The CRISPR Children: a chat with Dr. Kiran Musunuru

Kiran Musunuru
So there are three unpublished manuscripts that I know about. And I have two them.

Vivien
That’s Dr Kiran Musunuru of the University of Pennsylvania. He’s a physician-scientist who works in genetics and gene-editing. He has also co-founded a company called Verve Therapeutics. He has written a book about some children you may have heard about. The book is called The CRISPR generation The Story of the World’s First Gene-Edited Babies. It’s a book well-worth reading.

Hi and welcome to conversations with scientists, I’m Vivien Marx. This is another episode in a series about The CRISPR Children, the first people born with genomes edited before their birth. They were born in 2018.

I wrote an article for Nature Biotechnology about them--a piece I have been working on for three years. I have wanted to find out how these children are and how to assess any health effects they might experience form the gene-editing.

A lot of scientists refused to speak to me. But a number of them did and I am grateful to them. These podcasts are for sharing more of what I found out as I dug around on this matter.

What Dr. Musunuru is talking about is that he has seen two manuscripts by a scientist by the name of He Jiankui who with his lab team at Southern University of Science and Technology in Shenzhen gene-edited human embryos.

These embryos were implanted into the bodies of the women from whom eggs have been obtained from. The eggs were fertilized with their husband’s sperm in a lab dish. This is in vitro fertilization. That’s something that’s done around the world to help people who want to have a biological child and for whatever reason cannot do so.

Right after fertilization of the egg, though, in this instance the scientists injected CRISPR Cas 9 reagents into the fertilized egg. They wanted to make targeted gene- edits, the goal was for this to lead to people who would be resistant perhaps even immune to HIV, the virus that causes AIDS. The gene edit was supposed to disable a protein that acts as a kind of a molecular doorknob for HIV.

There are a total of three children who resulted from these experiments which caused an uproar that you may have heard about when it became public what He Jiankui and his lab team had done.

Kiran Musunuru wrote his book and he focuses on the two children, Lulu and Amy, non identical twins. He talks about his thoughts on this experiment and on gene-editing more generally. And he talks about the manuscripts he has seen.

He has this fun passage about scientific manuscripts that I am just going to read here. “If you’re ever having trouble falling asleep, try reading a scientific paper. They tend to be written in a convoluted style, chock full of technical jargon and couched in mealy-mouthed language that never quite commits to any conclusion. ….and then a bit later in that section he writes “On the flipside, if you’re a researcher and you come across a paper describing work in an area where you’re an expert, maybe work that directly impacts your own work—that paper will be an engrossing page turner that you won’t be able to put down.”

So that was the passage in Kiran Musunuru’s book

Such was the case for him and the manuscript describing the efforts by JK’s team to make the world’s first gene-edited babies.

JK is the short-hand sometimes used to refer to He Jiankui, who is serving a jail sentence related to these experiments. The manuscript that made Dr. Musunuru sit up was called Birth of Twins After Genome Editing for HIV Resistance.” To put it mildly, Kiran Musunuru is not a fan of this manuscript.

I call it manuscript because this is a scientific study that has found its way into news stories and into this book by Dr. Musunuru but it’s work that has not been published in any journal has not been uploaded to a pre-print server or a website. The data are not available for scientists to study.

A journalist at a magazine called MIT Technology Review had heard about this study and published his news piece on it. The Associated Press had been working on a story for a long time on this and then AP published its story. This was all in late November 2018.

And then a lot happened, some of which is not terribly clear. Still not, three years later. That always irked me, it’s why I kept asking people about this. Many scientists didn’t want to speak with me. Some scientists did including Kiran Musunuru.

So back to the manuscript Birth of Twins After Genome Editing for HIV Resistance.

Kiran Musunuru 4:52
That's the one that I that I recount in my book, that's the one that kind of made me very, very upset. As I related in my book, there is a companion manuscript that I do not have. I have not seen but apparently is a pre-clinical work in embryos, human embryos, monkey Embryos, mouse embryos left up to a quote unquote, clinical trial.

Vivien
It’s a quote unquote clinical trial because it seems to not have been run as a clinical trial with all the permissions that that entails. What is clear for Kiran Musunuru is that what was done is an ethics violation.

