In Utero Genetic Editing: Doc-to-Doc with David Stitelman, MD

Modern genetic editing technology is making the dreams of science fiction writers a reality by opening the door to editing fetal genomes while they are still the womb.

In this doc-to-doc interview, I discussed the current state of this technology, and the implications for the future of humanity with a leading expert, Dr. David Stitelman. A transcript follows the interview.

Perry Wilson, MD: The day when genetic diseases can be cured in utero is closer than you may think, with a recent study in Nature Communications demonstrating for the first time the successful cure of beta thalassemia in a mouse model. But humans aren't mice and the biological and ethical challenges surrounding gene editing of human embryos are significant. To discuss these issues with me today is Dr. David Stitelman.

Dr. Stitelman is a pediatric surgeon and researcher here at the Yale School of Medicine and the Surgical Director of the Yale Fetal Care Center, which specializes in treating children before birth. He's the co-author of that Nature Communications paper and is really on the cutting edge of research in this area. Dr. Stitelman, thanks for joining me on Doc-to-Doc.

David Stitelman, MD: Great, thanks for having me.

Wilson: Congratulations on this paper. This really struck me as a milestone achievement, a landmark achievement. Am I correct in saying that this is the first time that a genetic disease has been cured in utero in a mammal, which was brought to term, which resulted in a live birth?

Stitelman: Yes, this is the first time that a fetal mammal has been edited and a disease phenotype has been reversed by that editing.

Wilson: Congratulations. That's amazing. This is a mouse model of beta thalassemia, so this is a single-gene type of mutation, I assume.

Stitelman: Yes.

Wilson: Is it the same mutation that humans have?

Stitelman: It is. Yes, there are a handful of mutations that cause the disease in humans. This is a pretty common mutation that causes the human disease.

Wilson: How do the mice do? Are they totally normal when they're born, and what percentage of them actually make it? Is this 100% successful?

Stitelman: The mice that are edited, we see about 6% to 8% editing in their bone marrow. Their anemia goes away. Their CBCs look like a normal wild-type mouse of CBCs. The mice with beta thalassemia make blood in their spleens, and so their spleens are gigantic, and these corrected animals, their spleen is still a little bit large, but it goes down to a more normal size.

Wilson: That brings up sort of an interesting issue, where thalassemia -- and presumably maybe sickle cell is another human model disease -- is a disease where you don't have to fix every gene, right, in every cell? You just need kind of some population of cells acting normally.

Stitelman: Right.

Wilson: Is this the sweet spot for in utero editing right now?

Stitelman: The two things you need are you need to target the right tissue, meaning that to get blood diseases, you need to get those blood progenitors. Then the level of editing, meaning the percentage of cells that are corrected, will correct certain diseases. Things like hemophilia, if you get 1% of normal, you'll actually go from severe disease to mild disease. For these hemoglobinopathies, usually people say ... our paper is 6% to 8%, but typically people say about 10% and you're able to reverse the disease phenotype. Things like cystic fibrosis -- 10% or 15% editing turns cystic fibrosis to chronic bronchitis, where 25% gets you probably a cure. Then there are other diseases where you may need higher levels of editing. But a lot of these diseases, to get that level of about 10%, you can probably do quite a bit of good.

Wilson: That's really amazing. Let's turn towards humans. I think a lot of people are kind of really interested in this technology for humans, but also, perhaps, reasonably nervous about it and the applications. What's the current regulatory environment like in terms of testing these things on human fetuses?

Stitelman: As far as I know, to maintain your national NIH funding you're not allowed to edit human cells or human embryos. All of our work is currently in mice and in rodents.

Wilson: But there have been some papers I've seen where there's been sort of editing of fetuses or maybe even blastocysts, like a few cells of humans which are not then implanted. They never go to term. They never turn into a newborn baby. Is that done extra-governmentally?

Stitelman: The reports that I've seen are groups in China that are regulated in a different way than we are. I think that they're using agents that are different than the agents that we're using, so they're using different gene-editing technologies, things like CRISPRs and TALENs that are a little bit different than what we're using. The concern with those agents is that if they edit your disease-causing mutation, that's great, but if they alter some other site in the chromosome while those cells are developing and becoming a human, then maybe...

Wilson: It could cause more problems than...

Stitelman: Then that wouldn't be safe. The safety is a major...

Wilson: Obviously.

Stitelman: In terms of the ethics of all this, you need to make sure that your therapy is safe before you proceed with any type of clinical application.

Wilson: Of course. Is that the Achilles heel of a gene-editing technology, in general, or CRISPR specific, is it the off-target effects that we all need to really worry about?

Stitelman: In terms of safety, I think so, yes. I think in terms of efficacy, of it actually working, you need a certain level of cells or chromosomes edited. But in terms of safety, you would worry about changing the regulation of some other gene.

Wilson: What happened in your mice? How did they...?

