additional fine of about 14,000 U. S. dollars. That’s not was IRBs like to see in a consent form, right? [audience laughing]

Not a good idea. It’s not clear that he even had IRB approval. This is tricky, because when the scientific community started reacting so negatively to his announcement, pretty much everybody immediately threw him under the bus. And so the hospital where the babies were supposedly born said, “We’ve never seen this guy. ” [audience laughing] The university said, “He’s taking a year leave. ” He’s basically independently wealthy, he can do that. That looks actually to be true, but so they didn’t want anything to do with this, “Didn’t happen here, “didn’t happen on our watch. ” And the IRB, or the equivalent, the ethics committee over there basically said that he forged the documents. Whether they did that just to cover their own behinds or whether that was really true, how do you know? So it’s not clear that he had IRB approval.

The dad was HIV positive, and so there was some suggestion that maybe this would help mitigate the risk of the children getting HIV from the dad, but that doesn’t make any sort of scientific sense, because there are other, well-established protocols for preventing any kind of HIV infection from a dad to an embryo. You can do this sperm washing where you remove any trace of the HIV. The children had no active disease, so if you think, if I haven’t convinced you, if you think that treatment is essential, well he wasn’t treating a disease here, right? They didn’t have any condition; they were normal embryos that he’s trying to enhance. So if you think that matters, that’s one thing. And they weren’t at any significant risk of infection, so any more than anybody else in that socioeconomic status in that geographical area. Turns out the gene editing itself didn’t even work in one of the children, and yet they implanted the embryo anyway, and the other child, it appears that some cells were edited but not all. So you ended up with a mosaic organism that had some genes in one cell and different genes in another cell, and it’s not clear what kind of problems that might cause. The edit also introduced a mutation in the CCR gene that’s different from the one that gives resistance to HIV, so that is a particular Delta-32 mutation where you delete 32 bases and that’s what confers resistance as far as we know. He didn’t delete all 32 bases. He deleted like five, and we don’t know, maybe it has an affect, but we don’t know ’cause it hasn’t been studied.

The CCR mutation also makes you more susceptible to influenza, which is a problem in that geographical area. No peer review of the trial, as far as we can tell. It wasn’t even published in the Chinese registry of trials until shortly before Regalado found it, ’cause he was regularly reviewing these, but that was when the babies were born, right? So the trial happened nine months ago, at least, and it wasn’t registered. And he basically has been accused, as sort of, publication by press release, as opposed to actually getting stuff published, but at this point, because of the ethics concerns, none of the major journals are gonna touch anything he writes anyway. They’re just not gonna do it. And then, one final parting shot, he also announced that there was a third pregnancy, which should have come due in August, given when he said there was a third pregnancy. And we know nothing about that, so. Thank you very much. [audience applauding]