Kiran Musunuru
There were a lot of ethical problems with what he did. My sense is over time is that this is going to be a textbook example. I think it's for generations of students talk about, you know. Times when the ethics principles were heavily violated.
Of course, there have been a number of trials, of course, there's been Tuskegee, but I think this will actually be part of the discussion as well.

The chapter in my book that really focuses on the ethical issues and if that's what drove me to write it that way, was thinking that maybe someday that chapter can be cleanly taken out as and actually serve as a relatively short summary that students can read about the various principles of ethical medical research and then how in this case study they were all violated.

Vivien
Ethics violations are serious in science and especially so in medicine. There have been awful experiments. For example as Kiran Musunuru mentions The Tuskeegee trial. It was run by the US Public Health Service. It started in 1932, ended in 1972 because of a whistleblower. And for 40 years African American men with syphilis were studied. Ugh, sorry, this experiment is awful. Treatment for syphilis is penicillin and this treatment was withheld from these men.

The Nazi experiments on people another awful violation of an ethics violation in medicine. In the so-called Doctor’s Trial 23 physicians and scientists went on trial and were convicted because they engaged in the murder and torture of numerous individuals such as by testing chemical weapons on them or inflicting wounds that were intentionally infected and not treated. Ugh also awful experimentation on people.

Which Kiran Musunuru also talks about in his book. So for him and for many these gene-editing experiments are as unethical as those experiments.

I wondered about if and how these unethical experiments have been published. Indeed The Tuskegee experiment information is accessible due to a whistleblower. A scientist by the name of Tywanna Whorley investigated this, she is at Simmons College.

And these Nazi experiments are accessible, too. So it has been striking me and odd that the information about these experiments with the gene-edited children is not accessible. And yes these things are always complicated and the privacy and dignity of the children must be respected and maintained. For this gene-editing experiment to become a lesson, it seems it would benefit from relaying detail about what was done and also investigating how the victims are being cared for.

The children are here and alive and maybe they and their parents could be helped. To investigate if they have been harmed by the gene editing experiment they would have to be examined. But the question is how? What kinds of tests might be appropriate? There have never been people who were gene-edited before birth. And, of course, there is their privacy and dignity to maintain. Their parents would have to consent to any monitoring of the children’s health. I asked Kiran Musunuru about all this:

Kiran Musunuru 8:49
They're human beings and they have a certain dignity that inheres in being a human being. Part of that is not being treated like an experimental subject unless there is consent.

Now, here it's all tricky because other growing up, it's really their parents that are responsible. Right. Right. And so it's gonna be on them to decide, hopefully they'll have some freedom to decide to what degree do they want their children monitored? How invasive do they want that monitoring to be.

From a purely scientific perspective, I'm being very careful I n what I say here. Purely scientific. If you want to o to know what happened with these kids. You know, there are the issues of mosaicism and so forth you would have to be fairly exhaustive, extensive and perhaps invasive in sampling throughout the body to get a sense of what is the extent of CCR5 throughout their body. And I'm not advocating for that in any way, shape or form. And that's a lot to put on a kid purely to answer a scientific question.

The flip side is you treat them like you would any other children. They get regular pediatric visits and, you know, on an annual basis or what not, just make sure everything's okay.

Vivien
From a source I can’t name, I know that the He Jiankui lab set up a health plan for the children that resulted from his experiments. They were to get health exams until they are 18. And then the hope was that they consent to further health check ups. Ot’s unclear if the health plan is being adhered to. The plan stipulates that the lab will pay for any heath exams and some have taken place.

One big challenge the children face is that they are likely genetically mosaic. Our cells tend to be all genetically identical but because of the way the gene-editing was done, these children, and just a reminder there are three such children that we know of in order to understand the effect mosaicism might have in their bodies wone would need to sample their organs. Not possible.

And not only would you need to sample them you still would not know about that organ because you might have sampled a part that is made up of edited cells. But right next to where the biopsy needle went in, there are cells that have a different genetic code.

It might be a change in the CCR-5 gene that He Jiankui tried to edit, but the change could also be elsewhere in the genome, an off-target effect or just a mutation that took place and is not connected to gene editing.

Children have many new gene variants, they are not clones of their parents. It’s hard, perhaps impossible to know if which cells in their body of these gene-edited children are mosaic and which might cause disease or raise the risk of disease.