Stitelman: They did very well. The editing reagents that the Glazer Lab designed are called peptide-nucleic acids or PNAs. They work in a little bit of a different way than CRISPR/Cas9 or these other things. You can design them so that they can bind near mutations and edit mutations specifically, but they use the cells' own editing enzymes and editing equipment. They're probably about a thousand-fold higher fidelity than other reagents. In our paper, we saw 6% to 8% on-target editing, meaning that the beta thalassemia gene was corrected. We saw no off-target effects...

Wilson: None whatsoever.

Stitelman: ... in the homologous sequences that we looked at.

Wilson: Wow!

Stitelman: We saw no off-target editing.

Wilson: That's another real feather in the cap of the lab, that they're able to do this so precisely. I'm going to put you on the spot here. If the U.S. government is not allowing this research to proceed, are we going to lag behind other places in the world, places like China? Obviously, we want to do this carefully and deliberately, but should we start opening that door slightly to more robust research in human embryos?

Stitelman: I think that the next step is working out safety and efficacy in larger animals, and then once you've demonstrated that safety, then you can move on to the next step. I think, in my mind, the priority is doing no harm and not racing to do this before somebody else.

Wilson: You're a pediatric surgeon. You are taking care of human babies all the time, many with genetic conditions. Do families ask you about this? Have they looked you up? Do they know the kind of thing that you do and say, "Stitelman, when can I...?"

Stitelman: No. No, they haven't. There have been some. The Glazer Lab and the Saltzman Lab have been working on these agents for a very long time, and there have been families who have approached them to ask about trials or off-label use, but no family has approached me to ask.

Wilson: Larger animals are next, so non-human primates? Are we at that sort of stage?

Stitelman: I think so. We're having some conversations to figure out how to do that.

Wilson: Very exciting. On the opposite side of families coming and asking you, do you or does anyone in the lab get people telling you you're doing the wrong thing, that you're playing God here?

Stitelman: It's something that we think about quite a bit. The focus of what we're doing right now is curing genetic diseases, meaning to take it a step back, as an example, sickle cell anemia. I learned about sickle cell anemia in high school in the '90s. They'd worked out the mutation. It makes the protein fold funny. It makes the cells funny. The cells get clogged up in vessels. As a pediatric surgeon, I take care of these patients. I take out gallbladders when they get stones. I see them in the emergency room with these abdominal pain crises. I place lines for blood transfusions. I unfortunately place feeding tubes in cases where these children have had strokes and can't swallow. Seeing it on that level and knowing that the mutation is known and you could rationally, fundamentally cure the problem of the disease is what drives the lab and what drives our work. But we are thoughtful about are we, again, really focusing on safety, are we causing any harm in what we're doing?

Wilson: Let's imagine a hypothetical world where safety is addressed, the technology is highly specific, there's no off-target effects. We've gone through the regulatory process. We can potentially cure some of these monogenic diseases in utero. There, I can imagine, will come a time when the question is asked, "Can we edit other genes?" Genes that, perhaps, aren't disease genes with a phenotype that is so dramatic, but perhaps, genes that are associated with a risk of alcoholism or genes that might affect your IQ or something like that. Will there be a demand for these things, and where do you draw the line? Is there a clear, ethical place where you say, "Yes, we will edit this gene, but no, we will not edit this gene?"

Stitelman: I see where you're going, in that if you're willing to cure muscular dystrophy, are you willing to improve the muscles more to make someone stronger or something like that? If you're willing to reverse some genetic disease that results in developmental delay and you're willing to make that person brighter, are you willing to change anything to make ... as people understand how these things work, are you willing to change those? I don't have a perfect line in the sand of what you would correct and what you would not, although just if it is a disease, we would want to cure it, whereas if it's augmentation of...

Wilson: Normal...

Stitelman: ... augmentation of normal...

Wilson: I'm going to make it harder for you because I agree. I think the public, in general, would say, "Oh, no. You can't make someone smarter." But let's say that there is a gene, maybe not associated with a childhood disease, but a gene like ApoE4 that's associated with Alzheimer's disease late in life and maybe some other cardiovascular risk factors. Is that a gene that we should say, "Yeah, maybe we should change it?" If I were a parent, I might want to edit that in my child.

Stitelman: Yeah, I think so. I think if you have familial Alzheimer's, the idea of correcting that in the fetus is a little bit ... it seems like a treatment that's a couple decades too early. But if that's your window to treat it, that seems reasonable. If there's some developmental component to the disease that you don't notice, then perhaps you'd have even better neurologic outcomes than if you tried to edit that in a person who is already showing signs of dementia.

Wilson: Sure. It's a fascinating area. There's, obviously, a lot of questions still out there, but congratulations on your achievement. The first successful cure of a genetic disease in a mammal in utero is pretty remarkable. It does feel a bit like a brave new world. [LAUGHTER]

Stitelman: It's very exciting times.

Wilson: It is. We'll be watching closely moving forward and best of luck as you advance the science.

Stitelman: Great. Thank you so much.