Kiran Musunuru 11.40
Let’s say God forbid, one of them gets cancer in their teenage years. How will you know was it actually is because of really bad luck, because some kids do get cancer or because it was a direct result of the editing. It's not clear to me that you would know. you'd be able to figure that out.

Vivien
According to a source, the three children, there’s Lulu, Nana and Amy, the children are ok. Hard to know for certain but this is what I have been told.

But should they develop a disease, such as cancer it will be hard to know if the gene-editing caused that or contributed to that. So perhaps they need more intense screening than others. Kiran Musunuru and others I contacted said it is unclear to them how to determine what might have gone awry in their cells to the point that it causes disease.

The girls genomes have been edited to induce HIV resistance and we surely do not want to test if they are or are not when exposed to the virus.

There’s another aspect, the CCR-5 gene might fulfill other jobs in the body involving the immune system and the girls may, for example be more vulnerable to viruses, but that is unknown.

Let’s say, human embryos had been edited for a heart condition, then taken to term. Kiran Musunuru has worked a lot on the heart. So I asked him what would one, in theory, do to test And let’s say there was concern about mosaicism

Kiran Musunuru 13:15
You'd have to sample the heart if you you want to know if there are any mosaic mutations in there, biopsy, which is done for like heart transplant patients and patients with heart failure. It is an invasive procedure, there are some risks, but it's not something you would do lightly. And, you know, even, you know, taking a little biopsy from one part of the heart might not give you your whole answer.

Vivien
With mosaicism it’s hard to get whole answers. Mosaicism is a problem with embryo editing. There are bound to be cells that have edited genomes and cells with unedited genomes.

And there might be different on target and off-target mutations –unintended side effects so to speak—due to the gene-editing. It’s likely to occur and it’s difficult perhaps even impossible to assess. I have another podcast just on that aspect based with Dr Rudi Jan isch from the Whitehead Institute of Medical Research in Boston.

Also about the girls health. The team used the mother’s blood to assay the fetuses in the womb and they used cell free DNA and then ran cancer panels to look for specific mutations that might lead to cancer or heighten cancer risk. Here’s Kiran Musunuru talking some more about those tests and mosaicism issues.

Kiran Musunuru 14:25
That's the devilish about, you know, mosaicism and as I probably emphasize, too much on my book . You know, I harp on it over and over again is that, you know, all those cancer panels and everything you mentioned that was from mom's blood You know, that was cell-free DNA that emanated from the fetus to ok notices that ended up in mom's blood and it was measured indirectly. That's only that only reflects a sampling of cells that were shed by the placenta that actually is not cells from the kids themselves.

Amniocentesis would give me more information. But apparently mom refused amniocentesis when it was offered to her. Right. So you don't have that more direct information even that is just cells said from basically that skin, the precursor of the skin of the fetus. into the amniotic fluid.

But if there are mutations hidden away deep in the body. You'll never know it, through any of this analysis.. Right. And it can be in hidden there might and it might you know, it might cause unforeseen problems in the future if it ends up in germ cells. Right. Person was born might be fine but you go to the next generation. It could end up being propagated there. And then it wouldn’t be mosaic anymore it would end up in all the cells that that second generation potentially cause problems.

How do you assess for this, I don't know. the science is so new. mosaicism in general somatic mutations, clonal hematopoesis of indeterminate potential. A lot of things that are really hot in medical science. Right. All right. Well, we've never had a situation where somebody has or I wouldn't say intentionally, but, you know, actively created mosaicism.

Vivien
It was likely not intentional that he Jiankui and his team created mosaic embryos but it is what likely happened.

Kiran Musunuru 16:18
And so here's the problem. If you do anything in embryos, you know, what he did is he injected in single cell embryos. And that's what scientists typically do with animals like they're trying to make transgenic animals or us or modified animals, single cell stage. And the problem is that crisper remains active beyond that first cell, even if you tried to contrive to make it so that it's more limited in its duration, what's been seen over and over again, it remains active cell divides so that some others have proposed such questions. And the problem is there is absolutely no way you can be 100 percent sure that out of some cells that you sampled, wherever it is from the fetus, from the blastocyst before you implanted.

There is no way to possibly know what is going on for sure in the rest of the cells that you didn't sample. We do not have the ability to read DNA sequence without destroying cells. Yeah. It's yeah, it's beyond our capability.

So I would argue not I'm not supporting germline gene editing. But as I mentioned in my book, what I would argue is the safe way from the perspective of their mosaicism o do it because there's an embryo to do it in sperm cells, to do with an egg cells spermatogonial stem cells, because then you can let the editing happen to let enough time pass that CRISPR is gone. And then you make the embryo. You know, you have a single cell, a single genome that is stable. There's absolutely no crispier around. You will get back into a blastocyst. Then you can be pretty sure that if you sample yourself from a blastocyst, that it does reflect what's going on in the rest of the embryo. That's the bar you have to meet.

So there's a technological issue. I don't know that it can ever be truly safe if you're actually editing embryos. That doesn't mean you can't do germline editing. It just means you need to do it at an earlier stage or you need to develop fantastic technology that doesn't exist that can somehow scan the entire genome of an embryo. I don’t even know how you do that, but non-invasively do it without harming the embryo.

Vivien
If one wanted to do heritable gene-editing and avoid mosaicism one would need a way to non-invasively scan the embryos genome. That technology does not exist.

The alternative is you edit the genomes in germ cells such as sperm or egg as Kiran Musunuru mentioned. A number of scientists have pointed this out, such as Rudi Jaenisch form the Whitehead Institute for Medical Research and George Church at Harvard Medical School

Another event that can happen in gene-editing is off-target and on target edits, both of which are unintended. There are ways to screen for those but as Kiran Musunuru explains the team of He Jiankui did not do that analysis well.

Data about the third child, whom I call Amy, are unknown. In his presentation at the Second International Summit on Human Gene Editing Summit in 2018 He Jiankui said they found one off-target mutation. And he said they found this one off target in an intergenic region of chromosome 1 of Lulu’s genome. One off-target.

Kiran Musunuru 19:25
So there's no way they picked up they, just happened to pick up the one off target. The issue of mosaicism. The one off-target effect that was itself mosaic. Correct. So they just got lucky. And one of the five cells or whatever they sampled from the embryo appeared to have that off target at it. And they picked it up basically a trace of it. And so To think that out of the 300 cells in the embryo, they just got lucky and happened to pick up the one cell that had one off target effect, no way, that beggars the imagination.

You'd have to be incredibly lucky for that to happen. And then the sensitivity of next generation sequencing. You know what they did the whole genome sequencing. It was quite poor. It was very patchy.

It was nowhere near deep enough to reliably pick up anything that happened in the genome. There are a lot of areas of the genome that aren't you well-defined where you can be sure that what you're seeing is an off target effect versus genetic variation versus whatnot. I mean, again, that issue of mosaicism , If something's only present in one out of the five cells you may not pick it up because only 20 percent of the reads, well actually it might even be less, it might be 10 percent of the reads, because you have 10 alleles across 5 diploid genomes.
So only 10 percent of the alleles might actually have the variant allele and that could be interpreted as noise when you are doing next generation sequencing. . And then you have no idea what's going on in the rest of the embryo. So I feel pretty certain that no, did not pick up those off target effects

Vivien
So in addition to mosaicism there were additional mutations that were unintended by the gene-editing. The data about these experiments are not published, Kiran Musuru has seen the manuscripts but not many scientists have. And like Dr. Musunuru there are many questions unresolved about the gene-editing that took place.

To assess the embryos and now the children adequately. Well that’s a scientific and medical challenge in an off itself.

Kiran Musunuru 22:05
We just don't have the technology, yet, we're getting better. But we can't sequence widely enough and deeply enough , especially for mosaicism to be confident that you can entirely rule out off target effects. I think they got lucky they found one. I think, you know, it's like tip of the iceberg there' re probably more. swimming underneath that. They just you know, they were not able to pick up because of the limitations of today's technology. You know, let's face it He Jiankui was not exactly the world leading expert on sequencing technologies, but he sort of presented himself as such.

Vivien
The experiments by the He Jiankui lab have had consequences, living consequences there are three girls with genomes that were gene-edited before their birth.

One question is are there perhaps more children. I have some upsetting indication there are. That others are experimenting with human embryos. One aspect I want to mention is that he Jiankui and a physician scientist by the name of John Zhang, who is a friend and also a kind of mentor to He Jiankui, so I have heard,

Zhang is CEO and founder of New Hope Fertility, which offers in vitro fertilization services. Together the two of them had been planning to open an in vitro fertilization (IVF) clinic to offer gene-editing services to people who desire a biological child. Such services are illegal in many countries and unregulated in others. And in some countries, there are facilities where other regulations apply, such as military hospitals.

If an entrepreneur wants find and use CRISPR reagents they can. Here’s Kiran Musunuru talking generally about gene editing.

Kiran Musunuru 23.55
It’s so easy to do. This is what I like to say. It's very easy to do if you don't mind doing it badly, if you don't care about the consequences. If you want to do it well, you want to do it safely and rigorously. It's very challenging with today's CRISPR technology because it does have the limitations. it's a far from effective technology if we have a long way to go before it truly can pinpoint precision, before you sure that it does complete editing 100 percent where you want it it to zero percent everywhere else. We're just nowhere near that yet in a few years, maybe, but not certainly not there now.

Vivien
Clearly there is a lot more to find out about the experiments and a lot of developments will be necessary if one ever wants to consider heritable gene-editing.

Gene-editing in adults has been progressing. It’s not the same in so many ways but Kiran Musunru told me about this, too.

He works on a gene-editing project in adults project related to heart attack. This is somatic cell gene-editing. Not heritable gene editing, not gene editing in embryos.

One type of somatic gene editing is CAR T therapy. It’s for cancer patients. Some T-cells are removed from their blood and in the lab these T cells are engineered to fight the cancer and then infused back into people. It’s called immunotherapy and has led to striking positive results and cancer immunotherapy is anow a bustling field.

Kiran Musunruru has a different disorder in mind. Among the genes he studies is PCSK9 which produces a protein, mainly it produces this protein in the liver, and this protein appears to play a role in a heritable form of high cholesterol.

One idea he and others have is that if you decrease the activity of this protein you can decrease LDL cholesterol, the so-called bad cholesterol. One can use gene-edtiting to block this protein in the bodies of adults. He has co-founded a company called Verve Therapeutics that is exploring gene-editing in adults and there are a number of companies in this space.

Kiran Musunuru 26:00
So, I mean, we have to be very clear about the distinction between germline gene editing. Right. So things that are inherited through generations that happen in embryos or potentially in sperm or eggs or whatnot. Right. And somatic gene editing, that is, you know, unambiguously in living people, adults, maybe in some situations children, but typically adults who have active illness or who are at high risk for illness. "Because I think they're are entirely different situations. "

So the work that I'm doing is entirely about really adult gene-editing. That's about patients with heart disease or who are at high risk for heart disease because of genetic conditions like familial hypercholesterolemia.

And there is a very well-established path to developing therapies. We’ve seen it with basically every drug that's been considered by the Food Drug Administration. We've seen it with the first gene therapy applications that have now been approved with CAR-T immunotherapy. And we're starting to see with somatic gene editing.

And so there's a well-established path, requires lots and lots of preclinical work, years of work in pre-clinical models, whether it's mice or non-human primates, etc. Generation of so much data takes up thousands of pages that you submit to the FDA or in Europe, the EMA the equivalent in Europe before you get approval to proceed with a clinical trial.

And then even when you're doing the clinical trial, there is so much oversight. You know, everything is reviewed, all the Ts are crossed, all the Is are dotted. It's a very, very different scenario than what He Jiankui did. I mean, he basically, you know, quote unquote, clinical trials if you want to call it that, it was done with no oversight, no supervision. It was just him thinking that he could run a clinical trial despite having no medical training or no experience in clinical trials.

There was some question of an ethics committee maybe giving approval. That seems to be under debate whether that actually really happened or not. And, you know, he just did what he thought he knew what he was doing, and in retrospect, he didn't really know what he was doing.

Whereas, you know, any effort I make, any effort that, you know, my fellow scientists at either companies that are doing gene editing, developing applications or institutions like my home institution, the University of Pennsylvania, where there are some clinical trials underway. Everything is being done in a very, very responsible way that it takes years to unfold.

And so and it takes a lot of money, a lot of resources to generate all the data, preclinical models to establish safety, to satisfy regulators before even allowed to touch a human being. And then you have to go through all the various phases of clinical trials and satisfy the FDA once again before they'll allow you to actually market it to patients.

Vivien
At Verve Therapeutics Kiran Musunuru is senior scientific advisor and he has remained at the University of Pennsylvania. The company it was spun out of the University of Pennsylvania into a UPenn incubator space at UPenn hand oved to Cambridge Massachusetts where it is headquartered. The company plan is to develop gene editing delivered to patients who are at risk of cardiovascular disease.

There are ‘gene variants a person might have in his or her genome that lowers their risks of artherosclerosis and it protects against heart attack. The goal the company has is to mimic this and use gene editing to lower lipids in the blood to reduce this risk.

The genes their technique addresses are expressed in the liver and so these genes would be changed. In particular the genes PCSK9 and ANGPTL3.

The approach will make changes in the cells of an adult’s liver but it’s not a change that is passed to that person’s children. This process does not edit embryos or sperm or eggs. One day there might be clinical trials in people and to file what is called an investigational new drug application with regulatory authorities in the US and in Europe.

Since I spoke with Kiran Musunuru he and his colleagues published a study in which they used a different kind of gene-editing called CRISPR adenine base editing to target the PCSK9 gene.

They created a mutation in that gene that lowered the level of protein that gene produced and it also lowered LDL cholesterol in the body of the monkeys for eight months. The results the authors of the study point out, it was published in nature, The results “ provide a proof of concept” for the delivery of base editors to the primate liver. It’s somatic base-editing, not germline gene-editing. Potentially a therapy, still being tested, of course.

It is gene-editing and it is not at all like the work He Jiankui did.

Kiran Musunuru 31.05
We're very unambiguous. And anyone we talk to and it's right there on the website, we have absolutely no intention whatsoever of going anywhere near the germline here. You know, we are very specifically unambiguously about treating adult patients with heart disease or at risk for heart disease.

Vivien
His work a U Penn and at Verve is not at all about germline gene-editing. But he is of course concerned about what He Jiankui did and about the need to learn from this.

That would mean making the data somehow available to the biomedical community. But there are lots of issues with that. Some are specific to China, such as whether or not the data could even be permitted to go to a labs outside of China to be analyzed. And there is the need to privacy about the children and they have to remain their dignity.

Kiran Musunuru 31.45
It’s this tension between privacy, respect for privacy, which I fully appreciate . I’m a physician I appreciate that as much as anybody else. And the need for people, scientists, really. But really, I think the general public to understand what happened here to the extent that it doesn't to cause you know, cause unintentional harm to the people involved, which in this case is the twins or the parents.

There's also a feeling that you do want to put too much data out there. Right. You mentioned Mi-seq data. That gets tricky because that's not that's definitely not published. Right. The chromatograms, you know, some of the stuff I talk on the book. All of that was actually publicly presented by He Jiankui himself with a live stream audience of more than a million people still actually watch it.

Vivien
Some data about the girls about what He Jiankui can be found, sometimes in unexpected places, one can watch the presentation and some people like Kiran Musunuru have the manuscripts. The manuscripts are not to be found in a scientific journal not on a preprint server. Here’s Kiran Musunuru.

Kiran Musunuru 33.05
There's nothing I really talk about in my book with respect to data, with respect to, you know, like actual, you know, chromatograms or information or whatnot that isn't already out there. Well, I saw comfortable with that. Once you talk about MiSeq data, once you talk about, you know, genetic variation and, you know, some of that said the manuscript, s ome of it is on the Chinese clinical trial registry site, believe it or not.

You know, you're inching towards, you know, how much is too much information to be sharing about live human beings even though the data is not directly from the human beings. It’s this from embryos, placenta, someone that you don't we don't actually know that that that matches up perfectly with what's in the kids. It’s tricky, I don’t think there’s ever go ing to be a consensus on what the right answer is here.

Vivien
That was Conversations with scientists an episode in the series called The CRISPR children. Today’s episode was with Dr. Kiran Musunuru of the University of Pennsylvania and who also co-founded the company Verve Therapeutics.

There are multiple episodes about The CRISPR children and also a story by that name in Nature Biotechnology.

And I just wanted to say, just because there is confusion about these things sometimes, the University of Pennsylvvania did not pay to be in this podcast.This is independent journalism produced by me in my living room. I’m Vivien Marx thanks for listening.